BIOLOGY UBRARY G IMMUNITY METHODS OF DIAGNOSIS AND THERAPY AND THEIR PRACTICAL APPLICATION CITRON GARBAT IMMUNITY METHODS OF DIAGNOSIS AND THERAPY AND THEIR PRACTICAL APPLICATION BY DR. JULIUS CITRON ASSISTANT AT THE UNIVERSITY CLINIC OF BERLIN, II MEDICAL DIVISION TRANSLATED FROM THE GERMAN AND EDITED ' BY A. L. GARBAT, M. D. ASSISTANT PATHOLOGIST AND ADJUNCT VISITING PHYSICIAN, GERMAN HOSPITAL, NEW YORK SECOND EDITION, REVISED AND ENLARGED 30 ILLUSTRATIONS 2 COLORED PLATES AND 8 CHARTS PHILADELPHIA P. BLAKISTON'S SON & CO. 1012 WALNUT STREET I COPYRIGHT, 1914, BY P. BLAKISTON'S SON & Co. THE. MAPLE. PRESS. YORK. PA TO PROFESSOR FRIEDR1CH KRAUS AS EVIDENCE OF DUE HONOR AND THANKFULNESS THIS BOOK IS DEDICATED BY THE AUTHOR ON THE OCCASION OF THE OPENING OF THE NEW II MEDICAL DIVISION 3C0085 PREFACE TO THE SECOND GERMAN EDITION. The first edition of "Immunity, Methods of Diagnosis and Therapy," was, I am glad to say, favorably received. The book apparently fulfilled a demand, as not quite two years have elapsed and a second edition is required. An Italian translation by G. Volpino and an English edition by A. L, Garbat have in the meantime appeared. A Spanish and a Rus- sian publication are in preparation. During the last two years certain problems in this field have gained greatly in importance, and therefore require more detailed consideration. Special chapters have been allotted to tumor studies and to anaphylaxis. Probably, even this treatise on anaphylaxis will by many be considered too short. Without underestimating the distinct scientific value of the very numerous articles being written on this subject, I feel that the fundamental principles of anaphylaxis are still too theoretical and the practical application still too limited, to warrant a more exhaustive re- view in a practical book such as this. Chemotherapy, on the other hand, has attained such striking clinical importance, that I have dwelt in length upon its development. As ever, the practical elements are emphasized. My broad experience with salvarsan has aided me in distinguishing the points of importance for the practitioner. ' Slight changes and additions have been scattered throughout the book. I hope that this second edition will add new friends and retain the old ones. JULIUS CITRON. Vll PREFACE TO THE SECOND ENGLISH EDITION. The necessity for a second English edition has arisen after an interval of a year. It is a source of satisfaction to note the apparent general approval and demand which this small book on " Immunity" has fulfilled. The credit to Dr. Citron is well deserving. The text of the new English publication has been carefully reviewed and minor changes made, but it follows closely that of the German with its important additions. Like in the first English edition, however, so also in the second, topics of more recent development, treated only very slightly or not at all in the German, have been enlarged or inserted: gonococcus and typhoid complement fixation tests, agglutination and hemolysis tests for transfusion, prophy- lactic typhoid inoculation, etc. Friendly criticism to discuss some sub- jects with more detail has been gladly met and corrected, so that a be- ginner may have no greater difficulty than is only natural. It is hoped that this English edition will meet with the same approbation which I am happy to say has been accorded the former one. A. L. GARBAT. IX PREFACE TO THE GERMAN EDITION. This book is to serve a purely practical purpose. The methods of serum diagnosis, on account of their growing clinical significance, are constantly stimulating greater interest in all branches of medical science. While giving instruction in this subject, I realized that it would be of great help to both the medical student and physician if they possessed a short text-book which would review in a purely critical form the various methods of immunity diagnosis, especially those relating to tuberculosis and syphilis. The two systems of Kolle and Wassermann, and R. Kraus and Levaditi are doubtless the standards on the subject in German medical literature. On account of their size and price, however, these volumes come to be sought only by the specialist. It was therefore my aim in this book so to present the subject of immu- nity that the general medical man, who is even slightly acquainted with laboratory work, can learn the details of the various reactions and their significance. In selecting the different methods, I have taken up those which are used in the clinic for diagnostic, therapeutic, or prophylactic purposes. In addition there are herein included certain fundamental considerations of questions on immunity which for the present are only of theoretical interest, but which owing to the rapid development of the subject, may soon become of practical importance. I have endeavored especially to place before the reader a critical review of the results of the various methods. In the description of technical details, the original articles of the author have been selected; modifica- tions having been considered, only provided they exhibit distinct ad- vantages over the original method. I here wish to express my thanks to my teacher Prof. Wassermann, under whose guidance and stimulus I gained my laboratory experience; also to my chief Prof. Kraus whose clinical genius proved to me the practi- cal importance that this subject of immunity commands. To the publishers as well, whose kind cooperation in all my plans as regards publication and illustration, greatly simplified my work, I extend my heartiest appreciation. JULIUS CITRON. XI NOTE BY THE AMERICAN EDITOR. The study of " Immunity," once of merely theoretical interest and purely scientific importance, is to-day no longer such. A realm of prac- tical considerations, considerations which are constantly coming up and enlisting the attention of the busy practitioner have little by little sup- planted those phenomena at one time vaguely understood and mostly taken for granted. Gradually have the uncertainties so long dominating and obscuring an intelligent comprehension of the subject been cleared away; mistakes explained; and hypotheses re-established as proven facts. The methods employed for the necessary investigations have naturally improved and increased with such extreme rapidity that a severe task presents itself to one who desires to separate the more from the less valuable ones. It was therefore with extreme satisfaction that I greeted the opportunity of bringing out an English edition of this work- ing hand-book on the various, but most essential methods used in the applications of •" Immunity." The author of this volume has, by his exhaustive research and extensive practical experience as a teacher, treated his field with such fulness and preciseness of detail that it is of value not only to the laboratory student, but also to the clinician. Its already favorable reception in Germany will, it is hoped, be ex- tended to it in America, especially by those whose lack of familiarity with the German language has kept this work beyond their reach. The chapter on vaccines has been slightly revised and elaborated to conform more closely with the most recently advocated methods of Sir A. E. Wright, to whom the editor is indebted for his experience. Other- wise there has been no need to alter the original text, with the exception that here and there some features which may be of special interest to the English reading public, have been inserted. I wish in the present connection to express my deep thanks to my teacher Dr. Citron for offering me the privilege of this undertaking, and to the publishers, Messrs. Blakiston & Co., without whose hearty coopera- tion this would have been impossible, my sincere appreciation. A. L. GARBAT. Xlll TABLE OF CONTENTS. CHAPTER I. PAGE INTRODUCTION i Definitions of immunity and antibody. The law of specificity. The neces- sity of control tests. CHAPTER II. LABORATORY EQUIPMENT . 7 General technique. (Technique of injection. The methods of obtaining and preserving serum. Bacterial filtration. Dilutions. Measurement of small amounts of bacteria.) CHAPTER III. ACTIVE IMMUNITY ' ; -;..v'. V.V*:':V •'•=. '•'-; .-'-. '..". '•'.- . ;'.".t'.;. \ . 21 Immunization with living and dead virus. (Vaccination against small-pox; antirabic vaccination; antityphoid inoculation.) :':•%>•: CHAPTER IV. ACTIVE IMMUNITY . . ... 36 Immunization with bacterial extracts. Aggressin experiments. CHAPTER V. TUBERCULIN DIAGNOSIS ./'.-..:: ^'^''.. ^:-^'^, :,\'-,\^'^:\-. ... 45 Koch's method; cutaneous reaction; Moro's ointment reaction; intracutaneous reaction; ophthalmo-reaction; the specificity and prognostic value of the tuberculin reactions. Mallein. Tricophytm. CHAPTER VI. TUBERCULIN THERAPY 60 The technique of the tuberculin therapy; old tuberculin; new tuberculin; bovine tuberculin. Nastin. CHAPTER VII. TOXIN AND ANTITOXIN . . . . '.-:. •'••'. •'.'* -• . . :,••* .•: '.'..- .. -^. •> V-.". . :. :.-..:. Xi . 74 The serum therapy of diphtheria. CHAPTER VIII. TOXIN AND ANTITOXIN (continued) . . -. , ;. , ^. ',- . ............ 85 Definition of toxin. Tetanus toxin. Botulism toxin. Dysentery toxin. Staphylolysin. CHAPTER IX. THE TOXINS or THE HIGHER PLANTS AND ANIMALS AND THEIR ANTIBODIES. FER- MENTS AND ANTIFERMENTS . 95 Snake poison. Paroxysmal hemoglobinuria. xv XVI TABLE OF CONTENTS PAGE CHAPTER X. AGGLUTINATION Macroscopic test, microscopic test. Group agglutination. Bacterial 105 agglutinins. Hemagglutinins. Transfusion tests. CHAPTER XI. PRECIPITINS 120 Bacterial precipitation. Proteid precipitation. CHAPTER XII. BACTERIOLYSINS AND HEMOLYSINS (CYTOLYSINS) 131 Technique of bacteriolytic tests. Pfeiffer's phenomenon. Bactericidal test by plate method of Neisser and Wechsberg. Hemolysins. Cytolysins. CHAPTER XIII. METHOD OF COMPLEMENT FIXATION 152 Principles of this method. Antituberculin. Ehrlich's side-chain theory. Serum diagnosis of syphilis and of diseases caused by animal parasites. CHAPTER XIV. TECHNIQUE OF THE COMPLEMENT FIXATION METHOD 171 The original method of Bordet-Gengou. Wassermann-B ruck's modification. The technique of the serum diagnosis of syphilis. Echinococcus disease. Epidemic meningitis, tuberculosis, gonococcus infections, typhoid fever. The differentiation of proteids according to Neisser-Sachs. CHAPTER XV. PHAGOCYTOSIS. OPSONINS AND BACTERIOTROPINS 196 Technique of opsonic index determination and of "Wright's vaccine treatment. Neufeld's method of examination for bacteriotropins. CHAPTER XVI. MALIGNANT TUMORS 216 Studies in reference to their immunity; serum reactions; transplantation experiments. Meiostagmine reaction. CHAPTER XVII. ANAPHYLAXIS 221 CHAPTER XVIII. PASSIVE IMMUNITY. (SERUM THERAPY) 231 Bacteriolytic sera. Special serum therapy. CHAPTER XIX. CHEMOTHERAPY 240 Definition. Atoxyl. Salvarsan. Chemotherapy of malignant tumors. LIST OF ILLUSTRATIONS AND CHARTS. FIG. PAGE 1. A room in the laboratory of the Royal Institute for Infectious Diseases (Berlin) 7 2. Standard for measuring the size of platinum loops (Czaplewski) .... 8 3. Intravenous inoculation (after Uhlenhuth) . 9 4. Intraperitoneal inoculation (after Uhlenhuth) 1 1 5. Removal of peritoneal exudate in Friedberger's position (original) ... 12 6. Veno-puncture (original) 13 7. Wet-cup method for obtaining blood (original) 14 8. Test-tube for preservation of serum (original) 15 9. Pukal filter 16 10. Filtration through pukal filter . 16 11. Reichel filter 17 12. Lilliputian filter 17 13. V. Pirquet's tuberculin test (original) . . , ... ..... Y, :. . . . . 51 14. Ophthalmodiagnosticum for tuberculosis (original) 52 15-16. Diagram for the complement fixation reaction '.'•;• • • • 154 17. Diagram for the complement fixation in syphilis . . 166 18-20. Technique for the determination of the opsonic index according to Wright 204 21-22. Technique for the determination of the opsonic index according to Wright 204 23-24. Technique for the determination of the opsonic index according to Wright 205 25-26. Technique for the determination of the opsonic index according to Wright 206-7 27. Phagocytosis of tubercle bacilli 208 28-30. Technique for intravenous injection of salvarsan 205-2 CHART. 1. Example of a diagnostic tuberculin reaction 48 2. Example of hypersusceptibility by diminution in the tuberculin dose . . 65 3. Marked increase of weight in a tuberculous individual in spite of con- tinued fever . . .... . .v :.'"'..••• 67 4. Treatment with S. B. E., almost without reaction. Immunization against B. E. . . ". ....•%?,..•.•;! 71 5. Opsonic curve after a small dose of staphylococcus vaccine ...... 200 6. Opsonic curve during treatment with new tuberculin 201 7. Increase in the opsonic index for gonococci by Bier's hyperemia .... 202 8. Auto-inoculation with tuberculin after physical examination and massage 202 xvii IMMUNITY. CHAPTER I. INTRODUCTION. — DEFINITIONS OF IMMUNITY AND ANTIBODY. — THE LAW OF SPECIFICITY. — THE NECESSITY OF CONTROL TESTS. The diagnosis of infectious diseases can be approached in several ways. In addition to the aid obtained from clinical signs such as the course of the temperature, the changes in the various organs, the exanthemata, etc., the finding of the specific etiological agent of the disease, or the specific anti- bodies developed by the reaction of the organism, are of equal or even greater importance. • The course of an infection depends not only upon the nature, the number, and the virulence of the infecting agents, but also upon the behavior of the infected body. One must consider a disease as the result of the interaction of both of these factors without necessarily being able to attribute the various symptoms to either the one or the other. Although the general reaction of the organism is varied, it can nevertheless be shown that in spite of even individual differences, the characteristic bacteria and their products bring about a distinct symptom-complex which is usually concomitant with a significant defense on the part of the organism. The means which the body employs in this protection are cellular and humoral in nature. In fact, there is a group of infectious diseases in which the cellular reaction predominates, and another in which humoral changes are pre-eminent; and between these extremes are various intermediate forms. Thus, the constantly changing picture of tubercu- losis always shows the tubercle as its typical product of cellular reaction; similarly leprosy and syphilis have their peculiar cellular changes. More difficult, however, to recognize by the unaided eye or even the microscope, are the finer biological alterations which take place in the body fluids during the course of infectious diseases. Here, special methods are necessary to detect and differentiate the various humoral changes which occur for the main part in the blood serum. As is known at present, the humoral as well as the cellular immunity reactions are not limited to infectious diseases, but also express normal physiological and pathological conditions. With the conception of Ehrlich's side-chain theory the bridge of understanding for the humoral reaction was built, and it at once became evident how the physiological phenomena of nutrition and production of energy are identical in their nature with processes which . . ; ^**»* -INTRODUCTION *» * : > >< under pathological circumstances lead to the formation of anti-infectious bodies. In an analogous and no less ingenious manner, Metchnikoff has shown that the same cell group of mesenchymal origin which the organism stations against bacterial invasion has physiological and physio-patho- logical functions to f ulfill in the whole animal scale. In the lower animals, these cells aid in the metamorphosis of the body structure, thus leading to the disappearance of entire organs. In the female, they aid in the involution of the uterus after labor, while in the aged they destroy the nerve cells in the senile atrophied nerve centers or finally as chromophages turn the hair gray. The border-line between the physiological and patho- logical status is biologically not sharply demarcated. It is one single chain of manifestations which possess numerous transitional phases. As the methods of serum diagnosis can prove reactions much finer even than those accomplished by chemistry, their application has not been limited to the chapter on infectious diseases. By their means also, proteids, even though manifest in minutest traces, can be differentiated. Similarly, the secret of blood relationship has begun to be unravelled; and there is a possibility even of solving the problems of metaboh'sm. Closely associated with serum diagnosis is the serum therapy. Even though the general application of the latter is not as widely developed as that of the former, it must be remembered that through this medium diphtheria has been transformed from a fatal to a combatable disease, and incidentally made the name of Behring immortal. To-day, attempts are constantly being made to treat other bacterial and toxic diseases by specific therapy and it is to be hoped that success will soon be met with. The study of serum therapy and serum diagnosis is undertaken in various ways. It is comparatively simple to learn only the purely technical details. All large laboratories have trained assistants for the performance of certain reactions or groups of reactions with absolute precision. Although as we have said, they do such work as assigned to them with accuracy, they are nevertheless far from a thorough understand- ing of the subject of serum diagnosis. Unfortunately this blind method of procedure has recently been advocated to an alarming extent. In addition, the practical success which the Wassermann reaction has met with has inculcated the desire in a certain class of physicians, for carrying out this test alone, and thus to become independent of the use of large laboratories. To meet this demand, short courses have been established and the serum diagnosis of syphilis taught with lightning rapidity. That such a state of events is absolutely injurious is clearly evident. It is impossible for one to be a specialist in a certain reaction and at the same time be ignorant of the other phases in the study of C INCEPTION OF IMMUNITY 3 immunity. Unreliable and erroneous results are the inevitable out- comes of such unscientific work. The plan followed in this book consists in taking up all of the important principles and methods of immunity, even though at present some may attract no direct practical attention. The principle of the now widely important Wassermann reaction had been described years previously by Bordet and Gengou, but merely from a purely theoretical standpoint. Only with the development of the Wassermann test did it attain its prac- tical importance. To start systematically, it is necessary, primarily, to under- Conception stand certain terms frequently employed. First, the word of Immunity, immunity requires explanation: After an individual has recovered from an infectious disease, he passes into a state where he is less or even not at all susceptible to the same infection, although no macroscopical, microscopical or chemical change can be shown to have taken place in his system. This condition is one of immunity. And as the body itself by its own struggle with the invading bacteria has brought about this immunity, it is known as " active immunity." Jenner and Pasteur have employed this mode of immunity acquired spontaneously by overcoming an infection in their principle of prophylactic vaccination. The exact nature of this active immunity is only partially understood. It can be shown, however, that the indi- viduals thus actively immunized have within their organism reaction bodies of a specific nature directed against the infecting elements and their poisonous products. These reaction bodies, which circulate mainly in the blood serum, are known as Antibodies. The antibodies are of differert classes depending entirely upon their varied forms of activity. While some, such as the agglutinins and pre- cipitins, have the property of grouping their respective invading agents into small clumps or precipitates, without at the same time embracing protective powers, there are other antibodies which act essentially for the defence of the organism. They attain this by neutralizing the poison of the bacteria (antitoxins) or by destroying the bacteria (bacteri- olysins), or so altering the bacteria that the latter can be more easily destroyed by the cells (bacteriotropins, opsonins). The last three types of immunity can be designated respectively, as antitoxic, bactericidal,, and cellular immunity. Naturally there are many intermediate forms. It is very probable that besides these well-recognized forms of immunity there may be others, still unknown. Cellular immunity must surely have a far greater range of importance than is at present ascribed to it. There is, no doubt, a distinct cell immunity which acts without the aid of any serum substance and is known as "Tissue Immunity" ("histogene" Immunitat) . 4 INTRODUCTION If the serum of an animal which has been immunized, and containing antibodies, is injected into another normal but non-immunized animal, the latter acquires the power of being immune against the specific infective agent. In this case the immunity was not established by direct cell activity on the part of the animal, for the organism remained passive, and had, as it were, immunity thrust upon it. This form of immunity in contradistinction to "active immunity" is designated as "passive immunity." The forms of immunity thus far mentioned were all "acquired" either by the spontaneous recovery from the infection or the artificial transmission of the curative antibodies. In contrast, however, to this "acquired" immunity there is a "natural" immunity by which is under- stood that some animal species are not at all susceptible to certain infec- tions. Thus, man has a natural immunity against a group of diseases markedly fatal for some of the lower animals, e.g., chicken-cholera and hog-cholera. That this natural immunity is almost always cellular in character is undeniably true; and the most important form of this natural armament against infection is the powerful leucocyte, capable of engulfing and destroying the invading enemy. In other words, phagocytosis. Finally one should speak of a "local" and "general" immunity, meaning to express thereby the different resistance and susceptibility that various organs of the same individual display; and also of a "relative" and "absolute" immunity in order to differentiate quantitatively a transitory immunity from one that is of long duration. Another term very often employed is "antibody." This, as has already been explained, is a name used to designate the Antibody sPecmc bodies which the organism produces -as a reaction against the infecting agents and their toxic products. Anti- bodies are also formed when animals are injected with foreign proteids not of bacteria] origin, such as the blood from a different animal species, egg albumin, etc. In order that these antibodies may be obtained, the substances employed must enter the system "parenteral," i.e., some way outside of the gastrointestinal tract. In older literature the terms antibody and protective body were used synonymously. That is decidedly incorrect, inasmuch as not all anti- bodies possess the power of protection and not every actively immune organism, demonstrable antibodies. Furthermore, antibodies as the bacteriolysins which are generally considered to have protective powers, and correctly so too, can exist in a system in large numbers with out neces- sarily rendering that organism immune. How complicated various chapters in the study of immunity can be will be clearly evidenced by a few of the author's experiments with the hog-cholera THE LAW OF SPECIFICITY 5 bacillus. Rabbits rendered actively immune by inoculation with extracts of hog- cholera bacilli possess a serum which when injected into an animal of a different species, as the guinea-pig, will render the latter passively 'mmune. If, however, the serum is injected into another animal of the same class (another rabbit), no protective power is transmitted. In other instances it was shown that the rabbit which was being treated with the purpose of active immunity was in reality never immune, as it always suc- cumbed when injected with living bacteria even though its serum contained bodies which were perfectly able to passively protect guinea-pigs against the same deadly infection . Just as it is incorrect to consider an antibody and protective body as one and the same thing, it is equally erroneous to deny the existence of protective bodies, because their presence cannot be demonstrated by a certain method of laboratory examination. It must be kept in mind that there are still many unsolved problems in the subject of immunity and that therefore only the positive findings should be the basis for drawing conclusions. In order to learn the nature of these antibodies attempts have been made to isolate them chemically. Thus far all such trials have been unsuccessful. It is even uncertain whether these so-called antibodies are definite chemical entities. Only the effects of the serum as a whole are known, and the ingredients in it to which these activities are attributed are thought of as antibodies. For didactic purposes antibodies, as antitoxins, agglutinins, etc., will be spoken of in this book when the anti- toxic or agglutinating properties, exclusively, are meant. In spite of the individual differences which are ascribed to the The Law of various classes of antibodies, there is one quality possessed by Specificity. all — their specificity. To explain this by a rather crude example, may be mentioned the fact that typhoid antibodies will give their various reactions of immunity only when these are per- formed with the typhoid bacillus, and cholera antibodies only when per- formed with the cholera vibrio. Substances which lack this essential property of specificity cannot be considered antibodies, although they may fulfill all other requirements. There are indeed limitations to this fast rule, but these will be considered subsequently. ' For the present the following can be taken as a fixed fact; namely, that every true antibody is absolutely specific, and that all substances or bodies which are not specific cannot be real antibodies. The law of specificity is the funda- mental principle of serum diagnosis. As soon as the specificity of a reac- tion becomes doubtful, its diagnostic importance suffers greatly. In the following pages, therefore, the question whether or not a reaction is specific will be repeatedly discussed, and it will be the aim in every way possible, especially by the use of control tests and experiments to outline the limits of this specificity. Here, even at so early a stage of the discus- 6 INTRODUCTION sion, the absolute necessity of these control tests must be urged; though it may appear superfluous to the beginner, that for apparently simple experiments, controls are performed which consume more time than the actual diagnostic test itself. Probably also the desire will arise, and perhaps be satisfied, to omit these control experiments. Not- withstanding the possibility that for a long time perfectly good results will be obtained, it cannot be too often or too emphatically impressed upon all workers in immunity methods, that the only guard against mistakes and failures in diagnosis is necessarily found in control tests. And especially in doing research work, the latter are indispensable. For, experimental work which involves reasonable possibilities, or has any pretension towards plausibility, warrants no true scientific conclusion without the employment of such tests. The author has made it a rule, whenever new findings in serum diagnosis are published, always to look for the given control experiments. If these are insufficient, then no matter what the contents are, the value of the research is slight, for all its claims only may, but not necessarily must, be correct. CHAPTER II. LABORATORY EQUIPMENT. — GENERAL TECHNIQUE. Although some of the tests of serum diagnosis are comparatively simple and can be performed in one's office or even at the bedside, in most instances a laboratory equipment is essential. This of course does not at all imply the necessity of such elaborate apparatus as one is accus- tomed to find in our present up-to-date bacteriological or serological FIG. i. — A room in the laboratory of the Royal Institute of Berlin for the study of infectious diseases. laboratories where a great deal of complicated research work is done. For the practical application of serum diagnosis, as employed at the hos- pital or in private practice, an outfit much less costly is perfectly suffi- cient. As regards the question of a room, the selection of one with two windows, allowing the entrance of sufficient light, is indispensable. At the same time, however, some arrangement should be made in connec- tion with the windows in order that the direct rays of the sun be prevented from striking one's desk. Strong sunlight may weaken or even destroy 7 O LABORATORY EQUIPMENT the virulence of cultures, or bring about many changes in sera. Even diffuse daylight should not be considered as entirely inert. A general rule to be remembered is never to expose any biological reagent, be it a bacterial culture or any form of its derivative, a serum, or any other sub- stance to daylight any longer than is absolutely necessary. If this dictum is followed, one will avoid many a difficulty. To conform with this idea, it is wise to have upon the table a small closet into which the cultures and sera can be placed for the time that they are being used. Such a con- venient receptacle can be made out of a large cigar box, painted black inside and out, with its lid replaced by a small black curtain. The table or desk at which one works should be near the window, and covered with filter-paper, upon which should come a glass or asbestos plate. Instead of a wooden table it is certainly more elegant, but costlier to have a top plate of glass. Upon the table there should be a Bunsen burner, a microscope, a lamp for microscopic work at night, a dish filled with sublimate or cresol into which the infected substances, old cultures, used pipettes and graduates are placed. It is very convenient to have running water and a hood in the same room. Still neither of these is absolutely necessary. As for larger appa- ratus— must be mentioned, a thermostat, a mechanism for shaking, a dry sterilizer, a good autoclave, a water-bath, an instrument sterilizer, a water or electrical cent- rifuge, an ice chest, a closet for instruments and glassware, and finally animal cages of the kind that are easily cleansed. As for instruments and glassware the following are required: scalpels, scissors, forceps, glass-cutter, sterilizable syringes of various sizes, graduates of 10, 25, 100, and 1000 c.cm. capacity each, pipettes of i c.cm. with i/ioo divisions and pipettes of 10 c.cm. with i/io divisions, a sterilizable pipette retainer, Erlenmeyer flasks, Petri and KolleV dishes, test-tubes, dark glass flasks, ordinary water glasses, funnels, glass tubing of various sizes, and test-tube racks. Furthermore, a platinum needle and a platinum loop are required. For making a loop of a definite size, and one which can always be referred to, the small instrument devised by Czaplewski is of great help. It consists of four round metal bars i, 2, 3, and 5 mm. in diameter around which the platinum wire can be twisted in order to make a standard loop (Fig. 2). All instruments and glassware used for serum work should be per- fectly clean. It is best to have all the glassware plugged with non- absorbent cotton, and sterilized by dry heat. It is never advisable to clean the glassware with strong acids, alkalies or other strong chemicals. If this has been done, the chemicals must be thoroughly removed by wash- ing, as the slightest trace may interfere with the accuracy of some tests. All used glassware should at once be placed into a disinfecting solution. For this purpose, lysol, lysoform and cresol solutions are highly recom- FIG. 2. — Instrument (After Cza- plewski) for the standardization of platinum loops. TECHNIQUE OF INOCULATION 9 mendable. Sublimate is less efficient because it coagulates albumins and thus may lead to plugging of pipettes which may have contained blood rests. If highly infectious material has been examined, it is best to place the entire disinfectant solution containing the used glassware into the auto- clave, sterilize it there, then wash the supply thoroughly with soap, dry and resterilize it by dry heat for i to 2 hours at 120° C. The Technique of Inoculation. Both for serum diagnosis and serum therapy, the serum is required from animals which have been artificially immunized against the bacteria or their products of secretion. Almost without exception, this immuni- zation is produced by injecting the animal with the infectious virus. The method of inoculation is either intravenous, intraperitoneal, or subcutaneous. The technique of intravenous injection varies somewhat with Intravenous different animals. In rabbits, the veins running along the Injection. outer margins of the ears are most suitable. The assistant sits upon a chair, holds the hind legs and body of the rabbit tightly fixed between his knees and thus has his hands free to steady the rabbit's ears. Another method consists in placing the rabbit upon the FIG. 3. — Intravenous inoculation. (After Uhlenhuth.) table and firmly holding him there while the injection is made (Fig. 3). The ear is first struck gently with the fingers and washed with alcohol and xylol. If the hair is very long, it should be clipped. If the vein running along the outer margin of the ear is exceptionally small, it can be made more prominent by compressing it between the thumb and index finger at the root of the ear. No force should be used with the injections; the fluids should be allowed to flow into the blood stream very slowly. Glass 10 LABORATORY EQUIPMENT syringes, or such as can be sterilized easily, are preferable. Air bubbles are to be carefully guarded against in order to exclude the danger from air embolism. If infectious material is used for injection, it is advisable in such instances' to place a small piece of cotton moistened in alcohol or a 5 per cent, carbolic acid solution around the point of union between the needle and the barrel of the syringe to prevent the possible escape of any fluid which usually occurs at this point. After inoculation is completed, the needle should be quickly with- drawn, a small piece of non-absorbent cotton placed upon the needle puncture and compression applied. If non-virulent bacteria or albumin is injected, the bleeding may be almost instantly controlled by firmly squeezing the vessel above the puncture wound with the edge of one's finger nail. In guinea-pigs intravenous inoculation is more difficult, as here there are no large superficial veins. The Jugular or Iliac vein is therefore chosen, and must be dissected free. It is not necessary to tie off the vessel, but the wound should be firmly compressed by means of clean gauze or cotton. Morgenroth has substituted the simpler method of Intracardial intracardial inoculation. The point of maximum pulsation of Injection, the heart to the left of the sternum is made out by palpation and a thin sharp needle is inserted at the specified area. The spurting of blood indicates that the needle is within the heart. Thereupon the already filled syringe is carefully fitted on to the needle and the con- tents slowly injected. The syringe is then detached from the needle and blood is again allowed to spurt out in order to be absolutely convinced that the needle was still in the heart. It is next quickly withdrawn. By this method it is possible to inject about 1 1/2 c.cm. directly into the blood stream. In dogs, sheep, goats, horses, etc., the intravenous injection is given into the jugular vein directly through the skin which must be thoroughly shaved, cleaned and dis- .infected. Compression by the finger makes the vein more prominent. In dogs the popliteal vein is frequently selected. In man the intravenous injection is given into one of the veins on the anterior surface of the elbow joint. Several general rules are to be observed when giving intravenous in- oculations. First of all, only small quantities of fluids should be injected; secondly, the temperature of the fluids for injection should not differ from that of the body; thirdly, substances that are strongly hemolytic may pro- duce marked disturbances or even sudden death of the animal; fourthly, if an animal is to be frequently inoculated it is best to puncture the vein for the first inoculations as far peripherally as possible and give each subse- quent injection more centrally, for very often thrombi are formed at the site of inoculation. INTRAPERITONEAL INJECTION II INTRAPERITONEAL injection is employed most frequently Intraperito- among rabbits and guinea-pigs. The main danger associated neal with this method is possible injury to the intestines; but by Injection, heeding the following advice, this can be prevented. The animal should be fixed or held head down. In this position, the loops of intestines tend to sink toward the diaphragm. This is further helped along by gentle downward massage over the abdomen thus leaving an area, above the bladder, which is sometimes free from intestines. Another protective measure consists in using a blunt canula which can be made by breaking off the sharp point of the needle. As it is at times difficult to pierce the skin with this blunt instrument, it is advisable to pre- viously make a minute incision through the cutis and subcutis with a sharp pair of scissors and pass the needle through this small opening. The needle should not be plunged directly into the peri- toneal cavity, because at the with- drawal, the injected fluid easily escapes through the punctured opening. First, it is inserted sub- cutaneously upward, in the long di- rection of the animal ; then the hand is raised and the needle forced horizontally forward through the peritoneum, thus leaving the open- ing in the peritoneum at a different level than the one through the muscles and fascia, thereby making the escape of fluid more difficult. One readily realizes that he has gone through the peritoneum by a relaxation of the reflex ab- dominal rigidity (Fig. 4) . For the intraperitoneal injection in guinea-pigs, Friedberger has devised a procedure which is very satisfactory and furthermore does away with the necessity of an assistant. It can also be employed in Pfeiffer's test for the removal of exudates from the peritoneal cavity. The guinea-pig is allowed to creep into the breast pocket of the laboratory gown until its head and thorax are inside of the pocket. Its hind legs are grasped between the middle and ring fingers of the left hand and flexed on its back, thus giving a free exposure of the lower parts of the abdomen (Fig. 5). FIG. 4. — Intraperitoneal injection of rabbit. (After Uhlenhuth.) 12 LABORATORY EQUIPMENT SUBCUTANEOUS inoculation is the simplest of all methods. A fold of skin is elevated between the thumb and index finger °* tne ^e^ nan<^ an(^ tne needle plunged into the subcutaneous tissue. In rabbits and guinea-pigs the skin of the back or abdomen is chosen, as the subcutaneous tissue here is not tense. In goats, sheep, and horses the skin of the neck and shoulder region is preferred. Sub- cutaneous Injection. FIG. 5. — Removal of peritoneal exudate. Guinea-pig held in Friedberger's position. (Original.') The skin of the back and abdomen is to be avoided because following the injection edema frequently arises, which may extend to the lower extrem- ities and thus interfere with locomotion. If abscesses arise after subcutaneous injection, they should be opened, washed out with lysol solution and dressed with iodoform gauze. The Methods of Obtaining and Preserving Serum. Venesection or venous puncture is the method best adapted for obtain- ing blood from animals. The veins employed for that purpose are those OBTAINING BLOOD IN MAN which have already been mentioned in connection with intravenous in- jections. A simple, large, hollow needle is all that is required. Suction with a syringe is superfluous. Only in Morgenroth's method of removing blood directly from the heart of guinea-pigs is aspiration necessary. From rabbits enough blood can be collected by making an incision into the vein along its long axis with a sharp knife, or by dividing the vein transversely with the scissors. The blood thus collected is not absolutely sterile. In man, if only a small quantity of blood is required, it can be obtained from the finger or ear. If, however, a larger amount is necessary, puncture of one of the veins in the bend of the elbow with the Strauss canula is resorted to. It goes without saying that this area must be properly disinfected with soap and water, ether, alcohol or sublimate solution. Wright's method for collect- ing moderate quantities of blood will be reviewed in the chapter on opsonic studies. If the vein is prominent, the canula is thrust into the vein directly through the skin. Here the author has found it more convenient to point the canula upward, i.e., in the direction of the blood stream.1 In cases where the vein does not stand out it can be made to do so either by applying pressure with the finger upon its central part or placing a tight rubber bandage or rubber tube about the arm. This should not, however, be tight enough to obliterate the radial pulse. In very fat individuals, occasionally even these means may not suffice so that the vein must be dissected free and incised. After completion of the venesection the arm is elevated, slight pressure made upon the wound with sterile cotton and a bandage applied. If a small amount of blood is sufficient, and, as in most serological examinations absolute sterility is not essential, venesection can be replaced by the method of FIG. 6. — Puncture of vein. (Original.) Wet-cupping. wet-cupping. For this procedure a scarifier and Bier cup are required. The technique is as follows (Fig. 7). 1 The editor has found that more blood is obtained by thrusting the canula into the vein in the reverse direction. 14 LABORATORY EQUIPMENT Some part of the skin of the back is thoroughly disinfected and a well-fitting Bier cup firmly applied. Suction arises and the skin assumes a dark bluish-red appearance. After half a minute the cup is removed, the scarifier applied and the cutting edges set free. The scarifier is then reapplied, but this time at right angles to the previous incisions and the edges again set free. Suction is again made by the Bier cup and the blood is thus forced out from the multiple incisions. The blood obtained by any of the above methods is collected into a sterile vessel (graduate, flask, test-tube) and allowed to coagulate. The clot is then separated from the sides of the vessel by a sterile glass rod or platinum needle, the vessel plugged with absorbent cotton and placed into the ice chest. After 12 to 24 hours the serum begins to separate out from FIG. 7. — Obtaining blood by the wet-cupping method. the clot. If the serum is required immediately, the blood is allowed to flow directly into centrifuge tubes, the clot separated from the sides and the tubes centrifugalized. With a well-regulated centrifuge serum appears after several minutes. There are several rules to be kept in mind when using a centrifuge. Rules for i. The machine must be well oiled. the Use of 2. The counterbalance must be absolutely of the same weight. Centrifuge. 3. The centrifuge should never be suddenly stopped, but allowed to do so of its own accord. 4. In starting it, the motor should be gradually turned on. 5. If the centrifuge is slightly out of order it should not be used, but repaired at once, otherwise it may be ruined forever. 6. One should never centrifugalize with cotton plugs in the test-tubes. If the latter must be sealed, rubber stoppers should be used. METHOD FOR PRESERVING A SERUM 15 The color of a serum is greatly variable, depending mainly upon its hemoglobin or fat content. Blood taken at the height of the period of digestion shows a chylous serum. The serum of nursing women contains milk, that of icteric people contains bile. For most serological examina- tions these elements in the serum are inert, and do not interfere with the reading of the results. In precipitin reactions, however, the serum must be absolutely clear. If serum is to be kept for a long time, there are several ways that it may be retained without losing its activity. The method chosen depends upon the serum substance which is to be preserved. As will be pointed out again, substances are either thermostabile or thermolabile. The preservation of thermostabile substances (agglutinins, amboceptors) is usually very simple. It is sufficient to place the clear serum, which has separated from the clot, into a sterile test-tube plugged with absorbent cotton, and to put it into the ice chest away from the light. To reassure its perfect preser- vation one may add to it some phenol in such proportion that the carbolic is present to the extent of a 1/2 per cent, solution, e.g., to 9 c.cm. of serum add i c.cm. of a 5 per cent, phenol solution. The latter should be added drop by drop and ag- itated, so as to avoid the formation of precipitates. Another method, which the author employs almost exclusively for the preservation of sera containing amboceptors, consists simply in heating the sera at 56° C. for a half hour and then placing them into the ice chest. Inactivation has the advantage of stopping molecular changes produced by ferment actions of fresh serum. Furthermore, heating acts as a sterilizer for iso- lated air germs which may have found their way into the serum during the process of getting it. In this form, a serum can be kept in the ice box for several weeks without any material change. Occasionally one finds that a serum will undergo Tube used contamination in spite of inactivation, so that if a serum is to for preser- be preserved for several months, it is advisable to seal it in a vation of test-tube. For this purpose a brown glass tube slightly drawn out at its upper end is employed (Fig. 8) . The serum is placed into this sterilized tube and then the latter is sealed in the flame at its narrow part. Bacterial and organ extracts are well kept in this way. The best method of preservation consists in evaporating the serum to dryness in a vacuum desiccator. This procedure is rather complicated and can therefore be employed only in institutions. A vacuum desiccator with beatable plates is used. The serum is poured in very thin layers in sterile flat dishes and allowed to dry out in the desiccator at a tempera- ture of 30° C., later on at 35° C. in a vacuum of 3 cm. mercury. The dried serum i6 LABORATORY EQUIPMENT forms a yellowish-red horny mass which is scraped off from the dish and ground up in a mortar into a yellowish powder. The serum powder is then sealed in a brown glass tube. When this dried serum is to be used, the tip of the ampoule is broken off, and several drops of isotonic salt solution at a temperature of 30° C. are poured in, in just sufficient an amount to moisten the wall of the glass tube. By rolling the tube to and fro, one finds that the serum powder will easily stick to the moistened wall. The granules are allowed to swell up and after they have done so, enough isotonic salt solution is added to make up the original volume. For the preservation of thermolabile substances, the method of freezing has been suggested. Morgenroth has devised for this purpose a simple and handy apparatus named Frigo which can be obtained from Lautenschlager , Berlin. Although for most tests this method of preservation has been employed with success, Neisser's clinic reports that sera preserved in the Frigo with the idea of retaining their complement did not give as accurate complement fixation experiments as did similar fresh sera. Friedberger advises the addition of 8 per cent, salt solution for the preservation of the complement. When the serum is to be used it is diluted tenfold with distilled water, so that a 10 per cent, dilution of complement is obtained. By the addition of the salt, the resistance against harmful effects of light, room and body temperature, and chemical substances like phenol is increased, but the thermolability of the com- plement remains the same. Drying a serum in a desiccator is not to be advocated for the preservation of the complement, as during such procedure a portion of the complement is lost. Once the serum is in its dried form, however, the remaining complement is retained and in addition, has become resistant against high heat. Filtration of Bacteria. It is important in many serological studies to be able to separate bac- teria from their fluid media or suspension. This is accomplished either by centrifugalization or filtration. The first method does not completely FIG. 9.— Pukal filter. FIG. 10. — Filtration through a Pukal filter. free the fluid of its bacteria, but if this is desired the method of filtration is essential. In this connection, however, one must bear in mind that many albuminous, or albumen-like substances, few colloids and even some toxins, do not pass the filters and are therefore held back. Bacterial filtration is simplified by preliminary centrifugalization or passing the fluid through FILTRATION OF BACTERIA 17 filter-paper. Different porous materials have been used for bacterial filters, of which especially suitable are porcelain, infusorial earth and as- bestos. The filtration apparatus consists of the respective filter and the receptacle which receives the filtrate. Filtration takes place by differ- ences in pressure, where either the fluid is forced through by high pressure or sucked through by a vacuum formed in the receiving vessel. The following are some of the filters most commonly in use. 1. CHAMBERLAIN'S CYLINDER FILTER, F, used in the Pasteur Institute at Paris. The filter cylinder is made of infusorial earth and may be attached to any water outlet. 2. PUKAL FILTER, made of burnt kaolin, is used especially for the filtration of large quantities of fluid. The filter b is placed into the beaker e containing the toxin and bacterial fluid. The filter is then closed by a rubber stopper, perforated by a central opening through which runs a glass tube bent at right angles, and this in turn is con- nected with either an air or water pump for producing a vacuum inside of the filter. Between the filter and vacuum pump can be interposed a sterile jar a (Figs. 9 and 10). FIG. ii.— Reichel filter. FIG. 12. — Lilliputian filter. 3. THE REICHEL FILTER (Fig. 1 1) consists of a glass receiver A, having a side neck c and at' the bottom a tube-like outlet d. A porcelain filter B fits into the glass jar and rests upon the margin of the flask by means of a broad collar. The point of junction is made air tight by means of a rubber cap with a central opening, through which the cylinder can be filled. When in use d is shut off by a rubber tube with a pinch cock and c in which lodges a small piece of cotton is connected with a water pump that is instrumental in bringing about a vacuum. The function of d is to allow the removal of samples of the filtrate and finally to obtain the entire filtrate. 4. THE LILLIPUTIAN FILTER, candle-like in shape, and made of infusorial earth, is employed for the filtration of very small quantities. The filter is cemented upon a metal tube which is screwed, so that it is air tight, into a well-fitting glass cylinder open at the top. The tube is passed through a rubber cork which tightly closes an exhaust i8 LABORATORY EQUIPMENT flask. The fluid to be filtered is placed into the glass cylinder and sucked through into the flask by means of a vacuum produced here. For the purpose of collecting very small quantities a test-tube may be placed into the exhaust flask (Fig. 12). Preparation of Dilutions and Measurement of Small Amounts of Bacteria. All serological methods are to be considered on quantitative bases. In serum diagnosis as well as in the therapy, the amount of the serum- employed is the deciding factor. Similarly, the number or amount of bacteria required either for the purposes of immunization or serological reactions is of ex- treme importance. One cubic centimeter is the unit of measure for serum and all fluid mate- rial (Bouillon cultures, exudates, etc.). // small quantities are required, it is best to dilute the fluid with 0.85 per cent, saline solution. The exact prepa- ration of dilutions is one of the most essential technical procedures in serum diagnosis. Some general rules may be of help. 1. Never should amounts less than o.i c.cm. be measured out directly. For beginners even o.i is best measured in the form of a dilution, as errors are apt to occur very easily. 2. The decimal system should be adhered to as much as possible. 3. The dilution should be made just before it is to be used, inasmuch as many substances retain their activity best, or only, in concentrated form. The following is an example of correct forms of dilutions: Toxin Dilution of toxin, Dilution of toxin, Dilution of toxin, i no i :ioo i : 1000 o.i c.cm. = i c.cm. 0.05 c.cm. = 0.5 c.cm. o.o i c.cm. = o. i c.cm. = i c.cm. 0.005 c.cm. = 0.5 c.cm. o.ooi c.cm. = o.i c.cm. = 1.0 c.cm. 0.0005 c.cm. = 0.5 c.cm. The stock dilution of i : 10 is made by measuring off i c.cm. of toxin and adding 9 c.cm. of 0.85 per cent, of saline. The dilution i : 100 can be made by taking i c.cm. of toxin and adding 99 c.cm. of saline. It is more practicable, however, to take i c.cm. of the i : 10 stock dilution and add 9 c.cm. of saline. If the dilution i : ib is not present and only a small amount of the dilution i : 100 is desired, the latter is made by taking o.i toxin : 10.0 NaCl sol. Similarly 1:1000 = 0.1:100=1 c.cm. of the dilution (i : 10) :ioo=i c.cm. of the dilution (i :ioo) :io.o. PREPARATION OF DILUTIONS The following table shows the details of various dilutions: Dilution i : 10 Dilution i : 100 Dilution i : 1000 Dilution i : 10,000 T c.cm. +9 c.cm. NaCl i c.cm. +99 c.cm. Na- i c.cm. +999 NaCl = o. i c.cm.+999.9 c.cm. sol. = o.i c.cm. +0.9 Cl sol. = o. i c.cm.+ o. i c.cm.+99.9 c.cm. NaCl sol. = c.cm. NaCl solution. 9 . 9 c.cm. NaCl solu- NaCl= i c.cm. of dilution = 0.2 c.cm. +1.8 c. tion. o. 2 c.cm. + 199 . 8 c.cm. i : 1000 cm. NaCl solution. = 0.2 c.cm.+ i9.8 c.cm. NaCl sol. +9 c.cm. of NaCl sol. =0.3 c.cm. + 2.7 c.cm. NaCl solution. = 2 c.cm.+ i998 c.cm. =0.1 c.cm. of dilu- NaCl solution. = 0.3 c.cm. +29. 7 c*.cm. NaCl sol. tion i : 100 NaCl sol. = i c.cm. of dilution +9.9 c.cm. NaCl sol. = 3 c.cm. +27 c.cm. = i c.cm. of dil. i : 10 i : 100+9 c.cm. NaCl = i c.cm. dil. i : 100 NaCl sol. +9 c.cm. of NaCl. sol. sol. =0.1 c.cm. dil. +99 c.cm. NaCl. sol. = 10 c.cm.+go c.cm. = 2 c.cm. of dil. i : 10 i : 10 = 0.1 c.cm. dil. i : 10. NaCl sol. + 18 c.cm. NaCl sol., +9.9 c.cm. NaCl sol. +99 . 9 c.cm. NaCl. sol. etc. = o. i c.cm. of dil. i : 100+0.9 c.cm. ' NaCl solution. In preparing these dilutions, it is best to measure off the small quantities o.i- i c.cm. with a pipette; allow this to run into a well-graduated measuring glass and add enough saline to make the required dilutions. For example, if 30 c.cm. of a dilution i : 100 is desired, 0.3 c.cm. should be measured off with a pipette and aUowed to flow into a 50 or 100 c.cm. graduated cylinder and saline solution added up to 30.0 c.cm. It should always be one's aim to get along with small quantities of the substance to be diluted. If, for example, 8 to 10 c.cm. of a toxin dilution i : 100 are required o.i c.cm. of toxin +9.9 c.cm. of saline should be taken and not i c.cm. of toxin and 99 c.cm. of NaCl sol. Before making any dilution one should always calculate the total amount of sub- stance required; as for example in the following experiment: 1. Animal o.i c.cm. Toxin subcutaneously 2. Animal 0.05 c.cm. Toxin subcutaneously 3. Animal o.oi c.cm. Toxin subcutaneously 4. Animal o.ooi c.cm. Toxin subcutaneously Here, the total quantity of toxin necessary is found, by adding, to be 0.161 c.cm. This represents the minimum amount. It is always advisable to make an allowance for some loss and at the same time bring up the amount to a round or even number. 0.2 c.cm. of toxin would fulfill all these requirements. This amount is measured off by a pipette, placed into a graduated cylinder and saline added up to 2.0 c.cm., making a dilution of i : 10. Then 0.2 c.cm. of this dilution (i : 10) is taken, placed into another graduate, and again diluted with saline up to 2.0 c.cm. thus making a dilution of 1:100. The above problem therefore of injecting the various animals, ran be com- pleted as follows: 1. Animal receives i c.cm. of dilution i : 10 2. Animal receives 0.5 c.cm. of dilution i : 10 3. Animal receives i c.cm. of dilution i : 100 4. Animal receives o.i c.cm. of dilution i : 100 20 LABORATORY EQUIPMENT The unit for measuring the amount of bacteria grown upon a solid medium is represented by a standard sized loop. This platinum loop takes up about 2 mg. of bacterial substance. It is prepared as explained in Fig. 2. If smaller amounts of bacteria are used, dilutions must be made. For instance. 1/4 of a loopful of bacteria is desired; i loopful is suspended in i c.cm. of saline and 3 c.cm. of saline added. As a result, i c.cm. of this emulsion contains 1/4 of a loopful of bacteria. If 1/16 of a loopful is necessary i c.cm. of the above dilution is added to 3 c.cm. of saline, thus making i c.cm. of this last mixture contain 1/16 of a loopful of bacteria. CHAPTER III. ACTIVE IMMUNIZATION. IMMUNIZATION WITH LIVING AND DEAD VIRUS. Active immunization depends upon the principle, that an organism in overcoming a slight infection, either naturally or artificially ncip e acquired, develops enough protective bodies to withstand a of Active . ., . , . , ^. Immun- similar, severer, natural, or acquired infection. Moreover, ization. ^ serves primarily the purpose of prophylaxis. In labora- tories, active immunization of animals is also frequently under- taken with the view of obtaining sera for diagnostic and therapeutic purposes. In the manufacture of large quantities of serum, the horse is the animal used almost exclusively. Occasionally cows, sheep, donkeys or mules' are selected. In small laboratories usually rabbits, guinea-pigs, white mice, rats, and only occasionally goats or sheep are employed. The process of immunization evokes a marked disturbance in the gen- eral health of the animals. For this reason they must be well kept in warm places, and well fed. As far as their power of producing antibodies is concerned, there are individual differences even among the same species of animals; thus if five horses are immunized against diphtheria, some will give much better curative sera than the others. In general, the younger animals are preferable. Any substance which, when injected into an organism, can stimu- late the production or formation of antibodies, has been conveniently termed " antigen." After the injection of such an antigen, special notice should be taken of the animal in reference to temperature, weight, the excitation of diarrhea or the occurrence of abscesses, infiltrates, edema or paralysis. If an animal dies, a careful postmortem, and if possible, a bacteriolog- ical examination should be made. It should be the aim to ascertain if death was induced by the inoculated antigen, by contamination or secondary infection. One should always keep in mind the possibility of some of the animal epidemic diseases. Epidemic diseases occurring in rabbits are: Animal i. RABBIT SEPSIS. — Presents itself in the form of bronchopneumonia and Infections, marked nasal catarrh. It is very infectious. Sick animals should at once be isolated or killed and their cages thoroughly disinfected. 21 22 ACTIVE IMMUNIZATION 2. COCCIDIOSIS gives changes in the liver due to the settling of the coccidi ova forms. The parasites are present in the pus and are easily recognized microscopically. Following labor, guinea-pigs are very susceptible to sepsis. IN RATS, trypanosomiasis is of frequent existence, but is not pathogenic. The antigens are injected either subcutaneously , intraperito- The Tech- neally or intravenously. Only on exceptional occasions is nique of another entrance path chosen. Active Im- As regards the amount to be injected, one cannot very well mumzation. g|ve generai rules. It is important to prevent severe re- actions, although the question is still a disputed one, whether marked reactions tend to produce a better immunity. It is certain, how- ever, that inoculations of antigens in such minute doses as to apparently give no reaction, can still lead to immunity and the production of antibodies. Occasionally a single injection suffices for immunization. Repeated inoculations are usually necessary, especially so when a "highly valent" serum is desired, i.e., one containing a great number of antibodies or having high protective properties. When repeated inoculations are undertaken, there are various methods of procedure. i a. A small dose of antigen is injected. If a reaction sets in, one waits until this reaction has entirely subsided, then (not before the fifth day) the second injection — a somewhat larger dose — is given. After an interval of 5 to 8 days, a third injection of a still higher dosage is administered, and so on, again. ib. The intervals are the same, but the amounts of antigen remain the same at each injection. Both of these methods give excellent results and therefore are most frequently used. 2. For several successive days, a small or medium dose of antigen is injected. Each injection produces only a slight reaction. This last scheme according to Fornet is especially suitable for obtain- ing precipitation sera. As is evident, it has the advantage of gaining the immunity rapidly. 3. Inoculations are given at very long intervals (intermissions of four weeks or more). This method produces good sera, but has the disad- vantage of requiring too long a time. The methods of active immunization can also be divided according to the nature of the antigen. 1. Immunization with a living virus, 2. Immunization with a dead virus, 3. Immunization with bacterial extracts, 4. Immunization with bacterial toxins. CLASSIFICATION OF BACTERIA I. Immunization with a Living Virus. 23 This method of immunization simulates most closely the immunity attained spontaneously in overcoming an infection. This immunity is very strong and lasts for a long period of time, but it is attained with difficulty; frequently the dose of virus injected causes serious symptoms of infection. Various procedures have therefore been advocated so to di- minish the toxicity of the immunizing agent that only immunization effects, and no toxic symptoms be obtained. This was attempted either by the reduction of the number of organisms employed, so that very minute doses were inoculated, or by the diminution of the infectious nature of these bacteria (virulence so called). The first method, however, was not found applicable to all cases. The infectious nature of the different bacteria varies markedly. The same bacterium reacts differently with different animals. While some animals possess a natural immunity against certain bacteria, others exhibit a distinct susceptibility to the same micro-organisms. The conceptions therefore of pathogenicity and virulence are purely of a relative nature. ' In talking of the pathogenicity of bacteria, one should always mention the class of animal for which these bacteria are pathogenic. Bail has used this principle of pathogenicity in classifying bacteria. He Bail's Classi- mentions the following three classes: fication of a. Saprophytes. Bacteria, b. Half or partial parasites. c. Whole or pure parasites. To the class of saprophytes belong all those bacteria which when injected even in larger doses do not produce any characteristic disease; these are also known as apatho- genic — e.g., hen cholera bacilli for human beings. Classed as half parasites are those bacteria, according to Bail, the infectious nature of which depends upon the quantity of bacteria injected. While the injection of a rabbit with i/iooo of a loopful of a typhoid culture will produce no evidences of disease, one-tenth of a loopful will result in slight increase in temperature, loss of appetite, and eventually a local redness at the site of the injection. One loopful may bring about the death of the animal. The manifestations are dependent entirely upon the number of bacteria injected. The smaller the number, the milder the symptoms, until one reaches the stage below which no disturbances at all are visible. Pure parasites are those which have no sublethal dose. Even the smallest amount, when injected, will produce death. As examples, the tubercle bacillus for guinea-pigs, and bacilli belonging to the group of Hemorrhagic Septicemia for rabbits. Of the last mentioned 1/10,000,000,000 of a loopful of some cultures kills a rabbit within twenty- four hours with the symptoms of a septicemia; in other words, the injection of i c.cm. of a dilution of , one loopful of culture in ten million liters of water suffices to kill the rabbit. Furthermore, the bacteria increase so greatly in the body of the rabbit that they can be demonstrated in every drop of blood and in all organs and body fluids. The same organism is a saprophyte for the human being and a half parasite for the guinea-pig if injected subcutaneously and a complete parasite by intraperitoneal injec- 24 ACTIVE IMMUNIZATION tion. The conceptions therefore of complete or partial parasite as well as of saprophyte are only relative and are dependent upon the bacteria, the animal species, and the mode of infection. It is now clear that immunization with living bacteria can Example of onty ^e undertaken if the latter belong to the class of half- Active Im- parasites. Pure parasites are excluded from this method, munization As an example of such procedure can be given the immu- with Living Cation of a guinea-pig by intraperitoneal injections with Bacilli. }iving typhoid bacilli. Preliminary to this, the virulence of the typhoid culture must be ascertained. a. Preliminary test to titrate the virulence of the typhoid culture. 1. Guinea-pig i./I. 1909 1/20 loopful of typhoid culture intraperitoneal. 2. /I. active. 8./I. alive. 2. Guinea-pig i./I. 1909 i/io loopful of typhoid culture intraperitoneal. 2. /I. active. 8./I. alive. 3. Guinea-pig i./I. 1909 1/8 loopful of typhoid culture intraperitoneal. 2./I. slightly sick, does not eat. 3./I. active. 8./I. alive. 4. Guinea-pig i./I. 1909 1/6 loopful of typhoid culture intraperitoneal. 2./I. sick, does not eat, hair raised. 3./I. still sick. 4./I. more active. 8./I. alive. 5. Guinea-pig i./I. 1909 1/5 loopful of typhoid culture intraperitoneal. 2. /I. sick, does not eat, hair raised. 3 ./I. very weak, when placed on side remains so. 4./L f 6. Guinea-pig i./I. 1909 1/4 loopful of typhoid culture intraperitoneal. ,./!. f From this experiment it becomes evident that the lethal dose of this particular strain of typhoid culture is 1/4 to 1/5 of a loopful for guinea-pigs by intraperitoneal injection. Immunization therefore must be started with a smaller dose — e.g., i/io of a loopful. b. Immunization. 1. Guinea-pig 8./I. 1909 i/io loopful of typhoid culture intraperitoneal. i6./I. 1/8 loopful of typhoid culture intraperitoneal. . 2 2. /I. 1/4 loopful of typhoid culture intraperitoneal. Animal remains active and healthy. 3O./I. i loopful of typhoid culture intraperitoneal. 5-/II. 2 loopfuls of typhoid culture intraperitoneal. Animal remains active and healthy. 2. Guinea-pig 8. /I. 1909 i/io loopful of typhoid culture intraperitoneal. i6./I. 1/4 loopful of typhoid culture intraperitoneal. Animal remains active and healthy. VACCINATION AGAINST SMALL-POX 25 3. Guinea-pig 8./I. 1909 i/io loopful of typhoid culture intraperitoneal. i6./I. 2 loopfuls of typhoid culture intraperitoneal. I7./I. animal is sick and does not eat. i8./I. animal is very weak. I9-/I. t Control animals always die within twenty-four hours, as in previous experiment, on injection of 1/4 of a loopful. From experiment with guinea-pig i, it can be learned, that by gradual increase of the immunizing dose, a state of immunity is reached which can overcome an infection produced by a high multiple of the dosis letalis. Experiments 2 and 3 prove that even a single preliminary injection suffices to pre- vent the death of an animal upon subsequent receipt of the lethal dose of the same bacteria; but this single inoculation is not sufficient to protect the organism against a very severe future infection. The attained immunity is therefore only relative, not absolute. Analogously it is possible to immunize by subcutaneous and intra- venous injections. The latter method is usually the one of choice when half parasites are employed, as the highest and quickest grade of immunity is thus reached. It carries with it, however, the greatest danger, and fre- quently results in death of the animal. The method of immunization with small doses of living, fully virulent bacteria, has thus far been made use of only in animals. In man this experience has not been carried into effect. It is j eared that the bacteria may increase very rapidly and give rise to severe disturbances. The method has therefore been altered and instead of using virulent material for immu- nization, only a weakly infectious or attenuated virus is employed. Vaccination against Small-pox. This is the best known example of active prophylactic immunization. To Jenner belongs the credit of having been the first one to apply this principle. Vaccination against small-pox consists in inoculation of an attenuated form of small-pox germs, the diminution in virulence being brought about by pass- age through the body of a calf, a less susceptible animal than man. The vesicles formed on the vaccinated person contain these attenuated germs. This lymph can be used for the inoculation of other individuals, as the germs do not regain their virulence by repassage through man. Inasmuch as it is not within the scope of this book to go into the details of the preparation of the lymph or the technique of vaccination, a brief survey of the benefits of vaccination will amply suffice and this may be seen from the table hereunto appended. The mortality from small-pox per 100,000 population was in the years 26 ACTIVE IMMUNIZATION in Prussia and Bavaria in Austria in Belgium in England in Sweden . . 1882-1896 0.7 38.6 18.2 2.9 1862-1876 Si.6 75-2 79-5 25.3 26.9 0.5 This method of immunizing against a virulent virus by inoculating with an attenuated form of the same, is known as Jennerization. Pasteur recog- nized that this method had general application and similarly used attenuated but still living cultures, "vaccins" so called, to immunize against hen cholera, swine plague, and anthrax. The same principle underlies Pasteur's anti- rabic vaccination. Antirabic Vaccination. In all civilized countries there exist special institutions, either directly under the city control or appointed by the city, where the Pasteur treat- ment for rabies is conducted. It is the duty of the general practitioner, on getting a suspicious case of rabies, to advise his patient to undergo this special therapy and to send the rabid animal, its head or brain preserved in glycerin, to the institute as soon as possible for the purpose of ascer- taining the presence of rabies. Up to very recently the actual cause of hydrophobia was unknown. Negri had described parasites, known as Negri bodies, in the large nerve cells of the cerebral cortex, cerebellum, etc.1 In man, infection usually occurs as a con- Treatment se°iuence °f the saliva of rabid animals (dog, cat, wolf, skunk) gaining entrance to wounds from bites or scratches. Roux and Nocard found that the saliva of experimentally infected animals is already infectious 2-3 days before the first symptoms appear. Thus, rabies may be transmitted to an individual by an animal apparently healthy at the time. Remlinger has formulated the following very instruc- tive outline referring to the indications for antirabic treatment. 1. has died within 10 days after the biting, 2. has been killed within 10 days after the biting, 3. has disappeared within 10 days after the biting, 4. is unknown to the individual bitten, (a) becomes ill with rabies, (b) dies with suspicious symp- toms of rabies or of another disease, If the biting animal 5. has remained alive and under observa- tion for 10 days, antirabic treatment is indicated. antirabic treatment is indicated. (c) becomes ill but does die within 10 days, not (d) remains well both during and after this period, 1 Noguchi has now succeeded in artificially cultivating the rabic virus. further observation, treatment if animal dies. no treatment. By inoculating cultures containing these granular, pleomorphic or nucleated bodies, he has reproduced rabies in dogs, rabbits and guinea-pigs. PASTEUR TREATMENT FOR RABIES 27 Pasteur found that rabies can be transmitted to dogs by injecting them subdurally with the brain substance of rabid animals. This ordinary virus substance is known as Street Virus. The incubation period of rabies is very long. It varies from about three weeks to [possibly] some years. By passing the virus through monkeys, the incubation period is considerably increased. After the successive passage through five or six animals, the virus becomes so weakened that infection is almost impossible. Reversely, increase of the virulence may be affected by passing the virus through a successive number of rabbits, as these are very sensitive to the disease. After passage through a large number of such animals, the incubation period is gradually shortened from about three weeks or a little less to a constant period of six or seven days. Further diminution in the period of incubation was impossible and there- fore Pasteur called this " Virus fixe." His first experiments in immuniza- tion were made by passing the weakened monkey virus through rabbits and then treating dogs with the spinal cords of the latter. Later on, Pasteur discovered that instead of passing the virus through monkeys, he could diminish its virulence by drying the spinal cords derived from rabid animals, for varying periods of time. In this way he could prepare an entire series of graduated strengths. The material used for this drying was not the street virus, but that obtained by successive pas- sage through rabbits or "virus fixe" which possessed very constant im- munizing and infectious properties. By drying the "virus fixe" over caustic potash at a temperature of 23° to 25° C. for five days, its regular incubation period of 7 days was very much prolonged. Increase in the length of drying caused the entire loss of virulence. Pasteur immunized dogs as follows: He began with the injection of a virulent spinal cord which had been dried for thirteen days and every following day injected subcutaneously some fresher spinal cord, i.e. (dried for a lesser period of time), until finally he used virus dried only for one day. The animals thus treated were immune against the bites of rabid dogs as well as subdural, subcutaneous, and intravenous infection with "virus fixe " and street virus. This procedure was strongly recom- mended by Pasteur, who brilliantly contributed the observation, that if an animal was' infected but did not as yet show symptoms, these could be prevented by a similar modus operandi, as above mentioned. In man, the inoculation is carried out on the same principle. The fact that the incubation period of hydrophobia is very long, makes the prophylactic inoculations of greater service. Only rarely is this period less than six weeks, usually considerably longer — up to 584 days, entirely dependent upon the virulence of the virus and the point of infection. 28 ACTIVE IMMUNIZATION Technique of Antirabic Vaccination in Man. The actual vaccine consists of i c.cm. (2-3 mm. length) of the substance of the spinal cord of a rabbit which has been killed by inoculation with the fixed virus, rubbed up into a fine emulsion with 5 c.cm. of sterile 0.85 NaCl solution. About i to 3 c.cm. of the resulting fluid are injected subcutane- ously into the skin of the abdomen. A cord dried for fourteen days is used for the first injection, emulsions of less attenuated virus are used on suc- ceeding occasions until finally a portion of a spinal cord dried for only three or four days is employed. Pasteur's schemes of the actual doses can thus be drawn up. a. For infections at points distant from the central nervous system (mild infections). Day of injection i 2 c; 6 7" 8 IO II 12 1-2 14 1C 14 12 10 8 6 5 5 4 3 5 5 4 4 3 3 Number of days cord was dried . . . + + + + + 13 II 9 7 6 Amount injected in cubic centi- 3-0 3-0 3-0 3-0 2.0 I.O I.O I.O I.O 2.O 2.O 2.O 2.0 2.0 2.O meters. b. For head wounds (severer infections). Day of injection. i 2 3 4 5 6 7 8 Number of days cord was dried. 14 12 10 8 6+6 5 5 4 3 4 13 II In A. M. In p. M. 9 7 In A. M. In p. M. Amount injected in cubic centi- meters. 3-0 3-0 A. M. P. M. 3-0 3-0 A. M. P. M. 2 c.cm. 2 c.cm. A. M. P. M. 2. 2. 2. i. 2. Day of injection TO TT 14 T f 1 6 18 TO Number of days cord 5 5 4 4 3 3 5 4 3 5 4 was dried. Amount injected 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. TECHNIQUE OF ANTIRABIC VACCINATION 29 The drawback to this classical method of Pasteur consists in using the virulent material rather late in the course of the inoculations. A more energetic treatment has therefore been advised. There is no added danger in doing this because the virus fixe in contrast to the street virus is not at all or only slightly infectious for man. Hogyes in Buda Pesth uses the virus fixe right from the start. He be- gins with marked dilutions (1/10,000) and gradually increases them to i/ioo. The theory underlying this procedure is, that the usual method of attenuation by drying alters the quantity of the virus but not its quality; hence the same result may be obtained by simple dilution. Ferran successfully employs the virulent virus in large doses right from the onset of the treatment. Especially in very severe infections, as in bites from wolves, is this procedure justifiable. The exact arrangement of doses varies a little at different institutions. In Berlin, it is considered that the virulence of the dried cord is lost on about the eighth day instead of the fourteenth. Hence in the hydro- phobia department of the Berlin Institute for Infectious Diseases, the authorities have adopted the following scheme, which stands midway between Pasteur's classical method and the extreme procedure of Ferran. Scheme for treatment of mild infections: Day of injection . . i 2 3 4 S 6 7 8 9 10 II 12 ** 14 15 16 17 18 19 20 21 Number of days 8-7-6 5-4 4-3 5 4 3 3 a 2 5 5 4 4 3 3 2 2 4 3 2 2 cord was dried. Amount injected 0.5 of i. 5 of 2.O 3-0 3- i.S 2 i I 2 2 2 2 2 2 I-S i-5 2 2 i-5 2 in cubic centi- each each 1.0 meters of an emulsion i c.cm. of cord in 5 c.cm. of sterile bouillon. Scheme for treatment of severe infections: i 2 3 4 5 6 ;7 8 9 10 II 12 X3 14 IS 16 17 18 19 20 21 Age of cord 8-7-6 4-3 5-4 3 3 -3. 2 '*•' 5 4 '•4; 3 3' 2 2 4 3- 2 2 3 2 Amount . . i-5 i-5 i-5 2 2 i I i 2 2 2 2 2 2 2 2 > ;2 2 2 2 ACTIVE IMMUNIZATION In severe injuries the entire treatment is repeated after one month's interval. There is at present no doubt whatsoever as to the value of these antirabic vaccina- tions. Compiled from a great number of statistics, the mortality of those infected or exposed to infection but untreated is 15 to 16 per cent., while the death rate of those treated at the Berlin Institute during 1898 to 1901 was 0.55 per cent. Similar figures are given by the other institutions. The serum of individuals who have taken the Pasteur treatment con- tains antibodies that can neutralize the toxic effects of the rabies virus. If a virulent strain of the latter is mixed with the serum and injected into an animal, no symptoms will develop. A similar serum is manufactured at the Pasteur Institute by the intravenous injection of sheep with emul- sions of the street virus or virus fixe. It is used mainly for immunization of animals by the so-called " simultaneous method/' whereby mixtures of a virulent virus and the serum are employed. In this way an immunity is attained much more rapidly than by the classical method. A. Marie and Remlinger have made use of this simultaneous method also in man, with good results. It is of value especially when a rapid immunity is essential as in severe infections (bite of wolf, injuries of the face) or neglected cases. The time elapsed between the injury and the onset of treatment is an im- portant deciding factor as to the final result. The following table of Diatroptoff demonstrates this: The bitten individual started treatment during the Number of patients treated Of these died Percentage First week 4602 26 o ^6 Second week QQI 16 1.66 Third week . •217 10 7 IQ The immunity attained by vaccination is of comparatively short duration; probably only several years. It is advisable therefore in case of reinfection not to depend upon the previous immunity but go through another treatment. Attempts have been made to employ this principle of virus attenuation for other infections. Behring and Koch tried immunization aSamst bovine tuberculosis by inoculation with living human Tuberculosis tubercle bacilli. These can be bought under the name of Bcvovaccine (v. Behring) and Tauruman (Koch). L Tauruman is prepared by the Hochst Farbwerke and is put up in sealed glass tubes which contain 0.02 to 0.04 gm. of living tubercle bacilli suspended in 10 c.cm. of normal saline solution. This Tauruman is previously examined in Ehrlich's Institute and note is taken of its purity, quantity of bacteria, virulence against guinea-pigs and avirulence against rabbits (characteristics of the human type of tubercle bacilli). SENSITIZED BACTERIA 3! To this class of experimental work belong also the attempts of Fried- mann to immunize against human tuberculosis by the use of the tubercle bacilli of cold-blooded animals, and those of Wassermann, Ostertag and the author, to inoculate against hog cholera with living cultures of mouse typhoid. Besides the preceding mode of virus attenuation by passage through animals, there are other methods employed for the diminu- Other tion of the toxicity of the virus. Growing the bacteria at Methods of too high a temperature, or exposing bacterial emulsions to Vaccine light, disinfectants or moderate heating, accomplishes the Preparation. same purpose. The mixture of bacteria with their specific serum (i.e., serum obtained from animals that have been inoculated with these bacteria), also diminishes the virulence of the bacteria. Such bacteria are des- ignated by Bordet as " sensitized." In this mixture, the bacteria attach their specific antibodies; after 'centrifugalization, the added specific serum now devoid of its specific antibodies is removed, and the sensitized bacteria can be used as vaccines. Inoculations of the latter rarely produce any infiltration. The same object can be accomplished by injecting bacteria and at the same time also their specific serum. This is technically simpler and is known as the "Simultaneous Method." It has shown itself of great value in Lorenze's prophylactic inoculations against swine erysipelas. 2. Immunization with Dead Bacteria. — Immunization with dead bacteria was first undertaken by Toussaint, Salmon and Smith, and Chamberland and Roux. This method is to be distinctly separated from those already discussed. Bail claims that the immunization with living bacteria as well as by aggressins (to be mentioned later) is an immunization against the infectious disease; while the immuniza- tion with dead bacteria is an immunization against the bacterial bodies. While this holds true for some bacteria, it is, to say the least, questionable whether it can be considered a general rule. Whenever a real immunity is desired — that is, protection against disease, a vaccine either in the form of living or attenuated bacteria should be given the preference. Up to a certain degree, the extracts of living bac- teria, and the natural and artificial aggressins can be similarly employed. If, however, no real immunity, but just a serum containing a great number of antibodies is wanted, as in serum diagnosis, for agglutination, bacteri- olysis, complement fixation, etc., then immunization by dead bacteria is just as, if not more so, efficient. Recently, the question has been raised whether the antibodies pro- duced by immunization with heated antigens are identical with those ob- tained with unheated antigens. The experiments of Obermeyer and Pick, 32 ACTIVE IMMUNIZATION which will be referred to under proteid immunization, seem to prove that they are not alike. For laboratory work it is advisable to use living cultures only in cases of absolute necessity. In heating bacteria to destroy their virulence and thus be suitable for inoculation, we must be very careful not to raise the tempera- Death of ture to such a degree that not only the toxicity but also the Bacteria immunization power is destroyed. It is best to employ the by Heat, minimum amount of heat which will kill the respective bacteria. For most of these as Typhoid, Paratyphoid, Colon, and Dysen- tery bacilli, Cholera Vibrios, Meningo-, Staphylo-, Strepto- and Pneumo- cocci, one hour at 60° C. is sufficient. The bacteria are grown upon agar cultures and the required amount is removed and suspended in sterile physiological salt solution or bouillon. This suspension is then placed into a hot water bath or thermostat regu- lated at 60° C., for one hour.' If the bacteria employed are highly infectious, one must be sure that all bacteria have been killed. This must especially be noted when giving prophylactic inoculations in man. Several drops of the emulsion are therefore transferred to agar tubes and incubated for a day or two. If a growth appears, the emulsion is to be reheated; if not it can be considered sterile. The mode of immunization with dead bacteria is the same as has been described for the living ones. In general the dosage to be used may be larger. Small doses are injected at first, followed later on by increasing quanti- ties at intervals of five to eight days, e.g. Intravenous inoculation of a rabbit with dead typhoid bacilli. Result. — Protection against living virulent bacteria, appearance of agglutinins, bacteriolysins, bacteriotropins and complement binding substances in the serum. i ./I. 1909. Rabbit No. i. i loopful of a typhoid agar slant culture killed at 60° C. and injected intravenously. 6./I. 4 loopfuls of typhoid culture killed at 60° C. and injected intravenously. I2./I. i culture of typhoid killed at 60° C. and injected intra- venously. 20./I. Infection with i culture of the living typhoid bacilli injected intravenously. Animal remains alive. Rabbit No. 2. Control. 20./I. Infection: 1/4 loopful of living typhoid bacteria intra- venously. 22./I. f(death). The use of killed typhoid bacteria for prophylactic immunization has recently been widely adopted. This has been stimulated to a great degree by the successful experiments of Wright, and Pfeiffer and Kolle. PROPHYLACTIC TYPHOID INOCULATION 33 Wright's Method of Prophylactic Typhoid Inoculation. The vaccine originally employed by Wright for these inoculations con- sisted of highly virulent cultures of Bacillus Typhosus grown in broth for twenty-four to forty-eight hours (sometimes even for four weeks), and sterilized by heating at 60° C. The vaccine was then standardized, i.e., the strength of the vaccine was fixed in accordance with another of known strength, the dosage of which had been gauged by inoculations in man. The early form of standardization consisted in determining the toxicity of the virus. Guinea-pigs weighing 250 to 300 gms. were inoculated subcutaneously with 0.5, 0.75, i.o and 1.5 c.cm. of the vaccine respec- tively. Death to some of the animals would come in twelve hours to three days. The amount required to kill a guinea-pig weighing 100 gms. or rather the proportional fraction of the dose which proved fatal to the one of 250 to 300 gms. was taken as the standard dose for injection in man. Wright subsequently found that better results were obtained, if the vaccine was prepared from twenty-four hour cultures grown upon the sur- face of agar. The growth is then washed off in physiological Preparation, saline solution. This emulsion is sterilized by subjecting it to a temperature of 55° to 60° C. for one hour, after which the number of bacteria is computed by Wright's method (see Standard- ization of Vaccines in chapter on Opsonins). After this the emulsion is further diluted with physiological salt solution containing 0.5 per cent, carbolic or 0.25 per cent, trikresol in such a manner that two or three different concentrations are secured, one containing 500 million killed typhoid bacilli per c.cm., one containing 1000 million per c.cm., and one containing 2000 million per c.cm. The particular strain of typhoid bacillus employed for the vaccine varies. Some use a strain isolated by Leishman in 1900. This is selected not on account of its degree of virulence but on account of its property of being able to stimulate the formation of a great amount of antibodies. The editor and others employ a mixture of eight or ten different strains with perhaps one or two paratyphoid strains. The typhoid vaccine or typho-bacterin as it is frequently called is administered subcutaneously, usually in the arm at the insertion of the deltoid muscle. The needk should not enter the muscle or find its way between the layers of the skin. The arm should be cleansed as for any other injection with alcohol and iodine. The dosage almost uniformly employed consists of 500 million Dosage, bacteria for the first injection and 1000 million for each of the two subsequent injections at intervals of eight to ten days. If, however, such an extended period of time is not available, then two in- oculations will suffice, the first dose 1000 million and the second dose 34 ACTIVE IMMUNIZATION 2000 million. Wright used the two larger doses. An objection raised to this method is that the general reaction obtained is more severe. The editor, however, has employed these larger doses for inoculating the nursing and medical staff of the hospital and a great number of laymen, without any ill effect. Women and children should receive a dose in proportion to their weight; a healthy man weighing 150 pounds being designated as the stand- ard of comparison. Local and general reactions follow the inoculations. Thus local redness and swelling of the skin, lymphangitis and enlargement of the neighbor- ing glands are the usual consequences. The inflammation can at times be severe enough to simulate erysipelas. The general symptoms, on the other hand, may consist of a general feeling of illness, headache, little fever, and occasionally nausea, not infrequently accompanied by vomiting. These signs of indisposition, however, pass off rapidly without leaving any permanent ill effects. Debilitated persons frequently present the most profound reactions. Occasionally latent and chronic diseases of a non-typhoidal character may be made active by inoculation. These exacerbations are not serious and usually by diminishing the quantity and increasing the number of doses, these effects can be avoided. Six to eleven days after the injection, an increase in the number of agglutinat- ing, bacteriolytic and bacteriotropic bodies can be demonstrated in the blood of the inoculated individual. The immune bodies reach their height in two to three months after incculation, and then fall rapidly. In a series of forty cases Garbat was unable to demonstrate complement fixation bodies with any regularity in the blood after pro- phylactic injection. As to the results of antityphoid vaccination, opinion is somewhat divided. According to Wright's statistics infections have been diminished by about one-half, and in single series to one-sixth or even one-twenty- eighth of the former, or control number. The mortality too is much lower. Out of 1758 individuals who had been vaccinated, only 142 or 8 per cent, died; out of 10,980 who had not been, 1800 or 16.6 per cent, met death. The immunity attained is not absolute, for according to Russel, in 1911, among 80,000 p.ersons vaccinated in the United States Army there were twelve cases of typhoid with one death (due to intestinal hemor- rhage), and in 1910 six cases occurred with no fatalities. Had it not been for the prophylactic immunization, there would have occurred at the pre- vailing rates of incidence about 250 cases. Similar favorable statistics have been collected in England, Germany and France. The period of immunity lasts from two to three years. PROPHYLACTIC TYPHOID INOCULATION 35 Pfeiffer-Kolle's Experiments. Pfeiffer and Kolle prepare their vaccine by growing typhoid bacilli on agar cultures and suspending a twenty-four hours' growth in physio- logical NaCl solution. The normal platinum loop is the unit of standard- ization. A full grown agar culture is considered as 10 normal loops and as such it is diluted in 4.5 c.cm. of saline. This emulsion is placed in a thermostat at 60° C . for two hours and then tested for its sterility. Suffi- cient 5 per cent, phenol solution is next added to the suspension to bring the contents up to a 0.5 per cent, carbolic solution, and the final emulsion is again heated at 60° C. for thirty minutes. One c.cm. of the vaccine is thus equivalent to two normal loops of culture. The amounts of vaccine to be injected have not yet been definitely decided upon. The best dosage so far is the following: For the first injection: 0.3 c.cm. of the vaccine. For the second injection: 0.8 c.cm. of the vaccine. For the third injection: i.o c.cm. of the vaccine. The injection is made subcutaneously between the breast and clavicle. The local and general reactions are the same as those observed with Wright's method. As a result of the injection only increased agglutinins and bacteriolysins have been found in the blood serum. Bacteriotropins have not as yet been examined for. The effects of these inoculations seem to be very good. Protection is prolonged according to the increase in the number of injections, and if inoculated individuals do become infected, they run a very much milder course of the disease. The following statistics as given by Kuhn indicate the results: Inoculated. Non-inoculated. Very slightly ill 186 (50. 13 per cent.) 331 (36. 55 per cent.) Moderately ill 96 (25.88 per cent.) 225 (24.85 per cent.) Severely ill 65 (17.52 per cent.) 234 (25,80 per cent.) Deaths 24 ( 6.47 per cent.) 116 (12.80 per cent.) 371 (100 per cent.) 906 (100 per cent.) The' prophylactic immunity according to Kuhn lasts one year. Kolle has under- taken similar experiments against cholera. CHAPTER IV. ACTIVE IMMUNIZATION. Immunization with Bacterial Extracts.— Aggressin Experiments. The marked infectious nature of the organisms belonging to the class of "pure parasites" makes it very difficult to produce an immunity against them. They possess no sublethal dose in their living state, and if used when dead, will produce no prophylactic immunity. By artificial atten- uation of these living virulent bacteria Pasteur succeeded in obtaining vaccines of several of them. The methods that he employed were, however, totally impracticable, for not infrequently, by the use of the vaccine, the disease which it was the object to prevent was instigated. It was there- fore a distinct and important triumph when Bail and Weil showed that immunity against these parasites could be attained by using as vaccine antigen, the so-called "aggressins"; i.e., exudates from animals that had been infected with the respective bacteria. Bail's explanation of the aggressin-immunization method is entirely theoretical. He believes that during an infection, the bacteria secrete certain agents which counter- act or entirely destroy the infected organism's protective powers, especially phago- cytosis. These bodies he called aggressins and they were distinguished by the fact that they were formed by living bacteria, and only in the living body. According to Bail, the pathogenicity of bacteria depends upon their power to produce these aggres- sins. If this theory be correct, it should be possible to demonstrate aggressins, espe- cially in infections where the protective power of the organism is almost nil, as for example an infection produced by the bacteria belonging to the group of hemorrhagic septicemia. Unfortunately, in actual practice this is not so. The following experiment gives an idea of the true nature of these aggressins and how they are obtained. At first, an infecting agent — the bacillus of swine pest, may be chosen. This micro-organism belongs to the same class as chicken cholera and fowl plague, and is distantly related to the human pest. For rabbits, this bacillus is a pure parasite, for giunea-pigs, by subcutaneous inoculation, a half parasite. The Obtaining of Aggressins. One drop of a twenty-four-hour brcth culture of this swine pest bacillus, in 5 c.cm. bouillon, is injected intrapleurally in a rabbit in the following manner. 36 OBTAINING NATURAL AGGRESSINS 37 A small incision is made in one of the intercostal spaces on the side of the chest, and through this wound a long canula is introduced into the pleural cavity. The animal as a rule rapidly succumbs to the infec- tion. On autopsy, the pleural cavity is found to contain an exudate of a reddish-brown color (hemorrhagic) on the side where the inoculation was given, and of yellow serous fluid on the other side. The bloody exu- date, about 15 c.cm., is removed with a sterile pipette, placed in a sterile centrifuge tube to which is added 1.5 c.cm. of 5 per cent, carbolic acid drop by drop (making the entire solution a 1-2 per cent, carbolic acid dilution), agitated continually in- order to prevent precipitation, and followed by centrifugalization at a high speed for many hours until it becomes very clear. The upper clear part which is now free of bacteria, or very nearly so, is pipetted off and heated for three hours at 44° C. Its sterility is then tested and if no growth appears after forty-eight hours, it is considered sterile. First Fundamental Aggressin Test. (Its Power of Increasing Severity of Infections.) No. Animal Date Amount cf infective material Aggressins Result i Guinea-pig. 6/IV'os. i/ioo loopful of swine Remains alive. pest subcutaneously. 2 Guinea-pig. 6/IV'os. i/ 100 loopful of swine + 1.5 c.cm. t on third day. pest subcutaneously. of aggressins subcutaneously. 3 Guinea-pig. 6/IV'o5. -j- 1 . 5 c.cm. Remains alive. of aggressins subcutane- ously. 4 Guinea-pig. 6/IV'o5. i/ioo loopful of swine +3 c.cm. sub- f on second day. pest subcutaneously. cutaneously. 5 Guinea-pig. 6/IV'o5. '. 3 c.cm. subcu- Remains alive. taneously. 6 Guinea-pig. 6/IV'os. i/ 1000 loopful subcu- -f-2 c.cm. sub- On 7/IV very ill; taneously. cutaneously. on 8/1 V very ill; 9/IV very ill; zo/IV very thin; i | V m~ : • ; ' n/IV begins to pick up slowly and remains alive. Marked infiltration around point of injection. 38 ACTIVE IMMUNIZATION It can be deduced from this experiment that i/ioo of a loopful of swine pest culture, which represents i/io of a fatal dose for a guinea-pig by sub- cutaneous injection, can be converted into an acutely fatal dose by injecting the aggressin simultaneously or a half hour before the swine pest culture. The aggressin itself is only slightly toxic, and the quantity injected is well borne by the guinea-pig. Its power of increasing the virulence of the infective material varies directly with its quantity, i.e., the greater the dose of aggressin, the more rapidly is death occasioned. If, however, only small doses of the culture are given, and in addition to this, the aggressin is in- jected, the animal does not die, but becomes exceedingly ill, thus indicating the effect of aggressins. In this connection it might be well to add that the aggressin may be given twenty-four hours previous to the time of infection. On microscopical examination of the aggressin exudate, only very few cells, but a great number of bacteria are present. The bacteria here have increased during the short time after the infection to a far greater extent than they would have done in an artificial medium. The body, continu- ally in combat against their increasing toxicity, finds itself powerless when its limited fighting capacity, decreasing in proportion to the rise in strength of the hostile micro-organisms, is expended; and ultimately succumbs to the infection. During the struggle between the protective forces of the organism and the invading bacteria, many of the latter are destroyed and these disintegrated bacteria are found within the exudate. From this fact Wassermann and Citron formed the conclusion that the aggressins are not as Bail claimed, secretory products of live bacteria produced during the conflict between the bacteria and the body organism, but rather the prod- ucts of broken down bacteria. Therefore, Bail's supposition that aggres- sins are only obtained in the living body is erroneous and can be shown to be so by the fact that aggressins may be reproduced whenever the essential requirements can be had, and these are: 1. Large numbers of bacteria. 2. Non-poisonous agents which can disintegrate these bacteria. Aggressins thus obtained are known according to Wassermann and Citron, as "artificial" in contrast to Bail's " natural" ones. Wassermann and Citron Method of Obtaining Artificial Aggressins. Cultures are grown in mass on Kolle's flask'plates. A Kolle's agar plate is equivalent to twelve agar slants. For the inoculation of these flasks a long platinum loop is needed which transfers some of the culture to the plate. The transferred material is then spread over the entire surface of the flask by a large triangular platinum loop. The latter is made by in- serting into a holder both ends of a not too thin platinum wire, about 20 OBTAINING ARTIFICIAL AGGRESSINS 39 cm. in length which is then shaped into a triangular form. While still red hot, this triangular loop should be introduced in the flask and allowed to cool there. Before the culture is spread, it is advisable to bend the entire loop to a slight angle by pressing it against the upper wall of the flask, thereby preventing the hot end of the loop holder from coming in contact with the agar surface. It is best also to test the platinum loop upon the surface of the agar in order to ascertain whether it is still too hot. After twenty-four hours of incubation there is usually a pronounced growth upon the plates. This culture is then washed off either by serum or distilled water (" serous" or " aqueous aggressin"). The former may be obtained fresh from a rabbit. Usually 10 to 12 c.cm. of fluid per flask is required; 3 or 4 c.cm. are first poured upon the culture growth and the mass scraped gently but quickly with the triangular loop. Then the remainder of the fluid 7 to 8 c.cm. is poured in to release the still adherent bacteria. The turbid milky emulsion is collected either in a small dark glass Erlenmeyer flask or a brown bottle. This is then placed into a proper apparatus and shaken for one to two days at room temperature. Enough 5 per cent, carbolic acid is added to make a 1/2 per cent, phenol solution, and the emulsion is centrifugalized and sterilized in the same manner as has been described for the natural aggressins. The tendency of aggressins toward increasing virulence ("infektions bef orderung ") is the same whether these aggressins are artificial or natural. From the following experiment it can be seen that the bacteria contain some substance which is easily soluble in the body fluids and in distilled water, and which increases the infectious nature of the respective bacteria when injected simultaneously with them. In small doses, this substance is not poisonous, in large doses it may be, but is not necessarily so. There is no definite relation between the poisonous qualities of the aggressin and its power to increase the virulence of an infection. This disproves the as- sumption of some authors that the action of the aggressins is dependent upon the toxicity of the endotoxins. ACTIVE IMMUNIZATION Experiment. No. Animal Date Amount of in- fective material from agar cul- ture Serous aggressin Watery K.( aggressin j i Guinea-pig. 2Q/V '05. i/ 200 loop of i Re Results I 2 3 4 5 6 7 8 Guinea-pig. Guinea-pig. Guinea-pig. Guinea-pig. Guinea-pig. Guinea-pig. Guinea-pig. Guinea-pig. 2Q/V '05. 29/V '05. 2Q/V '05. 2/VI '05. 2/VI '0S. 2/VI '05. 2/VI '05. 2/VI '05. I/2OO loop Of swine pest sub- cutaneously. i/ 200 loop ot swine pest sub- cutaneously. i/ 200 loop of swine pest sub- cutaneously. 1/200 loop of swine pest sub- cutaneously. -j-2 5 c cm Remains alive. f after twenty-four hours Remains alive. f after three days. 1 f after three days. Remains alive, t in twenty- ; four hours. Remains alive. 1 . ; subcutane- ously. -{-2 5 c cm. subcutane- ously. 2 c.cm. subcu- taneously. 3 c.cm. subcu- taneously. . . . . 3 c.cm. subcu- 1/200 loop of swine pest. taneously. 4.5 c.cm. sub- cutaneously. Second Fundamental Aggressin Test. (Its Property of Active Immunization.) Bail and his pupils believe that when bacteria invade a normal organ- ism, it is the aggressin power of these bacteria which determines whether or not, by their multiplication, disease will set in. If infection does take place, it continues until the "aggressive" nature of the bacteria is curbed. As there are some bacteria which on injection do not produce any disease, Bail attributes this phenomenon of immunity to the missing " aggressive" action of the respective bacteria. It is not merely the presence of bacteria which is the criterion for the existence of disease; as long as they are void of their " aggressive" property, they have actually become saprophytes. Accordingly, Bail believes that the bactericidal immunity is no true immunity because it can be obtained by injection of dead micro-organisms or by live bacteria in such minute doses that no specific symptoms are pro- duced, i.e., no aggressins are produced within the body. "If the im- munity lacks the "anti-aggressive" component, which alone governs the existence of disease, one gains only an apparent immunity against the ex- citing factor of the disease, but not against the disease itself" IMMUNIZATION WITH NATURAL AGGRESSINS 41 Bail places the utmost stress upon the difference between an immunity directed against the exciting agent of the disease (bactericidal immunity) and that against the disease itself (anti-aggressive immunity). Immunization against the disease is only possible if the aggressin reaches the body of the animal to be immunized. This is possible either by employing Pasteur's method of vaccine inoculation, i.e., the injection of bacteria, the "aggressive" nature of which has been weakened but not destroyed, or by direct inoculations of aggressins. The latter is by far the simpler and more reliable mode of procedure, being productive of a true immunity. s Nowhere does this problem appear of such extreme importance as where immunity against a pure parasite is contemplated, as in Immunity the case of swine pest and chicken cholera. While it is ex- ^air ceedingly difficult, in fact almost impossible to immunize Parasite against these bacteria either with dead or living germs or vaccines, this task is readily accomplished by the injection of non-poisonous aggressins, ina smuch as they are well tolerated. In addition, these bacteria are of help in definitely deciding whether or not an aggressin immunity is at all possible. Weil, a co-worker of Bail's, has carried out these experiments for chicken cholera, while the author has done the same for swine pest. Example of Active Immunization with Natural Aggressins. a. Slow Immunization. Rabbit I. 6./IV. 1905 ist injection: i.o c.cm. natural swine pest aggressin intraperitoneally. ly./IV. 2d injection: i.o c.cm. natural swine pest aggressin intraperitoneally. 25./IV. 3d injection: i.o c.cm. natural swine pest aggressin intraperitoneally. i./V. 4th injection: 2.0 c.cm. natural swine pest aggressin subcutaneously. I2./V. 5th injection: 2.0 c.cm. natural swine pest aggressin subcutaneously. i6./VI. ist infection; with i/ioo loopful of swine pest culture intravenously. i y./ VI. Perfectly weU. 8./VII. 2d infection; with i loopful of swine pest culture intravenously. I5./VII. Perfectly well. 22./IX. 3d infection; with i loopful of swine pest culture intravenously. 3./X. Perfectly well. Rabbit I. Controls. Rabbit II. i6./VI. 1905 1/100,000 loopful of swine pest culture intra- venously. i y./ VI. f found dead. 8./VII. 1905 1/10,000 loopful of swine pest culture subcuta- neously. 9./VII. f found dead. 42 ACTIVE IMMUNIZATION b. Rapid Immunization. Rabbit II. 8./IV. 1905 Injection of 4 c.cm. of natural swine pest aggressin subcutaneously. lo./IV. Animal shows slight ill effects. I3-/IV. Perfectly active. 26./IV. ist infection: i/io loopful of swine pest culture subcutaneously. i6./VL 2d infection: i loopful of swine pest culture intravenously. Animal remained active. Controls. Rabbit III. 26./IV. 1905 1/10,000 loopful of swine pest culture subcutaneously. 27./IV. f. These experiments prove conclusively that by the method described above it is possible to attain a high grade of immunity. In this connection, however, it is very important to adhere to what Bail pointed out, namely, that a long period should elapse between the last inoculation with the aggressin and the first injection; the reason for that being, that during the period of im- munization, and following it for a long time, there is a condition of hyper- susceptibility to infection. Example of Active Immunization with Artificial Aggressins. Rabbit i. 3-/VI. 1905 ist injection: 4 c.cm. of watery extract of swine pest bacilli subcutane- ously. I4./VI. 2d injection: 2 c.cm. of watery extract of swine pest bacilli subcutane- ously. 2 5.7 VI. Removal of some blood. 4./VII. 3d injection: 3 c.cm. of watery extract of swine pest bacilli subcutane- ously. 2I./VII. Infection: i/io loopful of swine pest culture subcutaneously. 3./X. Animal alive and healthy. Rabbit 2. ig./VI. 1905 ist injection: 2.5 c.cm. of serous extract of swine pest bacilli subcutane- ously. 9./VII. 2d injection: 2.0 c.cm. of serous extract of swine pest bacilli subcutane- ously. I2./VII. 3d injection: 4.0 c.cm. of serous extract of swine pest bacilli subcutane- ously. 24./VII. ist infection: i/io loopful of swine pest culture subcutaneously. 22./IX. 2d infection: i loopful of swine pest culture intravenously. Animal remains perfectly well. Rabbit 3. i6./VI. 1905 Injection: 2.5 c.cm. of watery extract of swine pest bacilli subcutane- ously. THE AGGRESSIVE IMMUNITY 43 4./VII. Infection: i/ioo loopful of swine pest culture subcutaneously. I5./VIL Small local infiltrate. Very active. 3-/X. Animal alive and perfectly well. Control animals inoculated on the days of infection died within twenty-four hours after inoculations of 1/100,000 loopful of culture intravenously, 1/10,000 loopful subcutaneously. It is evident from the above, that an immunity against pure parasites can be obtained just as well by one or several injections of extracts of living bacteria, as by injections of natural aggressins. Since the production of aggressins by a struggle between the bacteria and distilled water can be excluded, it can be taken without further explanation that in the development of those substances which have a tendency to increase the virulence of bacteria, or which can be used to produce an immunity, the bacteria play a passive role, in that they are only extracted by the dis- solving agent. The difference between the anti-bacterial and anti-aggres- sin immunity is therefore not a qualitative one, as in both instances they are the substances that are set free from the bacteria which stimulate the formation of antibodies. When living virulent bacteria are injected for the purposes of immunization, they increase so rapidly that a proper dosage is impossible and the animals frequently die before enough antibodies are liberated. In addition, antibodies are also generated against the capsule of the bacteria (bacteriolysins) . The only difference between immunization with morphologically well preserved but dead bacteria and that with aggressins is that within the latter the bacterial substances which tend to bring about the immunity have not been altered by previous heating, but exist in their natural easily absorb- able form. Moreover, by using the extracts one does away with certain toxic substances which are found within the bacterial capsules, and which are rather toxic to subcutaneous tissue, producing necrosis and marasmus. The Third Fundamental Aggressin Experiment. Here, it is demonstrated that the serum of animals immunized, by aggres- sins either artificial or natural, contain antibodies which (i) can neutralize thai-property of aggressins whereby they increase the virulence of bacteria; (2) produce a passive immunity against infection with living bacteria. As for the biological structure of these antibodies, or anti- aggressins as they may be called, it may be said that they belong to the class cf ambo- ceptors, shown by the complement fixation methods. The practical employment of aggressins as a method of immunization offers distinct advantages, namely: 1. Absence of any possible dangerous effects. 2. Absence of or only very slight local and general reactions. 44 ACTIVE IMMUNIZATION 3. The high degree and long duration of the immunity gained by pro- phylactic inoculations. 4. The possibility of immunization against pure parasites. 5. The facility with which the inoculation material is preserved. The disadvantages, however, may be summarized as follows: 1. The manufacture of the inoculation material is rather complex and with some pathogenic bacteria (pest), not without danger. 2. The increased susceptibility during the interval between the inocu- lation and the onset of immunity. The last point applies not only to aggressins, but equally to other methods of active immunization. In times of an epidemic, aggressin immunization should never be undertaken. When one bears in mind the great advantages derived from the employ- ment of this form of immunization, its extensive use should be expected; especially so as animal experimental work with the most important of in- fectious bacteria: typhoid, cholera (Bail), colon (Salus), dysentery (Kiku- chi), staphylococcus (Hoke), has proven it to be quite successful. It is therefore no false prophecy, to say that this method will be em- ployed more and more frequently in the future; particularly for pest, results obtained in animal experimentation by Hueppe and Kikuchi have more than sanctioned its employment in man. Other methods of immunization -based upon the Aggressin principles have been advocated, but none have .attained any practical significance. Mention however must, in passing, be made of the work of Brieger's Brieger and his co-workers Mayer and Bassenge. Brieger Bacterial had made extracts of typhoid and cholera bacilli, in the main Extracts, identical with artificial aggressins. As far as his sterilization was concerned, he obtained that by filtering the extract through the Pukal filter. One should remember that by this procedure many important substances are lost, but in spite of this, his results of inoculation in man have been most encouraging, and there is a possibility that his method may take the place of Wright's or Pfeiffer and Kolle's, as the reac- tions are very much milder. Entirely different from the extracts of living bacteria are those made from previously killed ones. Neisser and Shiga among others have immu- nized against half parasites in this manner. This is not surprising since the dead bacterial bodies can be similarly used for this purpose. As a general rule, wherever dead bacterial bodies cannot be used for immuniza- tion, their extracts will also be found inefficient. The oldest bacterial extracts in use are the tuberculins. CHAPTER V. TUBERCULIN DIAGNOSIS. As a member of the class of bacterial extracts, tuberculin merits especial consideration, because it is used not only for immunization, but also for diagnostic purposes. Tuberculin diagnosis can be employed in several ways. 1. As Koch's subcutaneous method. 2. As the cutaneous reaction (v. Pirquet) and ointment reaction (Moro and Doganoff). 3. As intracutaneous reaction. 4. As ophthalmo reaction (Calmette). Koch's Subcutaneous Method. In the chapter on aggressins it was shewn that when a normal animal was inoculated with a certain definite quantity of bacterial extract, it could readily withstand any effects of such inoculation. If, however, a similar quantity was injected into an animal previously infected with the same bacterium, dangerous symptoms would be in evidence and if the dose were large enough, death would be likely to follow. With these facts for reference, the following experiments will be easily understood. A number of tuberculous guinea-pigs, and a number of normal ones as control, are injected with varying doses of tuberculin. After twenty-four hours some of the tuberculous animals are dead, others very ill, while the normal guinea-pigs remain perfectly active. Just as in the aggressin experiment, we have here a bacterial product in itself possess- ing only slight toxic qualities but which has so increased the virulence of the infection already existing, that an ailment which is usually of a slowly progressive nature becomes transformed into an acute one, terminating in the death of the animal. The close analogy between the experiments with aggressin as the in- jected substance, and that of the tuberculin, will become more clear when the nature of the latter is perfectly understood. Four to six weeks old pure cultures of tubercle bacilli grown Derivation of in 5 per cent, of glycerin bouillon are filtered, and the fil- Tuberculin. trate then evaporated down to i/io of its original volume. The resultant fluid, known as tuberculin, is dark brown and syrupy in nature, and keeps indefinitely. 45 46 TUBERCULIN DIAGNOSIS It consists, therefore, of a 50 per cent, glycerin extract of the soluble products of metabolism of the tubercle bacillus. A part of the glycerin has, however, been used up for the nutrition of the bacteria and thus it is highly probable that after four to six weeks the bouillon contains less than 5 per cent, glycerin and the evaporated solution less than 50 per cent. The specific substances contained within the tuberculin have not been definitely established. As probable elements, however, may be recorded products of secretion of the living bacteria, of degeneration of the dead bacilli and finally the glycerin soluble substances extracted from the bacterial bodies during the heating. No doubt, all these substances and many others about which we lack information, are directly concerned in the activ- ity of the tuberculin. Another of the many unsolved questions which here present themselves may be mentioned: whether any substances exist in the filtrate which are thermolabile, and therefore destroyed or modified by the heating? According to Bail's researches, the aggressin of the tubercle bacillus differs from all other aggressins in that it is not thermolabile and can moreover withstand high grades of temperature. In spite of this, though, attempts to eliminate the heating during the manufacturing of the tuberculin should merit consideration. If merely the term "Tuberculin" is used, one always has in mind the above filtrate tuberculin, also known as Old Tuberculin. The experiment with the tuberculous guinea-pigs has its analogy in the use of tuberculin in the case of man. Here, however, in order to avoid dangerous symptoms far smaller doses of tuberculin are selected. If therefore of two individuals one is tuberculous and the The Tuber- other not, and both are injected with the same amount of old culin Reac- tuberculin o.ooi c.cm., the healthy individual remains per- tioninMan. fectly normal while the tuberculous person shows a typical symptom complex which can be described under, 1. General reaction. 2. Focal reaction. 3. Local reaction. The General Reaction consists of, fever, headache, malaise, nausea, insomnia, cough irritation, palpitation, etc. The most constant symptom is increased temperature; the other manifestations may only be very mild or even entirely absent. The Focal Reaction exhibits evidences of a fresh inflammatory process in the suspicious or old tuberculous foci. In cases of lupus, laryngeal, and iris tuberculosis, this inflammatory reaction can be distinctly seen. In pulmonary tuberculosis the previously vague physical signs may now be- come definite; rales may appear, dulness may be increased, and eventually pains in the chest may arise. The Local Reaction is noticed at the point of inoculation. In spite of the sterile needle and thorough disinfection, the skin around the site of the injection becomes red, swollen and painful. That this is not due to dirt infection is proven by its absence in non-tuberculous individuals. HAMBURGER'S LOCAL REACTION 47 [The local reaction has been recently advocated by Hamburger as a very delicate diagnostic method. He carries out the test as follows: i/io c.cm. of a i: 10,000 dilution of tuberculin is injected just beneath the skin of the forearm or back. The needle should be of fine caliber and the syringe should never have been used for more concentrated solutions of tuberculin. If the reaction is positive a subcutaneous infiltration appears within twenty-four hours. Furthermore, there is a reddening at the site where the point of the needle rested (" Depot reaction" of Hamburger). If there is no reaction within twenty-four hours, i/io c.cm. of a dilu- tion i : 100 should be injected. This subcutaneous local reaction may also be carried out in the form of an intracutaneous test (see later)]. Of the three types of reaction the general and focal symptoms are the most constant. Both are so characteristic for the existence of tuberculosis, that their appearance justifies the diagnosis. In practice, however, it is the general reaction, or almost exclusively the manifestation of fever, which is taken as the guiding symptom in Koch's subcutaneous method. The focal reaction in all non- visible tubercular lesions is determined by subjective methods, while increase in temperature is alone an objective finding. In carrying out the subcutaneous tuberculin test, one must remember several practical points which are of help for the correct interpretation of the results. These may be summed up thus: Inasmuch as the rise of temperature is of diagnostic importance, no patient with any fever should be subjected to the inoculation. For several days previous, the patient's temperature should be taken every three hours and only if the temperature does not exceed 37° C. per axilla should the tuber- culin diagnosis be undertaken. The quantity of tuberculin to be injected is also of the utmost consequence. Too high doses should be avoided, as the specificity of this osage o reaction, like all other biological reactions, is limited quanti- tatively. While small doses of tuberculin will give a rise of temperature only in tuberculous individuals, larger doses may give the same rise even in healthy people. In addition, too large doses as a rule produce a general reaction which might be very severe and injurious. The dosage advised by Robert Koch for the diagnostic tuberculin reaction is as follows: 1. o.oooi c.cm. T. (for very weak individuals and children). 2. o.ooi c.cm. T. 3. 0.005 c.cm. T. 4. o.oi c.cm. T. 5. o.oi c.cm. T. 48 TUBERCULIN DIAGNOSIS CHART i. — Example of a diag- nostic tuberculin reaction. The dose chosen at the first injection is o.ooi c.cm. T. Very weak individuals, i.e., those in an advanced stage of tuberculosis or those who have experienced a recent hemopty- sis, as well as children, should receive an initial dose of only o.oooi c.cm. T. Bandelier and Ropke, who have a wide experience in this field, advise 0.0002 c.cm. T. as the primary dose. Few patients show a distinctly positive fever reaction even with this small dose; by a positive reaction is meant an increase in the temperature so that the latter is at least 0.5° C. higher than the highest point before the injection. If the temperature has not increased, the re- action is negative, and after an interval of two to three days of normal temperature the second inoculation of 0.005 c.cm. T. is given. If as happens occasionally after the first inoculation there is a doubtful reaction, i.e., there is an in- crease of 0.2° to 0.3° C. then the dosage at the second injection should not be increased to 0.005 c.cm., but the same amount o.ooi c.cm. T. is to be repeated. In a tuberculous individual this repeated injection of o.ooi c.cm. frequently results in a distinctly positive reac- tion, while in a non-tuberculous patient instead of the former doubtful, a distinct negative re- action is obtained. The general rules given for the first inocula- tion also apply to the second with 0.005 c.cm. In a doubtful reaction with this dose, one does not directly proceed to the o.oi c.cm. dosage, but the 0.005 c.cm. dose is repeated and only after a negative reaction with the repeated 0.005 c.cm. dose is the o.oi c.cm. in- jected (see accompanying Chart i). This rep- resents the maximum amount of tuberculin to be used for diagnostic purposes. Koch advises repetition of this dose if no reaction is ob- tained. The majority of authorities, however, abstain therefrom. In fact some investigators claim that a reaction obtained after inoculation of o.oi c.cm. cannot be considered specific, be KOCH'S SUBCUTANEOUS TUBERCULIN REACTION 49 cause there are non-tubercular individuals who respond to this quantity of tuberculin. Most tuberculous persons react after a dose of o.ooi c.cm. to 0.005 c.cm. T.; those, however, who are very far advanced or who suffer from severe cachexia, remain unresponsive to even much greater doses; in addition, patients whose serum contains antituberculin, do not react be- cause the inoculated tuberculin is quickly neutralized. According to Loewenstein the tuberculin reaction does not depend so Loewenstein's much upon the quantity of the tuberculin, as upon the frequency that Dosage it is injected. He, therefore, advises that the same amount, about Scheme. 0.0002 c.cm. be inoculated four times during the course of twelve to sixteen days. In by far the greater majority of tuberculous patients a typical reaction appears after the third or fourth injection. The author has no per- sonal experience with this method, but the reports of other authorities do not exhibit as favorable results as those claimed by Loewenstein. The inoculation is always to be given subcutaneously, and the Technique of back Qr breast js the best site for it. The dilution is made j . . immediately before the injection, with physiological salt solu- tion or 0.5 per cent, carbolic solution. In interpreting the result of the reaction one must exclude Value of rises of temperature due to extraneous influences such as Reaction. Angina, Influenza, etc. Furthermore there are individuals, especially hysterical ones, in whom any injection as such is apt to produce a rise of temperature. To guard against such a possibility an injection cf physiological salt solution should be made and thus quiet any suspicion of error. The diagnostic use of tuberculin is indicated when one is Indications, dealing with adults who present clinical symptoms, or clinically suspicious symptoms of tuberculosis, but who run no tem- perature and tubercle bacilli cannot be found. Tuberculin is contra-indicated in patients with high fever, Contra- and during or shortly after hemoptysis or hematuria. In 'indications, epilepsy, marked cardiac or renal affection, arteriosclerosis, diabetes, and similar conditions, inoculation should be under- taken' only under the strictest indications and with great care. A positive general reaction means that the individual is infected with tuberculosis, but does not throw any light upon the site, the extent, or the prognosis of the infection. The focal reaction allows the diagnosis of the position of the lesion. The Cutaneous Reaction. The cutaneous reaction was first introduced by v. Pirquet, who noticed that by scarification of the skin and application of tuberculin, tuberculous 50 TUBERCULIN DIAGNOSIS children would develop a distinct papule at this point, while in non-tuber- culous conditions such a reaction would be absent. The Technique of the Cutaneous Reaction. "The patient's forearm on the inner side is cleansed with ether; two drops of the pure undiluted old tuberculin are placed upon the skin about 10 cm. apart, and then the skin is scarified first between the two drops, for the purposes of a control, and next within each of these drops. — [A boring scarifier, devised for this, works very easily.]1 Finally a piece of cotton is placed upon each of these drops and allowed to remain there for ten minutes after which the cotton is removed. A dressing is not necessary." Interpretations of the Reaction. Scarification of itself produces the so-called "traumatic reaction" i.e., a small wheel with a rose-colored margin appears around each of the three points of scarification. This reaction passes away after several hours and only a small scab remains surrounded by a red rim. This "traumatic reaction" is to be sharply differentiated from the "specific reaction." The latter is noticed only upon the upper and lower points where the tuberculin has been applied and consists of a red, indu- rated papule which rapidly extends in size and elevation, measuring 10 to 30 mm. in diameter. (Fig. i, Plate I.) The papule may be round or have irregular margins. Scrofulous children show small, irregularly raised follicular infiltrations around the specific reaction. This is known as the "scrofulous reaction" It may appear as early as within three hours, but usually occurs within twenty-four hours. It arrives at its maximum within forty-eight hours; occasionally it is delayed and may not develop fully until the third or fourth day and then it begins to fade. Fre- quently a small pigmented spot remains. General and focal reactions are practically absent. Moro-Doganoffs Ointment Reaction. Moro and DoganofT found that a 50 per cent, ointment of tuberculin in lanolin rubbed into the skin without scarification, would give a reaction which consisted of small nodular or papular efflorescences after the nature of Lichen Scrophulosorum. In accordance with the number and size of these nodules as well as the time of their appearance, three grades of reaction are described. In employing the ointment it should be heated to 25° C. and a quantity about the size of a pea is thoroughly rubbed into the skin of the 1The flat end (2 mm. wide) of sterile wooden toothpicks can serve the same purpose. THE INTRACUTANEOUS REACTION abdomen or the region of the mammilla, for almost a minute. The diag- nostic value of the reaction is variously interpreted. An almost analogous reaction, described independently of Moro, by Lignieres and»Berger is to be found in thoroughly rubbing in concentrated old tuberculin into the shaved skin of tuberculous cattle. The Intracutaneous Reaction. Tuberculin even in very weak dilutions when injected directly into the skin of tuberculous indi- viduals produces marked inflam- matory infiltrates. This was first observed by Mendel and Man- toux and has been carefully studied in cattle and guinea-pigs by Roemer. In guinea-pigs the test per- formed is as follows: the hair of the abdomen is removed by calcium hydrosulfid, i/io c.cm. of a 20 per cent, tuberculin solution (0.02 c.cm. tuberculin) is injected with a very fine needle directly into the skin. The result is noted after 48 hours in order to allow the trau- matic effects to wear off. Roemer and Joseph differentiate three grades of reaction: (a) Very susceptible animals (+ + + ) show after 18 to 24 hours a pinkish wheal about the size of a half dollar with a very deep red center. In 48 hours the dark center which represents a blood extravasation attains a greenish hue. After four days a superficial necrosis sets in which leads to sloughing and final scarification. (b) In less susceptible animals (++) there is no central dark area of the wheal and the latter appears after 48 hours. These animals show a very slight necrosis later on. (c) The mildest form can be differentiated from the traumatic re- action only in that it does not disappear after 48 hours but remains several days. In the human being the intracutaneous method has been advised by Hamburger. The results are as yet insufficient to form definite conclusions as to the clinical value of the test. FIG. 13. — Inoculation with tuberculin for the Pirquet reaction. 52 TUBERCULIN DIAGNOSIS The Ophthalmo Reaction. At the discussion which followed v. Pirquet's presentation of his cuta- Historical. neous reaction, Wolff-Eisner remarked, "that by instilling some 10 per cent, tuberculin into the conjunctival sac, a local conjunctivitis was obtained and occasionally, also a general reaction. The marked severity of the reac- tion, however, and its apparent lack of specificity, made its diagnostic value improb- able." Calmette, who believed that Wolff-Eisner's failure in obtaining accurate results lay in the fact that glycerin was contained in the old tuberculin employed by him, obtained by alcohol precipitation a glycerin-free dry product, which he used in a i per cent, solution equivalent to 10 per cent, old tuberculin. It was he, therefore, who first established the clinical diagnostic value of the reaction. But his hypothesis was erroneous, as the mild reactions which he obtained were not due to the absence of glycerin, but because the Lille tuberculin is much weaker than the German prepara- tion. The author was able to show that the old tuberculin could very well be used for the Ophthalmo reaction if, instead of the 10 per cent., a i per cent, dilution was made. Thus employed, the reaction is exceedingly mild and specific. Eppenstein later advised a 2 and 4 per cent, dilution in cases where the i per cent, solution gave no reaction. Technique of Reaction. It is of extreme importance to have freshly prepared sterile dilutions of the old tuberculin (Hochst Farbwerke). All the ready-for-use preparations on the market should be discarded. This applies also to the "Tuberculin Test" Calmette' s sold by Poulenc Freres. The mishaps and low grade of specifi- city often ascribed in literature to the ophthalmo reaction can in a great major- ity of cases be explained by the employ- ment of preparations other than the i to 2 per cent, fresh dilutions of the old tuberculin advocated by the author, and in still another number of cases to its em- ployment in conditions where it was dis- tinctly contraindicated. The prepara- FIG. 14.— Ophthalmodiagnosticum for J F ^ tuberculosis. (After Citron.) tion of fresh tuberculin dilutions is very much simplified by the "Ophthalmodiag- nosticum for Tuberculosis ," of the firm P. Altmann, Berlin N. W. 6 (Fig. 14). This outfit consists of twelve sealed glass tubes each containing o.i c.cm. old tuberculin, a cylinder for the dilution graduated in percentages, and a pipette measuring o.i c.cm. fitted with a rubber bulb. One of the sealed ampoules is shaken so that the tuberculin is collected into its broader part and then broken at the designated point near the narrow end. The tuberculin is drawn up to the mark into the pipette and then trans- OPHTHALMO REACTION 53 f erred into the cylinder. Boiled water or sterile saline is added to the i, 2 or 4 per cent, dilution mark. The pipette is washed clean in the solution by successive aspiration and expulsion in order to free it completely of the remaining concentrated tuberculin, and can now be employed as the eye dropper. The solution should be used only on the day it is prepared. The tuber- culin in the sealed tube can be kept indefinitely. The pipette and graduate are sterilized by dry heat, boiling or by thorough washing in boiling water. One drop of the tuberculin dilution is deposited in the inner angle of the eye, and care should be taken that the drop is not immediately expelled, but evenly distributed in the conjunctival sac. In tuberculous individuals the reaction appears in twelve to Gradation twenty-four hours, and according to its intensity can be of the Oph- ,. . , , . A ^, , thalmoRe- dlvlded mto three Srades' action. First Grade. — Reddening of the caruncle and inner side of the lower lid (+) (see Fig. 2, Plate I). Second Grade. — Same as above but additional involvement of the con- junctiva of the eyeball (++)• Third Grade. — Conjunctivitis purulenta, phlyctenulae and other such severe manifestations (+ + +). The reactions of the first and second degree occur most frequently. The manifestations associated with the former of these are so mild that the patient himself does not usually notice them. If the proper dilution is used and the contraindications of this test are observed, a reaction of the third degree is obtained only in exceptional cases. Fever never occurs. The other eye serves as a control. It is advisable therefore before under- taking the reaction, to note carefully any differences that may exist in the conjunctive on both sides. It must be remembered that Selection i- The greater the dilution, the more specific is the reaction, of Correct 2. The test should not be repeated upon the same eye, even Dilution, jf there was no reaction at all at the first instillation. The following procedure should be adopted. A drop of the 2 per cent, tuberculin dilution is placed in the left eye. If a positive reaction takes place, it is of great probability that the patient is suffering from an active tuberculous process and thus the diagnosis is established. If, however, that proves doubtful, and further corroboration is required, the patient should receive, after the first reaction has entirely subsided, one drop cf a i per cent, tuberculin dilution in the right eye. If a negative reaction is obtained at the instillation of the 2 per cent, dilution, one drop of the 4 per cent, dilution is placed in the right eye. A negative reaction with the 4 per cent, mixture speaks almost conclusively for the absence of tuberculosis except in far advanced cachectic condi- tions. A positive result does not, on the other hand, indicate the presence 54 TUBERCULIN DIAGNOSIS of tuberculosis, as there are many normal individuals who react to a 4 per cent, tuberculin concentration. Instead of the tuberculin solution, Wolff-Eisner recommends a 2 per cent, old tuberculin lanolin ointment. The lower lid of the eye is pulled downward and a pea- sized mass of the ointment is gently placed into the conjunctival sac by means cf a sterile glass rod. The lid is held fixed for about a minute. The ophthalmo reaction is indicated in all suspicious cases of Indications tuberculosis where the presence of bacilli cannot be demon- , , , -r, strated and where the subcutaneous reaction either on account tnalmo Re- actions. °f the presence of temperature or other reasons cannot be undertaken. This test is much milder and more agreeable to the patient than the sub- cutaneous one, and in ambulatory work more significant, inasmuch as it does away with any necessity for considering as a guide the temperature taken by the untrained and usually unreliable patient. The ophthalmo reaction is contraindicated in all diseases of Contraindi- ^e eye^ tuberculous or otherwise. If one eye only is affected, ca1tl°J?s °r the reaction should not be undertaken upon the healthy eye. the Oph- r,. £ J , thalmo Re- Similarly, patients who have had some eye disease, even though action, many years ago, those who by reason of their occupation are readily exposed to eye diseases, or who live in districts where trachoma is prevalent, should be excluded from the test. The reason being that in those individuals the conjunctival mucous membrane becomes a locus minoris resistentiae and therefore easily inflamed. Repeated instillations of tuberculin into the same eye may set up very severe disturbances. Scrofulous children often show reactions of the third degree, inasmuch as they possess the constitutional tendency which makes them easily susceptible to conjunctivitis or phlyctenulse. In patients with a positive ophthalmo reaction that has subsided, a recurrence of the con- junctival inflammation is frequently observed when they begin to receive subcutaneous inoculations of tuberculin for therapeutic or diagnostic purposes. The Specificity of the Tuberculin Reaction. I'he one real essential for the practical application of all biological reactions is the specificity of the reaction. There is, however, as will be repeatedly pointed out further on, no single absolutely specific reaction. In fact, it would be more exact to consider these biological reactions only relatively specific; the latter depending upon the quantity of the required antigen and the reacting organism. In this connection it may also be said, that it is never possible to draw an exact line between the specific and non-specific biological reactions. There always will be a doubtful zone. As a general rule, however, the smaller the quantity of antigen that is required and the stronger the resulting reaction, the more probable is the biological specificity. SPECIFICITY OF V. PIRQUET S REACTION 55 In tuberculosis this problem is rendered still more complex by the pathological anatomical findings, whereby it is shown that an extraordinary high percentage of individuals have undergone tubercular infection at some time during life. The clinical consideration of tuberculosis, however, does not deal with the diagnosis of these harmless, practically healed tuberculous foci; what the clinician desires to know is whether or not a group of symp- toms manifested by a patient is of a tuberculous nature or not. In other words, it is not the latent, inactive, but the active form of tuberculosis that is to be diagnosed. It is that one must view the merit of the various tuberculin tests from this standpoint. The reaction of least specificity in adults is the v. Pirquet' s cutaneous reaction. In children it is far more specific. V. Pirquet has made the following very interesting observation.. Out of 747 children in Escherich's clinic in Vienna upon whom the reac- tion was tried, there were: Clinically tuberculous 130, out of which 113 (87 %) showed a positive reaction; Clinically non-tuberculous 512, out of which 104 (20 %) showed a positive reaction; Doubtful 115, out of which 56 (48.6%) showed a positive reaction. Almost all of the tuberculous children who did not react were cachectic. As for the positive reaction in non- tuberculous cases, the age of the child in large part explains the great differences found. Whereas healthy infants up to the sixth month almost never give a positive reaction, healthy children of 1 to 2 years react in 2 per cent, of cases. 2 to 4 years react in 13 per cent, of cases. 4 to 6 years react in 1 7 per cent, of cases. 6 to 10 years react in 35 per cent, of cases. 10 to 14 years react in 55 per cent, of cases. In adults one meets with a positive v. Pirquet's reaction in more than 70 per cent, of all cases. V. Pirquet explains this by the presence of latent tuberculosis. // therefore becomes self-evident, that the cutaneous reaction in adults is void of any diagnostic value. A negative reaction only, can be fully relied on, and that, if no cachexia exists. Ellerman and Erlandsen have attempted to improve upon the diagnos- tic value of the cutaneous reaction by a quantitative titration of the tu- berculin hypersusceptibility. They aimed to get the weakest dilution which still gave a distinct cutaneous reaction. Their results showed wide variations. Of those who reacted to a 1-5 per cent, tuberculin dilution the great majority were clinically tuberculous; at the same time there were many in whom not the faintest clinical suspicion of tuberculosis could be entertained. 56 TUBERCULIN DIAGNOSIS In young children on the other hand, v. Pirquet's method should be the one of choice. In addition to its being entirely harmless, and easily applied, it possesses a high diagnostic value. As for Koch's subcutaneous reaction, it is specific, inasmuch as it is a rare exception to get a negative reaction in an active tuberculous process. This occurs only in cases either with very severe cachexia or those with freely circulating antituberculin in* the blood. If the latter two possi- bilities are excluded, the absence of a positive reaction speaks decidedly in favor of the absence of tuberculosis. The interpretation of a positive reaction as to the existence of clinically active tuberculosis cannot be so definitely answered. From the recent work of most authorities, however, it seems to be taken for granted that a positive reaction does mean an active tuberculosis; still, this statement requires a great deal of consideration and limitation. In this connection the statistics of Franz are of interest. Out of 400 apparently healthy soldiers in one of the Austrian regiments who in 1901 — their first year of service, received an inoculation of 0.003 c.cm. of tuber- culin, a positive result was found in 61 per cent, of the cases. In the fol- lowing year (1902) 100 of the soldiers were re-inoculated and all of those who reacted positively the first time, did so a second time, in some instances even though the second dosage was smaller. Moreover, fourteen others who responded negatively the previous year showed positive results this time, making a total of 76 per cent. Out of 323 men inoculated for the first time in 1902, 68 per cent, reacted positively. It must be mentioned, however, that the majority of the members of this regiment came from a very tuberculous1 district. The same author also examined a Hungarian regiment in a tuberculous-free district, and under similar circumstances found a positive reaction in 38 per cent, of cases. Although these figures may be exceptionally high, they are without doubt conclusive as to the fact that Koch's reaction cannot be considered specific for "active" tuberculosis. Franz in addition gives important statistics concerning the health of the inoculated soldiers whom he examined for years following the inoculation. The appended charts taken from the most recent publica- tion of Franz (Wien. Klin. Woch., 1909, No. 28) tabulate what has been said above. SPECIFICITY OF OPHTHALMO REACTION 57 a During the period of three years, those o Positive that terminated their service through 0> No. of reaction death, invalidity or long leave of Regiment .£ soldiers absence showed. *o inocu- § > lated Negative reaction Tuberculosis Disease sus- picious of tuberculosis Other diseases Bosn. Inf. Reg. No. i 1901 400 ; +245(61%) \ -155(39%) 17 ( 8 deaths) 22 5 ( 4 deaths) 25 10 7d) Bosn. Inf. Reg. No. i 1902 323 / +222(68.7%) \ -101(31.3%) 13 ( 6 deaths' 4 28(1) 13 7(2) 5 Inf. Reg. No. 60.. .. 1902 279 f+io8(38.7%) I -171(61.3%) 4 3 ( 2 deaths 4 > 5 8 12 Total IOO2 J+575 I -427 34 (14 deaths 12 ( 6 deaths \ 54(i) ) 43 25(2) 24 d) Those who reacted in Regiment Time of observation No. ill with manifest tuberculosis i 1901 and 1902 to 0.003 c.cm. tuberculin Positive Negative Bosn. Inf. Reg. No. i. I Ser. (400 From 10. x. '04 10 6 4 men). until end of 1908. Same. II Ser. (323 men); Inf. From Oct., 1908, 6 5 i Reg. No. 60 (279 men). until end of 1908. 2 i i 18 12 6 • - :' ': '•--' T In regard to the specificity of the ophthalmo reaction, the condi- Specificity ^QU?> are more favorable than in both of the preceding tuber- ° Reaction Q°cu^n tests. The following short chart is explanatory. Posi- tive reactions were obtained in Audeoud Petit Citron (ist series) Eppenstein Schenk and Seiffert Tuberculosis Suspicious cases 94-6% 81.0% 94-3% 61.6% 80.7% 80.0% 72.3% 40.0% 78.6% 30.0% Normal cases . . 8.<% 18.4% 2.2% 0.0% 5.8% Calmette's preparation. i% old tuberculin. 58 TUBERCULIN DIAGNOSIS It is evident from the above figures that by the use of the i per cent, tuber- culin a grade of specificity is reached which can be considered quite high, as the non-tuberculous react only in a very small percentage of cases, while exist- ing tuberculosis is detected in 80 per cent, of the subjects. Clinical examina- tions of the positive reacting patients show that the latter belong to the group of active tuberculosis. Absolute reliance, however, in the determi- nation as to whether the positive reaction given is due to an active or latent tuberculosis, cannot even be placed on the ophthalmo reaction. According to several authors, it is claimed that typhoid fever, rheumatism, and syphilis (in the stage of eruption) are very prone to give a positive ophthalmo reaction, without the presence of a simultaneously existing tuberculosis. In conclusion, therefore, the author finds it difficult to make any general statement as to the preference of one or the other test for diagnostic purposes. In children the application of the Pirquet reaction, and in adults the ophthalmo reaction, are given preference to Koch's subcutaneous reaction; provided, no contraindications exist against the first, and that treat- ment with tuberculin is not to be undertaken. In the latter instance, the recurrent ophthalmo reaction when the tuberculin therapy is instituted, authorizes the use of Koch's subcutaneous diagnostic method. Mallein, Trichophytin. Similar to old tuberculin, the Mallein (Helmann and Kelning) has been obtained from cultures of Glanders bacilli and the Trichophytin (Plato) has been isolated from the Trichophyton fungi. Mallein has already attained its practical application for the diagnosis of glanders in veterinary medicine. Like tuberculin it is harmless in normal organisms, but brings about temperature and a local reaction at the site of the injection when inoculated into glanders stricken animals. Various general symp- toms may also appear. Its employment in a manner analogous to the ophthalmo reaction is also possible. The Prognostic Value of the Local Tuberculin Reactions. The fact that the tuberculin reaction in cachectic tuberculous indi- viduals is usually negative has led different observers to regard the degree of the local reaction in a given case as a guide to the prognosis. If, as is considered, the strength of a reaction depends upon the dose of tuberculin, the degree of susceptibility of the organism and its reactive power, the following theoretical possibilities may be considered. (a) With mild clinical manifestations, a reaction obtained only after large doses of tuberculin would speak for a favorable prognosis because the hypersusceptibility of the individual toward tuberculin is still mild. PROGNOSTIC VALUE OF OPHTHALMO REACTION 59 A severe reaction after a very small dose is less favorable, denoting a very much increased hypersusceptibility. (b) With very marked clinical manifestations } a reaction obtained only after large doses of tuberculin would speak for an unfavorable prognosis because the reactive power of the organism is weak. A severe reaction after a very small dose is more favorable, denoting a stronger reactive power of the individual. Wolf-Eisner and Teichmann formulated the following three hypotheses in regards to the conjunctival test. 1. A positive conjunctival reaction is indicative of an active tubercu- losis; associated with manifest clinical symptoms a strongly positive test would speak for a more favorable prognosis than if the reaction were weak or negative. The same interpretation can be applied to the cutaneous reaction. 2. A negative reaction (conjunctival or cutaneous) with manifest tuberculosis would point toward an unfavorable prognosis. The outlook is also unfavorable if the ophthalmo reaction is positive but the cutaneous test negative. 3. With the cutaneous test, the local reaction sometimes continues longer than four days. This is known as a " prolonged or continued reac- tion" and is found among normal subjects, among individuals in whom the lesions are undergoing a healing process and those in whom the tuber- culosis has existed for over ten years. The prognosis therefore is favorable. CHAPTER VI. THE TUBERCULIN THERAPY. Right at the beginning it must be made clear, that tuberculin is not to be considered as a curative agent against tuberculosis, but rather in the light of a bacterial extract for active immunization. In the previous chapter it has been shown that while there are some infectious diseases where im- munization can be accomplished by the use of bacterial extracts and dead bacteria, there are others where immunization is possible only when living vaccines or aggressins of living bacteria are employed. In both of these instances, however, healthy individuals are being treated to be protected from future infection. An exception is presented by rabies. In this dis- ease, the vaccination against the active symptoms is instituted after the infection has already taken place, but the redeeming feature about its treatment is the existence of the very long incubation period. Thera- peutic use of tuberculin, however, is a form of active immunization which belongs to neither of the above classes. The principle involved here is entirely different, and the question arises if it is at all possible to obtain an active immunity by the injection of an antigen in a condition where infec- tion has already taken place, and produced pathological changes. [In other words, where spontaneous immunization has failed.] An answer to this question is to be found in Koch's fundamental ex- periments which have been the basis as well as starting point of the entire tuberculin study. If a normal guinea-pig is inoculated with tubercle bacilli, the point of inoculation very soon closes. After ten to fourteen days there appears at this site a small hard nodule which finally ulcerates. This shows no tendency to heal and remains so until the death of the animal. If, however, an already tuberculous guinea-pig is similarly inoculated, while the point of inoculation also closes, no indurated nodule appears. Instead, a necrotic process of the skin sets in after the second day, which finally terminates in the casting off of the slough and the formation of a flat ulceration that heals rapidly. It does not matter whether living or dead tubercle bacilli are used for the second infection. In explanation of the above phenomenon it must be said that although the first injection had a fatal effect upon the animal, it must have stimu- lated certain immune reactions within the organism which became manifest after the second inoculation. That a condition similar to this, or even more favorable exists in man, is proven by the fact that while the large 60 IMMUNITY AGAINST TUBERCULIN PREPARATIONS 6 1 majority of people become infected with tuberculosis at some time during their life, only a small number show symptoms referable to the disease and the rest undergo spontaneous cure. Koch further showed that the injection of tuberculous guinea-pigs with large doses of tubercle bacilli produced rapid death, while frequently re- peated small doses evinced favorable effects upon the site of injection and the general condition of the animals. In this way he proved the beneficial influence which successive inoculations exert upon the primary infection. In the employment, however, of dead tubercle bacilli in man for thera- peutic purpose a serious difficulty presented itself. It was found that the inoculated dead bacilli were not absorbed, but remained for a long time at the seat of the inoculation instigating suppurative processes. On intra- venous application, formation of tubercular nodules was noticed. Koch realized that these harmful effects were due to the non-absorbable parts of the tubercle bacilli; in the main the bacterial capsules. He there- fore attempted to extract the immunizing substances, and in this way brought about the old tuberculin. It may be questioned, whether this old tuberculin is identical with tuberculous antigen; whether it is a feasible preparation for purposes of immunity; whether it contains all the important elements of the tubercle bacillus; if not, which are lacking? The specificity of immunity reactions has already been dealt with sufficiently to make it clear that immunizing a healthy individual with old tuberculin will bring about an immunity only against the substances contained within this prepa- ration. That that does not meet the requirement is proven by the fact that an animal immunized against tuberculin will not be protected against a later infection with living tubercle bacilli. It cannot therefore be expected that immunization of a tuberculous individual with old tuberculin will protect him against living tubercle bacilli. The expectation, however, that his immunity will be raised against old tuberculin only, is fully justified. Furthermore, we have seen in the aggressin experiments, that inoculation of animals with the aggressin antigen was sufficient to increase the immunity so that a subsequent infection was not attended by any harmful effects. In this case the injected living bacteria are not destroyed, but their ill effects upon the immunized organism have been paralyzed. In other words, the parasites have been transformed to saprophytes, That a similar state of affairs exists in the use of antitoxic sera will readily be seen. The antitoxic diphtheria serum, for example, neutralizes the toxin and thus cures the disease. The bacteria themselves, however, remain intact and also infectious for untreated individuals. Only later on are they absorbed by the phagocytes. When therefore in an individual who has passed through a course of tuberculin treatment there are found fully virulent tubercle bacilli in the sputum, it is no proof, if that is to be the only corroborative evidence, that the tuberculin treatment had been ineffi- cient. In fact, there are strong possibilities that the tubercle bacilli have become transformed into saprophytic bacteria. It is, however, a noteworthy and impor- tant fact, that immunization with tuberculin proves no protection against later infection with living tubercle bacilli, while in the case of aggressins and toxins this is possible. 62 TUBERCULIN THERAPY Although tuberculin cannot be considered as the aggressin or toxin of the tubercle bacilli, it simulates these substances with sufficient closeness to warrant its use in tuberculosis. It brings about an immunity against some of the poisonous products of the tubercle bacillus, leaving the others to be taken care of by the natural fighting powers of the individual. The knowledge that this old tuberculin represents only a partial aggres- Various sin, or toxin, and by that is meant that it does not contain all the neces- Tuberculin sary elements for the establishment of a true immunity, has led to the Preparations, production of a large group of preparations which are supposed to supply the missing properties of the old tuberculin. The more important of these preparations were originated by Robert Koch. Those which are of frequent use are: a. Old tuberculin (T. Tuberculin) — preparation described on page 45. b. Original old tuberculin (T. O. A. Tuberculin Original Alt.) The latter consists of the original nitrate of the tubercle bouillon culture and varies from the old tuberculin in that it is not heated and reduced to i/io its volume. The omission of heating is certainly not without effect, inasmuch as high heat modifies in some way the soluble bacterial substances. This preparation has not been used therapeutically by Koch himself. Spengler and especially Denys, who have made wide use of it under the name of "Le bouillon nitre," have been its main supporters. c. Vacuum tuberculin (V. T.) is the original tuberculin which has been reduced in vacuum to i/io its volume. d. The aqueous tuberculin of Maragliano (Tuberculina Aquosa) is closely allied to the above tuberculins. It contains all the water soluble extracts of the living tubercle bacilli obtained by extraction of the living bacteria in distilled water, followed by nitration. As is evident, it is prepared on the same principle as Brieger's bacterial extracts and Wassermann-Citron's artificial aggressins. The above mentioned tuberculin preparations are all very much alike in that they contain the soluble bacterial elements. Their action there- fore corresponds more or less to that of old tuberculin. Another set of preparations have as their basis the insoluble New elements of the bacteria and cannot as such, in either living or Tuberculin dead form, be absorbed. Since, however, the absorption Preparations. Of bacteria is a prerequisite to their proper action, it was necessary to so alter the body substances of these bacteria that they could be taken up. Koch found that this was best accomplished by thoroughly pulverizing the bacilli in large mortars. And by this means the first preparation which he obtained was e. New tuberculin T. R. (Koch) (Tuberculin Rlickstand or Residual Tuberculin). Cultures of young tubercle bacilli are thoroughly dried in vacuum and finely ground in mortars. The pulverized bacilli are agitated in distilled water and the VARIOUS TUBERCULIN PREPARATIONS 63 turbid fluid is centrifugalized. The sediment thus obtained composes the T. R. or the tubercle bacilli residue. T. R. therefore contains the aqueous insoluble components of the tubercle bacillus, while the soluble ones are retained in the opalescent supernatant fluid which Koch calls TO (Tuberculin Original). T. R. is readily assimilated by patients. If carefully administered it produces very little infiltration and only slight temperature and general reaction. Its price is com- paratively high (i c.cm. costs 8.50 marks). The first preparation which contained both the soluble and insoluble elements of the living bacilli was the /. New Tuberculin — Bacilli emulsion (B. E.) which consists of T. R.+ T. O. The living tubercle bacilli are first pulverized in a mortar and then suspended in salt solution. Centrifugalization is not necessary, but sedimentation is required; 50 per cent, glycerin is added for preservation purposes. Next to T. the new tuberculin B. E. has been most carefully studied. B. E. also lacks in being an ideal antigen inasmuch as immunity attained by its injections is not at all proof against subsequent infection. Closely resembling the B. E. is g. The Tuberculin of Beraneck. Beraneck produced two tuberculin preparations of which one is in the main identical with TOA, while the other is an extract of tubercle bacilli with i per cent, of phosphoric acid. These tuberculins are mixed and applied together. Sahli reports good results with this mixture. Although none of the described tuberculin preparations can Action of be considered a true antigen of the tubercle bacillus, they Tuberculin, undoubtedly have a favorable effect upon tuberculous individuals. To a certain extent the benefits are derived by the mechanism of partial immunization. This in itself does not, however, explain the entire phenomenon of their successful action. On examination of the tuberculous organs of animals treated with tuberculin, there will be found within the healthy tissue surrounding the tuberculous foci, a fresh inflammatory reaction. This consists of a sero- fibrinous exudate and a zone of leucocytes intruding to a certain extent upon the tubercular lesion. Tuberculin acts only upon tuberculous tissue which is still alive and not upon dead, cheesy or necrotic structures. If enough tuberculin is given so that death of a tuberculous guinea-pig occurs, the changes found are striking. On dissection, about the point of inoculation Koch reports a marked congestion of the blood vessels giving a red and often an almost dark violet appearance. This discoloration extends for a greater or less distance from the 64 TUBERCULIN THERAPY site in question. The neighboring lymph glands are similarly reddened. Besides the tuberculous changes present within the liver and spleen, these organs show on their surface many blackish- red spots varying in size from that of a pin-point to a hemp seed, and resembling very closely the ecchymosis found in some infectious diseases. On microscopical examination are found no blood extravasations, but very widely dis- tended capillaries directly surrounding the tuberculous foci. The capillaries are so densely plugged with red blood cells that it seems almost impossible for the circulation to h ave continued in these places. In exceptional cases only, are the blood vessels ruptured and the escaped blood found within the tuberculous foci. The lungs present similar changes, but not as regularly or of such characteristic appearance. The small intestine is often deeply and evenly congested. In all this symptom-complex, the never failing and almost pathognomonic feature is the hemorrhagic-like spots on the liver surface. Koch considered that the tuberculin brought about the death of the tuberculous tissue. He interpreted the disappearance of the reaction after repeated inoculations with tuberculin, as evidence that the entire tuber- culous structure had been destroyed; in other words that healing had set in. Accordingly, in the first tuberculin era, the erroneous tendency arose to consider fhose tuberculous patients as cured who after gradually dimin- ishing reactions to tuberculin had become entirely refractory to it. Actually, these individuals had merely become immunized against old tuberculin, and if another preparation such as new tuberculin had been in- jected, a reaction would have recurred. Basing their conclusions on experimental work, Wassermann, Bruck and also the author have shown that besides the factor of partial immuni- zation, it is the focal action of the tuberculin which is the beneficial agent in its therapy. The inflammatory hyperemia produced leads to a destruction of the tuberculous tissue, while at the same time the inflammatory process recedes. In addition there is a formation of connective tissue which en- capsules the focus and with it also is associated the local stimulation of antibodies. The Technique of Tuberculin-therapy. Three distinct periods can be noted in the history of this therapy. The first began in the memorable year, 1890, when Robert Koch made known his discovery of tuber- culin. At this time, the aim of tuberculin treatment was to cause very marked reac- tions and to continue with the injections until no further reaction was obtained. In lupus, glandular or bone tuberculosis 10 mg. was the initial dose. In tuberculosis of the lungs i mg. was the beginning. If the patient reacted to this amount, he received daily inoculations of this dose until no reaction appeared. Then 2 mg. T. were given and the same procedure repeated. Quite frequently, depending upon the strength of the individual concerned, 10 mg. was given as the primary inoculation in phthisis, and then rapidly increased. While Koch himself very soon recognized that this rather severe treatment was suitable only for incipient or moderately advanced cases, very sick and far advanced phthisis patients were similarly treated by many physicians. TREATMENT WITH OLD TUBERCULIN 65 Following such procedure, decidedly unfavorable results were obtained in the latter class of patients and consequently a marked waning in the enthusiasm which first greeted the tuberculin therapy was the inevitable outcome. Thus the once highly praised remedy was entirely rejected. During the second period only very few former followers of Koch continued their studies. They, however, made it their business to investigate the causes which accounted for the failure of tuberculin therapy, Their researches led to new prin- ciples in the treatment, and to more exact knowledge of its indications as well as contraindications. The success obtained by the untiring efforts of these investigators brought about after many years a revival of the interest in this therapy. It was again taken up (third tuberculin era) and there is no doubt that when properly handled, tuberculin in well selected cases is of decided benefit. Nevertheless, even at the present day, a conclusive opinion as to many of its details cannot be formed. Attempts are being made to set the individual treatment upon a biological instead of upon the more or less schematic basis thus far employed. Old Tuberculin (Tuberculin Koch 2Y). While it was the aim in the early era of tuberculin treatment to produce very strong general reactions, it is the general consensus of opinion at pres- ent that it is best so to arrange the tuberculin therapy as to avoid a general reaction and especially the fever. Gcetsch was the man who first called attention to this. February March April Warn* ^ 1 5 t -0) | 3 1 ^0.5 cj e 5 E ^ 5 | 39,5 jj jjj o 3 § t § § 38.5 £ « ft i\ ^" A °~ <* 0 ^ ^ 37,5 r ^ \ i / y j ^ / J !\ J I ^ \ \ 36.5 36.0 35.5 P V i\ i V ^ / V I/ ^ * V l^ / ^ V y ^ \s v * J v *~ 7 V I / -^ J ^ V" i ^ ^ i •4* ^ I/ \J Gewichh 50.8 kg K9 % SOi 'i Mi hg 50 1 Kg Zl CHART 2. — Example of hyper-susceptibility brought about by going back to a smaller dose. With such object in view, one must begin with small doses. Some men start with i/ioo mg., others with i/io mg. T. If no reaction is incited, the dose is increased in five to seven days to 5/100 mg. and then to i/io, 2/10, 4/10, 8/10, 1,2,4, 6, 8, 10, 20, 40, 80, 100, 150, 200, 300, 400, 600, 800, and 1000 mg. This last amount represents the maximum dose. If the patient still gives a focal reaction with such a dose, it is best to repeat it at intervals of two to four weeks, until finally no reaction is apparent. Occasionally one will advance with the doses at a more rapid rate, but in general, inoculations should not be repeated more than twice a week. 5 66 TUBERCULIN THERAPY If at any time a distinct or even doubtful reaction occurs, it is absolutely necessary to await the complete subsidence of the latter, and then the same dose is to be repeated. In such a case, an interval of at least eight to ten days should elapse. The dosage should under no condition be diminished, as thereby instead of immunity, hyper-susceptibility is the result. There- fore, patients who at a short time previously had a diagnostic tuberculin test performed should receive the highest dose employed in this test as the initial prescription for their tuberculin therapy. The following chart illustrates the condition of hyper-susceptibility occasioned by a diminution in dosage. (Chart 2.) Patient had a localized one-sided apex tuberculosis. At a diagnostic tuberculin injection, he reacted only when o.oi c.cm. T. was employed. After the interval of a month the patient was advised tuberculin treatment. Contrary to the rule just cited, he received as a first injection not o.oi tuberculin but 0.002 c.cm. T. With this small dose he already had an increase of temperature, although coming rather late, and not quite typical. After this reaction had disappeared, without any other manifestations, the same dose of 0.002 c.cm. tuberculin was repeated and as evident from the chart, a very marked response was inaugurated. This was accompanied by a chill, vomiting, headache, general pains and weakness. In addition there was a slight relapse after the aforementioned symptoms had disappeared. In order to immunize this patient against his hyper-susceptibility, it was advisable to repeat the dose of 0.002 c.cm. T. ; the reaction reappeared, but in a very much milder form. It was only after the fifth inoculation of the same dose that no reaction was in evidence. Thus was the hyper- susceptibility overcome and the patient treated in the general way. The danger of hyper-sensitiveness also exists if the same reactionless dose is too frequently repeated; especially so if the quantities injected are small. The higher the dosage, the less liable is the occurrence oj hyper- susceptibility. This question is above all to be considered when after a certain interval, a second course of tuberculin therapy is advised. In general it can be car- ried out after a period of three months, even though sometimes certain difficulties may be met with. Petruschky strongly recommended this treatment in successive stages. (Etappenbehandlung.) The author is of the opinion that it is best to retain the patient as long as possible at his acquired immunity (tuberculin) by stretching the course of treatment over a long period of time. He therefore repeats an inoculation of the maximum dose, every three or four weeks and when hyper-susceptibility arises, he changes the preparation and begins with a small dose again. As for the technical details of the treatment, several practical sugges- tions may be made. 1. The inoculation should, if possible, be given in the morning hours, for a restless night usually follows an injection in the evening. 2. It is best so to arrange the dilutions that the patient receives a frac- tion of i c.cm. at each injection. RULES FOR TUBERCULIN THERAPY 67 3. The site of injection should be alternated between the back and the breast. 4. The temperature should be every two or three hours and a kept. 5. Disturbances in the general condition taken chart of the patient without the presence of fever are to be considered in the light of general reactions just as much as fever without other disturbances. 6. The patient's weight should be taken regularly every week, and the dose should be increased provided no loss in weight has taken place. 7. In cases where the pulse increases in rate or becomes poorer in quality, the treat- ment should be undertaken very carefully and the pulse constantly kept as guide. Slowness of pulse can, as a rule, be con- sidered a signum bonum. Especially favorable for the tuberculin treatment are the individuals with a begin- ning, localized pulmonary tuberculosis, or cases of lupus, and renal tuberculosis, as re- ported by Lenhartz. The presence of fever leads some to consider such application as contraindicated. This is indeed incorrect, as frequently it is observed that a chronic fever entirely disappears during a course of treatment, and very often remains away. Even if the fever continues, a good result in the general condition of the patient is never- theless obtained. (Chart 3.) Patient H. — Nineteen years old with distinct tuberculous habitus, on admission to the medical service presented a marked infiltration and catarrh of two-thirds of the right lung with a cavity in the upper lobe; infiltration of the left lobe and a great number of tubercle bacilli in the sputum; marked weakness and continuous fever. In five weeks the patient had gained n kg. in weight — 8 kg. in one week. Simultaneously his general condition improved very much; the night sweats disappeared, and the cough diminished, but the number 68 TUBERCULIN THERAPY of bacilli still remained the same, and the physical signs of the lungs unaltered. Sub- sequently the patient received treatment with B. E.; the temperature finally sub- sided, the cough and sputum likewise, and the bacilli became few and at times en- tirely absent for several days in succession. In fact, the general condition became excellent. Objectively, there was no demonstration of catarrhal affection. It might be noted that such a remarkable increase in weight in so short a time is by no means the rule, although good effects are observed in many cases. Naturally the medical treatment should not be limited to the tuberculin therapy. Even in the immunization of healthy animals, attention is paid to their housing and feeding; so much more imperative is this consideration when applied to sick human individuals. Bearing in mind that in any treatment, success is only achieved by creating a favorable medium, the same good influences should not be neglected in tuberculin therapy. Rest and forced feeding are curative factors which one cannot omit, and the best places for obtaining these, at the beginning at least, are hospitals, sanatoriums or convalescent homes. When in such a way, the general status of the patient is improved, ambu- lant therapy may be instituted. As for the contraindications to tuberculin treatment, it is very Contraindi- difficult to set general rules. Here the opinions of various cations to authorities differ greatly. While for example, M oiler and Tuberculin others consider hemoptysis as a distinct contraindication, Therapy. Aufrecht and Kramer claim that under tuberculin therapy hemoptysis is decidedly improved. It is easy to understand this difference in attitude, if the changes in the focal reaction are considered. There is no doubt that hemoptysis may be excited by increased supply of blood and the inflammatory process associated with the inoculation of tu- berculin. The more severe the focal reaction, the greater is this possibility. On the other hand, the new formation of connective tissue and the absorp- tion of the tuberculous tissue will diminish the frequency of hemorrhage. With a general tendency toward hemoptysis, it is therefore best to wait a long time after the cessation of the latter, and then begin with small doses. The patient should be under careful observation and by constant physical examination any possible focal reaction should be controlled. If, in spite of this, hemoptysis does set in, one should not at once be discouraged. An interval of about fourteen days is to be allowed, and then the treat- ment again undertaken. Frequently, the hemoptysis will cease. If not, or if the patient loses in weight and becomes weaker, the tuberculin therapy should be discontinued. As further contraindications, Moller mentions marked general weak- ness, fever, heart affections, epilepsy, and hysteria. In full agreement with Bandelier and Rcepke, the author does not consider any of the above as TREATMENT WITH NEW TUBERCULIN 69 cause for the non-employment of tuberculin. Only where absolute ca- chexia, without any possibility for improvement exists, is this therapy to be omitted. In all other conditions, an attempt is by all means justified. Experience, of course, plays an important r61e in the selection of suitable cases. For a beginner, it is advisable to gain practice by the treatment of uncomplicated cases before undertaking those of greater difficulty. 2. New Tuberculin-Bacilli-Emulsion (B. E.) and New Tuberculin T. R. Treatment with new tuberculin follows along the very same lines set down for old tuberculin. New tuberculin T. R. is the mildest of all preparations. It is very suitable for the beginning treatment of susceptible patients. When the in- dividual does not react to large doses, it is well to start in with B. E. B. E. can also be employed without producing any reaction, although it is somewhat more difficult. The dosage scheme advised by Bandelier and Rcepke is as follows: i/iooo, 2/1000, 3/1000, 7/1000, 10/1000 mg., 15/1000, 2/100, 3/100, 5/100, 7/100, 10/100 mg., At intervals of i to 2 days; 15/100, 2/10, 3/10, 5/10, 7/10, 10/10 mg., At intervals of 2 to 3 days; 12/10, 15/10, 2, 2 1/2, 3, mg., At intervals of 3 to 4 days; 4, 5> 6, 7, 8, 9, 10 mg., at 4 to 6 to 10 days intervals. In susceptible patients, it is best to increase the dosage only by one- half mg. even when large doses are administered. Ten mg. B. E. represents the maximum dose. The author himself follows a different scheme from that of Bandelier and Rcepke. The injections are given less frequently, only about once a week, but the dose is always increased twofold, fivefold and even tenfold without any excessive reactions. Fever is obtained much less often with new tuberculin than with T. The reaction usually is in the form of lassitude, nausea, weakness, in- somnia, etc. The treatment with new tuberculin is particularly favorable in cases where a low continuous fever is present. It also is more potent in destroy- ing the bacilli of the sputum. The author therefore prefers the B. E. to all other tuberculin preparations. Bandelier and Rcepke have also ob- tained gratifying results with the B. E. therapy as is evident from the following statistics of 205 patients treated at the sanatorium at Kottbus. TUBERCULIN THERAPY Stage I II III Cured (A) 23 = ii. 22% IO = 37.O4% 13 = 10.48% o= o % Completely able bodied (BI) 98 = 47.80% 12 =44.44% 77 = 62.09% 9 = 16.63% Satisfactory result (A+BI) 121 = 59.O2% 22=81.48% 90 = 72.57% 9 = 16.63% Able bodied in terms of law (BII). . . 70 = 34.15% 5 = 18.52% 30 = 24.19% 35=64.81% Total improvement (A+BI-f BII).. Negative result (C) 191=93.17% I4= 6.83% 27 = 100 % 120 = 96.76% 4= 3.24% 44 = 81.44% 10=18.56% Bacilli in sputum on admission 114 = 55.61% 4=I4.8l% 63 = 50.81% 47 = 87.04% Absence of bacilli or sputum at dis- charge. 59 = 51.78% 4=100 % 49 = 77.78% 16 = 34.34% The objection has been frequently advanced that B. E. is absorbed with difficulty and tends to produce infiltrations. This may be readily over- come by preliminary use of T. R. or by the employment of sensitized B. E., as has been advised by Meyer and Rupple. By sensitized B. E. is understood a bacilli emulsion which has been mixed with the tuberculous serum of a horse or ox containing anti-tuber- culin. This mixture brings about a union between certain of the anti- bodies and substances contained within the bacteria. The tuberculous serum is then removed by centrifugalizing and washing the mixture with physiological salt solution. The sensitized B. E. (S. B. E.) is milder than B. E. and in its character is more like T. R. The infiltration is much less, or entirely absent, due, as a general rule, to the fact that sensitization of bacteria tends to neutralize those substances which produce infiltrations. This last has been demon- strated by the author in the case of mouse-typhoid, and swine pest bacilli, where marked infiltrations following their inoculation have been avoided by sensitization. The following chart illustrates preliminary treatment with S. B. E. followed by B. E. (Chart 4.) The patient was a female who at the time of admission presented double-sided apical and suspicious intestinal tuberculosis. Tubercle bacilli were present in the sputum. The patient was discharged from the clinic as relatively cured, i.e., all mani- festations of illness had disappeared with the exception of slight dulness over one of the apices which, however, could have been attributed to cicatrization. In addition, there was normal vesicular breathing, no temperature, no catarrh, and a good general condition. Undoubtedly it is difficult to say whether this case is cured. Only years of observation can prove this. A temporary latency of symptoms must always be considered. Suffice it to say, that the patient was a great deal improved and able to return to her work. TREATMENT WITH SENSITIZED BACILLI EMULSION On close observation of this chart it will be noticed that practically no reactions occurred in spite of the rather rapid increase in the dosage. Slight temperatures were mani- fest only occasionally (o.ooooi c.cm. S. B. E.). Even though the same dose was not repeated on account of the long intervals between the individual injections, no increased reaction appeared after < subsequent inoculation. When o.i c.cm. of S. B. E. pro- duced no reaction, the suscepti- bility to B. E. was tested. Doses of o.oooi B. E. to 0.5 B. E. were administered at short intervals, without any symptoms. Only after i.o c.cm. of B. E. was there a slight increase of temperature with rather marked general mani- festations— (headache, pains in the extremities, weakness, etc.), which subsided within twenty- four hours. On repetition of the same dose, no reaction occurred. Six days later, when for the third time i c.cm. of B. E. was given there set in a much more marked general disturbance, as evidence of hyper-susceptibility. (See Chart 2.) 3. Bovine -tuberculin. Koch's differentiation between bovine and human tuberculosis led; to the attempt at immuniza- tion of cattle with human tubercle bacilli. (Bovo-vaccine of Beh- ring, and Tauruman of Koch). Spengler tried to reverse this use and employ the milder, infec- tious bovine bacilli for the tuberculin therapy in man. He used these 1 rr! t-i o, < 1 S CnART-4- — E.H. 41 years. Intestinal tuberculosis and a double-sided pulmonary "tuberculosis.^Treatment with S. B.[E.; almost with- out any reaction. Immunization with B. E. Relative cure. Disappearance of cough, expectoration, bacilli, catarrhal symptoms of lung and intestinal symptoms. Increase in weight. / 2 ^ — 42*1 7^ wy:> U^ 7 s" s \ f / yqns 3l?« jo JO 'i *£_,, 3 & i •Jjr Msr> ' UI& jS Wf- 33 UJ3(f / ^ \ 2 7 £2 ; VK seq iyj9 "i SfQf}. jg U339 0 s s W1' 3ff U33L 0 s 7 & '•¥ ns3 ?S<£ JlOO >ou s V j s J <^ 'M ?• 9 8 ;:.:,:. 47 7-14 4i 21 20 '•.?v;r-5^ *'••?-' Indefinite 2 I 233 179 54 77 According to the researches of Doenitz, more recently confirmed and extended by Fritz Meyer, it was established that large amounts of antitoxin can even neutralize toxin already attached to the tissue cells. Men with practical experience like Heubner give 4000 units as the initial dose. In the United States doses as high as 10,000 to 100,000 1. E. have been admin- istered with good results. The view of large dosage is being gradually taken up also in Germany. At any rate it is by far better to give too much than too little. If the first injection does not suffice it should be repeated the next day. The only possible drawback associated with the use of excessive amounts is the possibility of serum sickness, to be men- tioned later. Netter has found that the administration of i gm. of calcium chloride on three successive days prevents serum sickness. The serum has as a rule been injected subcutaneously. This method is very practical and as far as anaphylaxis is concerned, is the least dangerous. The disadvantage, however, is that it is very slowly absorbed. Madsen and Hendersen-Smith have shown that but a trace of antitoxin can be found in the blood of the patient four and three-fourth hours after the injection, and only after two to three days can larger amounts be demonstrated. In view of this, Morgenroth recommends the gluteal intramuscular injection, for here a much more rapid absorption follows. In cases of dangerous illness intravenous injection may be undertaken. For this purpose Meyer advises a serum free of carbolic acid, although this is not absolutely necessary. The importance of the method of injection is clearly shown by the comparative experiments of Berghaus. In order to save a guinea-pig injected with a definite amount of toxin and followed in i hour by antitoxin, it was necessary to employ: 0.08 I. E. by intracardial injection (I. E. = antitoxin unit). 7.0 I. E. by intraperitoneal injection. 40.00 I. E. by subcutaneous injection. Thus the curative power was increased 500 fold by placing the antitoxin directly into the circulation. ,84 TOXIN AND ANTITOXIN The treatment of diphtheria must by no means be limited to serum therapy. A symptom of grave prognosis is the lowered blood pressure which must be counteracted by infusions of 1/2 liter of physiological salt solution containing five to six drops of adrenalin. The question whether the use of concentrated antitoxin is therapeutic- ally more efficient than the non-concentrated is still a matter for discus- sion. Numerous authors claim that sera of medium strengths (about 400 I. E.) are most efficient. The highly concentrated sera are much more expensive. For prophylactic purposes 500 to 1000 units injected subcutaneously usually suffice. Protection thus attained lasts about three weeks. CHAPTER VIII. TOXIN AND ANTITOXIN (continued). DEFINITION or TOXIN, TETANUS TOXIN, BOTULISM TOXIN, DYSENTERY TOXIN, STAPHYLOLYSIN. The diphtheria toxin and its antitoxin just discussed in detail is of great practical and theoretical importance, and can serve as a type of all true toxins and antitoxins. Bacterial toxins can be defined as poisons given off by the bacteria, the symptoms resulting from their action appearing after a certain incubation period. The invaded organism reacts by the production of specific antitoxins which neutralize the toxins in amounts, following the law of multiple proportions. Further analysis of this definition indicates that a substance can be considered a toxin only when it has a poisonous action, or in the words of Ehrlich when it possesses a toxophore group. This toxicity does not always manifest itself by necrosis or death as in diphtheria. More frequently the toxin has a somewhat selective action affecting a certain group of organs. Thus a toxin acting upon the central nervous system or blood is designated respectively as a neurotoxin or a hemotoxin. To differentiate a true toxin from other poisonous products obtained from bacteria, it is important to note that all true toxins are elements of secretion of the living bacteria, and can be separated from them by filtration. According to this definition, poisons contained within the bacterial bodies themselves, which may be liberated by various mechanical, physical, or chemical means, cannot be considered as belonging to the class of true toxins. These poisons are characterized by peculiar proper- ties and are known as endotoxins. In addition it may be remarked that inasmuch as a true toxin requires a period of incubation in order to manifest its action, those toxins which act spontaneously are to be ex- cluded from the former group. R. Krause nevertheless considered some of the poisons isolated from the cholera and cholera-like spirilla (El Tor Vibrio) as true toxins even though they lack an incubation period. The real essential property of a toxin is doubtlessly that one can immunize against it, and be able to demonstrate the presence of antitoxins within the serum of the immunized animal. Ehrlich claims furthermore that the amount of antitoxin produced follows the law of multiple propor- tions. By this is meant that the relationship between a definite dose of toxin and the amount of antitoxin just sufficient to neutralize it, is constant ; so that if ten volumes of toxin hold in bounds ten volumes of antitoxin, 85 86 TOXIN AND ANTITOXIN 100 volumes of toxin neutralize 100 volumes of antitoxin. This relation is best exemplified by the diphtheria toxin and antitoxin. With the other toxins, conditions are more complicated so that many objections have been raised against the above rule of multiple proportions. (Bordet, Arrhenius, Madsen, etc.) The true toxins causing infections in man are the 1. Diphtheria toxin. 2. Tetanus toxin. 3. Botulism toxin. 4. Dysentery toxin. 5. Staphylolysin and similar bacterial hemotoxins. Tetanus Toxin. The tetanus toxin is found within nitrates of bouillon cultures Tetanus of the tetanus bacillus. While partial erobiosis does not Toxin. entirely eliminate toxin formation, anerobic conditions are by far more favorable for it. The tetanus toxin is of two kinds; the tetanospasmin, and tetanolysin; the former a neurotoxin, the latter a hemotoxin. The tetanospasmin is the more important of the two for the reason that it is the agent which produces convulsions. If susceptible animals such as mice or guinea-pigs are injected subcutaneously or intra- muscularly with tetanus toxin, after a certain interval — the incubation period — they will begin to show symptoms due to tetanospasmin. They become hypersensitive to reflex stimulation; clonic convulsions and toxic rigidity of the muscles set in. In animals the spasms appear first in the group of muscles nearest the point of injection, while in man they almost regularly start in the muscles of the lower jaw. By intravenous and intraperitoneal injections, the tetanic spasm appears simultaneously in all muscles of the body; on intracerebral inoculation, Roux and Borrel describe the occurrence of epileptiform seizures, polyuria and certain motor disturbances — the entire set of symptoms being known as cerebral tetanus. Rabbits receiving very small amounts of tetanus toxin intra- venously die after gradual emaciation and marked cachexia. This type of infection is designated by Doenitz as tetanus sine tetano. If taken per os, tetanus toxin manifests no poisonous effects. Tetanospas- min is a distinct nerve poison especially affecting the central nervous system. Experiments by Wassermann and Takaki have demonstrated a close affinity existing between the tetanus toxin and certain organs. These organs differ in different species of animals. Thus in man, horse, and guinea-pig only the central nervous sys- tem, while in rabbits in addition to this, also the liver and spleen take up the tetanus poison. If an emulsion of brain tissue and a fatal dose of tetanus toxin are mixed and the mixture injected into mice, the latter remain unaffected. According to Doenitz TETANUS TOXIN 87 only the gray matter and not the white substance of the brain possesses this absorption power. If the brain emulsion is boiled, it loses this affinity for the toxin. Concerning the way by which the toxin reaches the central nervous system, opinions vary. Most writers, especially Meyer and Ransom, consider that the journey is made along the nerve paths. Zupnik on the other hand believes that it is distributed through the blood stream and is taken up not only by the nervous system, but also to a great extent by the muscles. That tetanus toxin is very labile is well known. According to Kitasato, five minutes at 65° C. or twenty minutes at 60° C. is sufficient to weaken the toxicity to a great extent, in fact almost to destroy it. Light has a similar effect upon it. Careful as its preservation may be, the soluble tetanus toxin soon becomes attenuated. Hence the best way of keeping it in stock is in a dry form. For estimating the strength of the toxin white mice are employed and are injected subcutaneously with fresh sol- uble toxin, the lethal dose being the amount which kills the animals in four to five days. Animals more susceptible than mice are horses, they being twelve times as sensitive and guinea-pigs six times as much. Hens possess greater power of resistance, being 30,000 times less susceptible to the toxin than mice. Tetanolysin acts upon the red blood cells and disintegrates them. The erythrocytes of goats, sheep and horses, are best suited for experiments to demonstrate this action. Ehrlich showed that the tetanolysin and the tetanospasmin are really two distinctly different toxins and not one toxin with a twofold function. When tetanus poison is mixed with red blood cells the tetanolysin is absorbed and the tetanospasmin remains free. Even the antitoxins of these two are different. As far as the standardization of the tetanus serum is concerned, it follows along the same lines as the diphtheria serum, i.e., the L + dose of toxin being the one employed. "In America the method of standardization was regulated by a law passed in July, 1908, based upon the work of Rosenau and Anderson at the United States Hygienic Laboratories at Washington. Their unit of antitoxin is ten times the smallest amount of serum necessary to save the life of a guinea-pig for ninety-six hours, against the official unit of standard toxin. This toxin unit consists of 100 minimal lethal doses of a precipi- tated toxin preserved at the hygienic laboratory of the Public Health and Marine Hospital Service. At the hygienic laboratory at Washington a standard toxin and antitoxin are preserved under special conditions, and standard toxin and antitoxin, arbitrary in their first establishment, are kept constant by being measured against each other from time to time. For details of this standardization the original article in the United States Hygienic Laboratory Bulletin 43, 1903, should be consulted." 88 TOXIN AND ANTITOXIN Also in regard to the efficiency of serum therapy in tetanus Serum opinions differ. There is, however, no doubt that a certain Iherapy of .. Tetanus amount of reliance can be placed upon this treatment. Fail- ures in successful application are ascribed to the different paths by which the toxin and antitoxin travel. The former is carried by the nerve fibers, the latter by the blood stream. Thus the serum instead of being given subcutaneously, as is the general rule, is admin- istered by intraneural, intracerebral, and subdural injections. 100 to 200 units should be injected subcutaneously at the site of the infection or its vicinity and in addition the nerve fibers supplying the infected region should be exposed and inoculated with moderate doses of antitoxin at various points along their centripetal course. The prophylactic use of tetanus serum has met with better results. Behring advises the administration of ten to twenty antitoxin units subcutaneously. Calmette sprinkles upon the open navel at birth a powder made of dried serum as a prophylactic against tetanus neonatorum. Bockenheimer advises an ointment containing the antitoxin as a dressing for suspicious wounds. The Botulism toxin is the poison produced by the bacillus Botulism botulinus. This is the exciting agent of a type of meat and Toxin. sausage poisoning described by van Ermenghem in 1896 as Botulism. The bacillus botulinus is a very actively motile anerobic bacterium which grows at room temperature and presents marked gas and toxin formation. A medium in which the toxin is readily produced consists, according to Ermenghem, of an alkaline bouillon made in the form of an infusion from ham with the addition of i per cent, of glucose, i per cent, of peptone and i per cent, of sodium chloride. The toxin can be demonstrated after 3 weeks of growth, and is then obtained by bacterial filtration. The cultures have a sour odor like butyric acid. The toxin deteriorates easily when exposed to air and light. It is therefore preserved in brown, sealed vials, and kept on ice; or in a dried form in vacuum. Heating the toxin for three hours at 58° C. or one-half hour at 80° C. destroys its toxicity. Acting unrestrained, the botulism toxin is one of the severest of poisons. It affects susceptible animals even in minutest doses. In contradistinc- tion to other toxins it is fatal even when taken per os. The characteristic symptoms produced by botulism intoxication consist of hyper- secretion of mucus from the mouth and nose, paralysis of eye muscles, urine retention, obstipation, dysphagia, aphagia, and aphoria. No fever, nor any sensitory disturb- ances are in evidence. Death takes place because of bulbar paralysis accompanied by respiratory and cardiac failure, The poison is absorbed or arrested in the central nervous system, o.i BOTULISM TOXIN 89 c.cm. of an emulsion of central nervous tissue neutralizes three times the fatal dose for mice. Lecithin, cholesterin, as well as fatty substances like butter and oil, act in a similar manner. Monkeys, rabbits, guinea-pigs, mice and cats are susceptible to the toxin. Cats usually exhibit the most characteristic clinical picture. Localized and almost pathognomonic paralyses occur in the form of prolapse of tongue, marked mydriasis, aphonia, aphagia, etc. In mice, paralysis of the hind extremities sets in after quite a small dose; and death follows in a few hours. In rabbits and guinea-pigs, moderate doses (0.0003-0.001 c.cm.) occasion no mani- festations during the first two to three days, but subsequently, the above-mentioned paralyses arise and in several hours the animals expire. With larger doses (o.i to 0.5 c.cm.) the incubation period lasts only a couple of hours and then dyspneic attacks usually succeeded by motor paralysis and death are the consequences. The strength of the botulism toxin is ascertained by injecting guinea-pigs subcu- taneously and observing the time when loss in weight, flabbiness of abdominal muscles and death occur. The following chart by Madsen exhibits the above principle. (+ means "death.") Dose in c.cm. Result Dose in c.cm. Result cxoois +on i st. day 0.0009 Weakness in 3 weeks. 0.0015 + after 11/2 days 0.0009 Loss in weight. 0.0015 + after 2 days 0.0009 Loss in weight. 0.0013 + after 2 days 0.0007 Loss in weight in 2 weeks. 0.0013 + after 5 days 0.0007 Loss in weight in i week. 0.0013 + after 6 days 0.0007 Loss in weight in i week. O.OOI + after 4 days 0.0005 Loss in weight in i week. o.oo i + after 5 days 0.0003 Practically no symptoms; only O.OOI -f after 51/2 days several days of weakness. Kempner immunized goats against botulism toxin and proved the presence of anti- toxins within their sera. Immunization of rabbits and guinea-pigs is only feasible if primary inoculations are made with a toxin previously attenuated by heat for one- half hour at 60° C. Recently Wassermann has immunized horses against this toxin. In mew of the high mortality and lack of any other specific medication, the use of this serum is strongly advised. In animal experimentation it shows itself of undeniable value. As for its effects in man, it has not been employed frequently enough to judge. The botulism toxin and antitoxin unite only very slowly. Otto and Sachs have shown that the inoculation of rabbits with a three hours old mixture of toxin and anti- toxin occasioned greater toxic effects when administered intravenously than when given subcutaneously. Only in mixtures twenty-four hours old was this difference overcome. pO TOXIN AND ANTITOXIN Dysentery toxin was first demonstrated by Conradi. Subse- Dysentery quently from experiments by Rosenthal, Todd, Kraus and Toxin. Doerr, etc., it became evident that this was a true toxin and not an endo toxin as was originally considered. Only the Kruse- Shiga type of bacillus forms a toxin; for the Flexner type, no definite toxin has as yet been isolated. Recent investigators, however, especially Kraus and Doerr, are inclined to consider the human dysentery of the Kruse- Shiga origin in the light of an intoxication or toxemia similar to diphtheria. The lesions in the large intestine where the bacteria accumulate can be compared to the diseased diphtheria tonsils, while the other manifestations, as the cerebral symptoms, cardiac disturbances, nervous sequelae, eye affections, etc., can be taken as expressions of the toxemia. Like the other described toxins, the dysentery toxin can be obtained by nitration of bouillon cultures. The meat infusion must be quite alkaline. The optimum alkalin- ity, according to Doerr, is obtained by adding 0.3 per cent, soda to litmus neutral bouillon. The precipitate thus formed which increases on sterilization should not be removed by nitration. Doerr also advises finely powdered chalk (20 gms. pro liter) to be added to the weakly alkaline bouillon before the last sterilization. The toxin is formed very gradually; the maximum is derived after two to three weeks. The gray white pellicle upon the surface of the culture can be taken as an indicator for the amount of toxin present. According to Kraus a good dysentery toxin can also be made by emulsifying the bacteria (grown upon agar) in physiological salt solution and filtering through Reichel niters. The toxicity of individual strains of dysentery bacilli varies greatly. The strength of the toxin is diminished by heating for one to two hours at 60° C. Higher temperatures destroy it: 80° C., in three minutes and 90° to 100° C. in one minute. Acids destroy the toxin probably by the formation of a non-poisonous compound. The addition of a strong alkali restores the toxicity. Its preservation can be accomplished in a fluid state under the cover of toluol. The action of dysentery toxin can best be studied by its effect upon rabbits after intravenous inoculation. Large doses kill the animals in very short time, six to seven hours. The ordinary lethal dose produces characteristic symptoms consisting of paresis, diarrhea, which may be bloody, paralysis of the bladder, hypothermia, etc. Death takes place in three to four weeks. Given subcutaneously, or intraperitoneally, the toxin has only a very mild action. The incubation period is especially prolonged. Given per os, no effect is in evidence. Besides rabbits the other susceptible animals (to large doses) are monkeys, cats and dogs; chickens, pigeons and guinea-pigs are, in the opinion of Kraus and Doerr, not at all affected by the toxin. DYSENTERY SERUM 9 1 The intestinal changes found at post mortem examination of the animals very closely simulate the pathological alterations occurring in man. A hemorrhagic ne- crotic enteritis is present which in rabbits is regularly localized in the appendix and ce- cum, while in dogs the entire intestinal tract and especially the duodenum is attacked, and in monkeys the lower part of the intestine is involved. The associated nervous manifestations are, according to experiments of Dopter, referred to changes in the spinal cord itself. These are of a nature similar to acute anterior poliomyelitis. Occasionally a polio-encephalitis is added. An antitoxic dysentery serum is obtained by immunization of Dysentery horses and goats. Various methods have been employed Serum. to obtain it. Of the older authors, Shiga and Kruse im- munized animals with dysentery bacteria and thus produced a serum which possessed besides its bacteriolytic and agglutinating properties also a weak antitoxic action. Rosenthal, Todd, Kraus and Doerr employed the toxin itself for immunization purposes. In standardization of the serum the properties to be determined are three. [Kraus and Doerr employ rabbits in this work.] 1. Its power of neutralizing toxin in vitro. — Toxin and antitoxin are mixed in various proportions; the mixtures allowed to stand fifteen minutes at room temperature and then injected intravenously. 2. Its power of neutralizing toxin in vivo. — The toxin is injected into the right vein and the antitoxin at the same time into the left vein. 3. Its curative power. — The antitoxin is injected at various intervals after the toxin. These three therapeutic factors do not appear simultaneously. The power of neutralization in vitro is first in evidence. Only very much later does the serum de- velop its curative strength and ability to neutralize in vivo. In animal experimentation, the antitoxic serum exhibits its neutralizing and cura- tive properties only if injected intravenously. Dysentery serum has been employed with fairly good results. Only infections caused by the Shiga-Kruse bacilli can, however, be benefited. The serum should be given subcutaneously and as early in the stage of the disease as possible. The dose advised by the authors varies greatly, on account of the difference in strength of the numerous sera and the severity of the infection. In cases of moderate illness, it is as a rule sufficient to give one to two injections of 20 c.cm. of a strong antitoxic serum which can neutralize toxin both in vivo and in vitro. Vaillard and Dopter have injected as many as 80 to 100 c.cm. in the severer cases. The good effect of the serum manifests itself by an improvement in both the general and local symptoms. If high fever exists, the tempera- ture sinks. If collapse temperature is present, it usually rises. The subjective complaints, especially the sleeplessness, improve. The blood TOXIN AND ANTITOXIN in the stools disappears; the movements of the bowels become less fre- quent and the severe pains are absent. Finally, the consistency of the stools changes and at the end becomes normal. Prophylactic use of the serum has met very favorable confirmation in the work of Kruse, Vaillard and Dopter, and Rosculet. Rosculet's statistics are especially interesting. In 1905 during a dysentery epidemic in Roumania, Rosculet injected eighteen apparently healthy individuals living at the homes where dysentery cases existed, with 5 c.cm. of the serum. Eighteen similar patients were removed from the dysentery surroundings, but received no serum. The results were that of the first group no fresh cases of infection arose, while of the control group fourteen became infected. It is rather premature to determine definitely the value of the dysentery serum therapy; enough has been seen, however, to advocate its use when- ever possible. Staphylolysin, or Staphylohemotoxin. — According to the experi- Staphyloly- ments of M. Neisser and Wechsberg the pyogenes staphylo- sin. cocci produce a typical hemolysin which is identical for both the aureus and albus cultures. By immunization with this hemotoxin, an antihemotoxin (antilysin) is obtained. Neisser and Wechsberg further discovered that human serum and serum of certain animal species normally contained antistaphylolysin ; less, however, in amount than immune sera. Working on the principle that in staphylo- coccus diseases, a hemotoxin is formed which incites the development of antihemotoxin for the protection of the animal, Bruck, Michaelis and Schulze attempted to employ the presence of antistaphylolysin in the serum as evidence of the existence of staphylococcus infections. As Staphylolysin, a twelve to thirteen day old bouillon culture of freshly isolated staphylococcus pyogenes serves very well. This can be preserved by adding 5 c.cm. of the following mixture to 100 c.cm. of the bouillon filtrate: 10 carbolic, 20 glycerin, 70 aqua. The hemotoxin content is approximated according to the following scheme: Amount of filtrate Fresh rabbit blood Physiological NaCl solution Result of hemolysis after 2 hours in incubator at 37° C. and 24 hours in ice box 0.2 c.cm. i drop. up to 2 c.cm. complete. o. i c.cm. i drop. '. up to 2 c.cm. complete. 0.05 c.cm. i drop. up to 2 c.cm. complete 0.025 c.cm. i drop. up to 2 c.cm. complete. o.oci c.cm. i drop. up to 2 c cm. incomplete. 0.005 c.cm. i drop. up to 2 c.cm. layer of red blood cells. STAPHYLOLYSIN CONTENT OF SERUM 93 Thus 0.025 is the smallest dose which can completely hemolyze the given quantity of red blood cells. The amount of antilysin is estimated by adding varying amounts of serum to the constant minimal hemolytic dose of the staphylolysin and determining what amounts of serum contain enough antilysin to prevent hemolytic action of the staphylolysin. It is best to allow the staphylolysin and serum to remain mixed for some time before adding the rabbit blood, so as to give the antitoxin a chance to neutralize the toxin. As every normal serum contains a certain amount of antilysin, it is necessary in order to obtain the pathological variations, to use a normal serum as a control. Such a serum, o.i c.cm. of which just suffices to neu- tralize twice the minimal hemolytic dose, was dried in vacuum and used by Bruck, Michaelis and Schulze, as standard serum. Estimation of antilysin content of the standard serum: Twice the minimum hemolytic toxic dose. 0.05 0.05 • 0.05 0.05 0.05 Standard normal serum O. 2 O. I 0.05 0.025 O.OI Result after 24 hours No No Slight Moderate Complete hemolysis. hemolysis. hemolysis. hemolysis. hemolysis. These mixtures were allowed to stand for one hour at 37° C. and then i drop of rabbit's blood was added, allowed to remain for two hours at 37° C. and twenty-four hours in the ice box. :' The standard serum was always freshly prepared in the form of a 10 per cent, solution in distilled water (o.i : i). In the above manner the antilysin content of the serum from patients with distinct or suspicious staphylococcus infections was estimated. The completely neutralizing dose of the standard serum (o.i above) was taken as i and the neutralizing dose of the serum for examination compared with this; if 0.05 c.cm. of a serum x neutralized the same amount of toxin as o.i of standard serum, the antilysin value of the serum x was 2. From the comparative studies of Bruck, Michaelis and Schulze it was concluded that most of the normal sera had values ranging from i down; occasionally results as high as 5 were obtained. Out oj twenty -five cases of staphylococcus infections nineteen gave values varying from 10 to 100. Fig- ures as high as these can, according to these authorities, become of valuable aid in diagnosis. Although these findings were corroborated by Arndt and others, the method cannot as yet be classed among those of clinical diagnostic importance. Similar study of other infections has not been undertaken. 94 TOXIN AND ANTITOXIN In addition to the toxins reviewed in these chapters, recent work has proven that toxins may, under certain conditions, be derived from bacteria other than those mentioned, e.g., cholera, typhoid bacteria and meningo- cocci. Problems such as these are still, however, open to scientific discussion; consequently no exact statements can be made here. CHAPTER IX. THE TOXINS OF THE HIGHER PLANTS AND ANIMALS AND THEIR ANTI- BODIES. FERMENTS AND ANTIFERMENTS. The toxins thus far studied were all secretory products of bacteria. This power of forming toxins is not, however, limited to bacteria alone, as there is a class of higher plants and animals that produce characteristic poisons against which immunization can be undertaken and an antitoxic serum obtained. Aside from snake poison, the members of this group bear little practical medical interest. The detailed study of these plant toxins (Phytotoxin) and those of animal origin (Zootoxin) has, however, greatly increased the theoretical knowledge of the phenomena of reaction and immunity. Phytotoxins. The most important phy to toxins are: 1. Ricin. 2. Abrin. 3. Crotin. Ricin is a deadly poison, of which the smallest fractions of a milligram Ricin. are sufficient to kill rabbits. Like bacterial toxins, ricin requires for its action an incubation period of at least twenty-four hours. The typical postmortem findings consist of redness and swelling of Peyer's patches. Ricin is a hemo toxin; if mixed, as an emulsion, with red blood-cells, the erythrocytes sink to the bottom and are agglutinated. Ehrlich succeeded in immunizing animals against ricin by first giving it to them per os in increasing doses for a long period of time, and later on by subcutaneous injection. The antitoxic serum thus produced neutralizes the poisonous action of ricin both in vivo and in vitro. Abrin, a vegetable poison, is obtained from jaquirity (Abrus precatorius) Abrin. and in its action closely resembles ricin, but is less poisonous. It is a marked irritant of the conjunctiva and was at one time employed in cases of trachoma. Roemer found that by repeated instillation of abrin into the same conjunctival sac, no reaction was ultimately obtained (local immunity), while the conjunctiva of the other eye retained its susceptibility. If the instillation was continued for a long period of time, a "general immunity" was attained which extended to the conjunctivas of both eyes. As a result, in the serum of such animals antiabrin could be demonstrated. 95 96 THE TOXINS OF THE HIGHER PLANTS AND ANIMALS Crotin is the seed of croton tiglium, a substance less poisonous than Crotin. either ricin or abrin. It does not agglutinate, but produces hemolysis of rabbits' red blood cells. Toward the red blood cells of other species (e.g., bird), it is entirely inactive. The immunization of Babbits is readily brought about by subcutaneous injections. Their serum neutralizes the hemotoxic action in vitro. The Zootoxins. Most important of the animal toxins are 1. Phrynolysin (toad poison). ] _ A T_ i . / ., . \ \ Simple hemo toxins. 2. Arachnolysm (spider poison), J 3. Snake poison. } _ 0 . . Lecithin producing 4. Scorpion poison, > _ . -n hemo toxins. 5. Bee poison, The one striking characteristic of toxins, that an immunity can be raised against them, is also possessed by these poisons. Aside from this fact they present many variations from the true class of toxins. Most of these poisons are complex, i.e., they contain more than one toxin and all are hemotoxic. Toad poison is obtained by rubbing up the skins of the Bombinator igneus; spider poison by trituration of the living "cross spiders" (Epeira diadema) in three or four times the amount of physiological salt solution containing toluol. The toad and spider poisons contain simple hemotoxins, that is to say, by the mixture of small amounts of this toxin with erythrocytes absolutely serum-free, hemolysis of the latter takes place. Not all species of blood are affected alike. The red blood corpuscles of sheep, goats, and rabbits are especially adapted for experiments with phrynolysin, while rabbits', rats', and human blood is more suitable for arachnolysin. Immunity of rabbits is easily attained. Snake Poisons. The most familiar poisonous snakes are the Cobras (Naja) of India and Indo-China which belong to the family of Colubridae, the European viper, and the American rattlesnake; the last two being of the Viperidae species. The poisons of these two families show great individual differences. Thus, those of the Colubridae group are decidedly thermo-resistant (temperatures as high as 100° C.) while the viper's poison is entirely destroyed at a tem- perature varying between 80 to 85° C., and markedly weakened at 7o°C. Snake poisons, as a rule, produce both local reactions at the point of the bite, and severe general disturbances. The cobra bite is only slightly painful. A characteristic feeling of stiffness extends from the point of infection over the entire body. In several hours a rapidly increasing weakness sets in terminating in deep coma and death. COBRA HEMOLYSIS 97 The viper bite incites a very severe local reaction. The point of infection is red, extremely painful and swollen. Convulsions, hemorrhages, followed by delirium which finally changes into stupor are manifest, and death takes place in one to three days. If the poison gets into the circulation directly, death is likely to occur in a few minutes. The prognosis of a snake infection depends largely upon the situation of the bite. The greater the blood supply of the infected area the more dangerous is the result. Bites received through the clothing are relatively less dangerous, as a great part of the poison remains adherent to the clothing. Snake poisons act primarily upon the nervous system and blood, although they exhibit a number of other toxic and ferment properties. Thus viper toxin occasions immediate coagulation of the blood by its action upon the vascular endothelium and has for this reason been called by Flexner and Noguchi, "Hemorrhagin." Furthermore, all snake poisons have a hemolytic power. Cobra hemolysis represents one of the most interesting of Cobra biological phenomena, and since it may possibly be employed Hemolysis. in clinical methods of examination its action will be here reviewed. Cobra hemotoxin is characterized by its power of dissolving the red blood corpuscles of certain kinds of animals (ox, sheep and goat) only in the presence of serum. Other red blood cells do not require any serum for their hemolysis (dog, guinea-pig, man, rabbit, horse). If the red blood corpuscles of the first group of animals washed free of their serum are mixed with cobra poison, no hemolysis takes place. On subse- quent addition of any fresh serum, hemolysis is in evidence. (Flexner, Noguchi.) The agent which activates the hemolytic substance belongs undoubt- edly to the class of lipoids. Of these, lecithin stands pre-eminent. It is, however, by no means certain whether that is the only or the most im- portant activator. Some sera exhibit this activating influence only when first heated. In their unheated state they are entirely inactive. Other sera act in a manner decidedly the reverse. Kyes and Sachs mention that this depends altogether upon the nature of the lecithin union. The following table shows the various combinations and their resultant action. 98 THE TOXINS OF THE HIGHER PLANTS AND ANIMALS Combination Power of serum to activate the hemolysis Serum Red cells If serum is unheated If serum is heated at 56° C. 65 to 100° C. Horse Ox. ! I +weak hemolysis. Horse Ox Horse Horse Ox Ox. Sheep Ox Sheep Sheep Human Ox. Human Human Rabbit Ox Guinea-oiff Ox.. Guinea-pig Rabbit -{-Signifies hemolysis. — Signifies no hemolysis. Cobra Hemolysin Test. 1. Washing of Erythrocytes. — The blood is collected into sterile flasks containing sterile glass beads. It is shaken and thus defibrinated to prevent coagulation. The defibrinated blood is next centrifugalized and the serum separated and drawn off with a pipette. The red blood cell sediment is mixed with physiological salt solu- tion and again centrifugalized. This procedure is repeated several times until all the serum is removed. The red blood cells are used in a 5 per cent, suspension; i.e., i part of washed erythrocytes suspended in 19 parts of saline. 2. The Activating Agent. — As an activating agent 0.2 c.cm. of serum or a o.i per cent, lecithin solution is employed. The lecithin can be kept as a stock solution consisting of i g. lecithin in 100 c.cm. of methyl alcohol. A o.i per cent, solution of the stock mixture is made by mixing o.i c.cm. of the solution with 9.9 c.cm. of physiological salt solution. 3. The snake poison hemotoxin is resistant toward heat so that it may be heated to almost 70° C. without interfering with its activity. Cobra poison contains the greatest amount of hemotoxin. While i mg. of cobra toxin hemolyzes i c.cm. of 5 per cent, horse's red blood cells in five to ten minutes, a similar amount of viper toxin requires thirty minutes for the same action. V. Dungern and Coca explain this type of hemolysis by the existence of a ferment within the snake poison which breaks up the lecithin with the liberation of oleic acid. This acid has long been known as a hemo- lytic agent. The necessity for adding lecithin or serum to certain species of blood is explained by the variability in the lecithin content of the erythrocytes, or a variability in the lecithin union. MUCH AND HOLZMANN " PSYCHO-REACTION 99 Calmette notes in the blood of tuberculous patients more than the nor- Cobra loxm , ri ... . . . . . . . . . . mal amount of lecithin; for that reason their serum can be used in very small doses to activate the cobra hemolysin. By this means he attains , . a diagnostic reaction for tuberculosis. On examination of the blood of luberculosis. . , . ,. . , . . . 177 tubercular individuals he has found: 78 per cent, of positive reactions in the first stage of tuberculosis. 57 per cent, of positive reactions in the second stage of tuberculosis. 70 per cent, of positive reactions in the third stage of tuberculosis. Szaboky has confirmed these findings, but not enough control examinations of normal individuals or of other infections had been made to firmly establish the diagnostic value of the test. At the instigation of the author, Alessandrini repeated the work of Calmette but came to different conclusions. The reaction was not found specific for tuberculosis; it was given by various other diseases. Furthermore Alessandrini could not confirm the hypothesis that hemolysis was dependent upon the lecithin or lipoid content of the serum. By simple mixture of cobra poison and horse's erythrocytes, he found that the resistance of the cells toward any traumatism becomes greatly diminished so that hemolysis can be attained even by hypotonic salt solutions in concentrations which under normal circumstances are entirely ineffective for horse's red cells. The hemolysis of snake poison can be overcome or interfered with by the addition of large amounts of normal serum, cholesterin, and small amounts of snake poison serum. Much and Holzmann have recently described the so-called "Psycho- The Psycho- reaction" which can be explained thus — Normal serum, when added to reaction, a mixture of cobra extract and human red blood-cells will not interfere with consequent hemolysis. If, however, the serum is obtained from patients suffering from depressive mania, circular insanity or dementia praecox, and added to the cobra extract and human red blood-corpuscles, the expected hemolysis does not take place. One would naturally suppose that this fact would be employed for clinical diagnosis, but unfortunately it has been generally proven by most authori- ties that it is altogether impossible to do so for the simple reason that it is not abso- lutely specific. Bauer has found the same reaction with navel blood. It is probable that the interference with hemolysis is brought about by an increase in the cholesterin of the serum — a greater possibility in diseases of the central nervous system than in any physiological or other pathological condition. Cobra In immunizing laboratory animals one cannot start with Immunity, inoculations of the unaltered snake poison. Phisalix and Bertrand begin with subcutaneous injections of a toxin heated to 75° C. and after two days use one-half of the minimal lethal dose of the unaltered toxin. Calmette weakens the cobra poison by the addition of an equal amount of r per cent, gold chlorid, and after four such injections with increasing amounts at each time, the pure toxin in very small doses is employed. IOO THE TOXINS OF THE HIGHER PLANTS AND ANIMALS In the same manner Calmette immunized horses and obtained highly antitoxic sera. He tested the strength of these sera, as follows: 1. Upon Rabbits. — Each animal received an injection of 2 c.cm. of the serum into the vein of one ear, and after two hours i mg. of toxin into the vein of the other ear. A control animal was similarly treated with toxin only. The latter animal died in a half hour, while the former remained alive. 2. Upon While Mice,. — Diminishing amounts of serum were mixed in test tubes, with o.oooi gm. of toxin (in i per cent, solution) and the mixtures injected into the mice. The amount of serum which should completely neutralize the toxin must not be more than 0.03 c.cm. According to Calmette one can judge the efficiency of an immune serum by its antihemolytic power, inasmuch as the hemotoxic and neurotoxic actions run parallel. This is denied by Noguchi. The scorpion and bee poisons display properties similar to those of the cobra poison. They also combine with lecithin to produce hemolysis. Thus far, it has been shown that the lipoids, especially lecithin are actively associated in the hemolysis of erythrocytes; whether the toxin combines with the lipoids and forms a toxolipoid (toxolecithid) which is hemotoxic, or whether as v. Dungern believes, the hemolytic action is due to the fatty acid derived from the lecithin by the ferment action of sub- stances contained in the poison, has not been definitely proven. Quite recently it has been thought that pernicious anemia and parox- ysmal hemoglobinuria are closely associated with such toxolipoids. Tallquist obtained from a Bothriocephalus latus, a hemotoxic Pernicious poison of a lipoid nature which experimentally produced a Anemia, blood picture characteristic of pernicious anemia. But it would be incorrect to associate all forms of pernicious anemia with tape- worm poison; more probable is it that hemo toxins are formed within the organism itself. In paroxysmal hemoglobinuria a hemotoxin of very peculiar Paroxysmal properties is found circulating in the blood. Hemoglo- It can be demonstrated as follows. binuria. x> Ehrlich's Method. — One of the patient's fingers is tightened by means of a small tourniquet and kept immersed in ice-cold water for half an hour. Some blood is then collected into a capillary pipette, from the finger thus tied, and as a control, blood from a finger of the other hand is drawn off. This is allowed to clot and then centrif ugalized. The results are that the serum from the finger held in the ice water is tinged red from dissolved hemoglobin while the control serum is normally pale. 2. Donath-Landsteiner' 's method repeats Ehrlich's experiment in vitro. SERUM TEST FOR PAROXYSMAL' ttEMtfGrOBINURlA IOI The patient's and the control individual's 'sertfni are eUtfe tftixed with washed human erythrocytes in various proportions. It does not matter whether the red blood cells are obtained from the patient or normal individ- ual. The mixtures are allowed to remain for one-half to one hour in the ice box and then from one to three hours at a temperature of 37° C. The serum from the paroxysmal hemoglobinuria patient shows hemolysis. A control tube containing the same ingredients, in the same proportions and maintained at either cold or warm temperatures, but not at both in succession as above, exhibits "no hemolysis. The hemolytic process in this disease is of a complex nature. In the cold, one element combines with the erythrocytes, and at high temperature another unfolds hemolytic tendencies. Some sera lacking entirely or not having enough of the second element in the serum, demonstrate no hemoly- sis. But on addition of some normal serum hemolysis occurs. It can there- fore be concluded that the second factor which acts in the heat is present within normal serum, while the first substance, the specific one, is found only in the blood of those suffering from paroxysmal hemoglobinuria; (and according to Donath and Lands teiner in 10 per cent, of cases of general paralysis). It is, in addition, the author's opinion, that similar toxic substances exist in the blood of epileptics and idiots. Not all cases of paroxysmal hemoglobinuria possess this characteristic hemotoxin. In some it is only found periodically. No explanation has as yet been offered for these varying phenomena. Attempts have been made to ascertain whether the hemotoxin is stimu- lated by an external agent or by infection (lues, malaria, trypanosomiasis) or whether it is of endogenous origin. The answer is still for the future to disclose. The Antiferments. Ferments are very closely allied to toxins in their biological structure. By the immunization of animals with ferments in as pure a form as possible, antiferments can be demonstrated. Just like antitoxins, antiferments can neutralize their respective ferments in vitro. As to their presence, it is quite important to know that they are found in normal serum in certain small quantities (together with antitoxins). The difference in their pres- ence in a normal serum and that in an immune, is purely a quantitative one. The antiferments thus far demonstrated are Antilabferment. Antipepsin. Antitrypsin. Antisteapsin. Antifibrinferment. // is difficult to obtain by immunization an antiferment serum of very high 102 THE TOXINS OF THE HIGHER PLANTS AND ANIMALS strength. Probably the normal organism is so regulated that it compensates for any increased amount of antiferment. Till recent times the demonstration of antiferments bore no clinical Antitrypsin. interest. The antibodies of the proteolytic enzymes first began to attract attention when the inhibitory influence which blood serum has upon the autolysis of organs was proven. It was Jochmann and Muller who showed in connection with their studies of the proteolytic ferments of leucocytes, that apart from these, the serum itself possesses an inhibitory influence upon the leucocyte fer- ment. This is found to be especially marked in diseases associated with great destruc- tion of leucocytes. Following them, Marcus, as well as Brieger and Trebing, discovered a restraining influence in the serum upon the action of pancreas trypsin and proved that the so-called antitrypsin was considerably increased in carcinoma patients. Bergmann and Meyer, also working along these lines, then demonstrated that the wrongly called "carcinoma reaction" was by no means specific for car- Brieger's cinoma, but was found in a large number of other diseases. It cannot, Cachexia as Brieger later announced, even be considered as a criterion for cachexia Reaction, (cachexia reaction). Undoubtedly, the already normal antiproteolytic power of the serum can be considerably increased in animal experimentation by a group of well-known pro- teolytic agents, and especially by leucocyte ferment and pancreatic trypsin. To differ- entiate between antileucocyte and antitrypsin ferment in the narrow sense of the word, is impossible. The one "immune serum" (sit venia verbo, if one can speak of immune serum in this sense) neutralizes the other antigen. Clinically a high and tryptic titer of the serum is found in about 90 per cent, of carcinoma patients, and is almost regu- larly observed in infections with high fevers as typhoid, severe articular rheumatism, sepsis, etc. In penumonia there is found during the infection a marked change from an excessively high to a low titer. In Morbus Basedowii (as well as in experimental thyroid feeding) it is almost the rule to find a high antitrypsin content, but one must always keep in mind that even few normal individuals show a similar increase. The clinical diagnostic importance of the antitrypsin titer is slight in comparison with its experimental increase. In accord with the findings in Basedow's disease, and in thyroid feeding it may be considered as an outcome of increased proteid destruction (hyper-production of proteolytic ferments in the tissues ?) . Leucocyte ferment has been found of practical use in the treatment of cold abscesses, i.e., in processes where lympho- cytosis and failure to produce polynuclear leucocyte ferment is present. On the other hand antitrypsin or antileucocyte ferment or even normal serum is employed to counteract inflammatory processes, i.e., to neutralize the excessive production of the leucocyte ferments, with apparent success (Leucoantifermentin, on the market). According to recent findings, the antitrypsin titer of the mother's blood increases markedly during the period of labor, while that of the fetus remains unaltered. There are two methods for the antitrypsin determination. The first was devised by Jochmann and Muller for proving the presence of leucocyte ferment and its antiferment, and then similarly employed by Marcus in the study of pancreatic ferments. Its principle depends upon the digestive action of proteolytic ferments upon serum albumin. When a drop of trypsin is placed upon a LofHer's serum plate, after a little while, a clear spot appears where the trypsin was brought into contact with the plate. ANTITRYPSIN DETERMINATION I03 If to this trypsin an amount of serum is previously added, which fully neutralizes the digestive action, no clear zone appears upon Loffler's plate. The details of this procedure are as follows: The ferment solution consists of o.i gm. trypsin, well shaken with 5 c.cm. of undiluted glycerin and 5 c.cm. of distilled water, then left in an incubator for a half hour at 55° C., then again shaken and filtered. The serum is mixed in small test-tubes or upon a glass slide with varying amounts of the trypsin; thus i loopful of serum is mixed with 1/2, i, 2, 3, 4, etc., up to 20 loopfuls of the trypsin solution and of each of these mixtures one loopful is placed upon Loffler's plate. (Ox serum plate, should be three days old). The plates are then placed into the incubator for twenty-one hours at 55° C. The presence or absence of the clear zones determines the quantities of ferment which respectively have not or have been neutralized by the one drop of serum (e.g., i : 6 means that in the mixture of 6 loopfuls of trypsin and i loopful of the serum for examination the digestive power of the trypsin was still interfered with). The inequality in the strength of the Loffler plates, their variability in the degree of alkalinity, the measurement by loopfuls, all, might prove to be sources of error which may greatly influence the results. Thus the latter can only be taken as approximate, relative values. The second, more exact and satisfactory method was introduced pri- marily by Gross and Fuld for presenting the action of trypsin, and was modified by v. Bergmann together with Bamberg and Meyer for the deter- mination of an ti trypsin. Numerous workers have found it thoroughly reliable. Its principle is based on the digestion of a clear casein solution. If the entire amount of casein is digested, no more is left to be precipitated by the addition of acid and therefore the solution remains clear. If, how- ever, casein has been left undigested, the addition of acid will produce a turbid solution or even a white precipitate. The necessary reagents are: 1. Casein Solution. — One gm. of casein is dissolved under slight heating in 100 c.cm. of N/io NaOH; this solution is next neutralized by N/io HC1, litmus being used as indicator, and diluted with physiological salt solution up to 500 c.cm. (If sterilized, it can be kept for a long while.) 2. Trypsin Solution. — 0.5 gm. of trypsin (purissimum Grubler) is dissolved in 50 c.cm. of normal NaCL+o.o5 c.cm. of normal sodium hydrate solution and then diluted with physiological saline up to 500 c.cm. 3., Acid Solution. — Five c.cm. of acetic acid +45 c.cm. of alcohol +50 c.cm. of water. First the titration of the trypsin solution is undertaken in order to find out how much trypsin is required to fully digest a constant quantity of casein. Gradually increasing amounts of trypsin (from o.i to 0.6 c.cm.) are placed in six test-tubes and to each 2.0 c.cm. of casein are added. These tubes are placed in an incubator at 37° C. for one-half hour, and then several drops of the acid solution are placed into each tube. The first tube, and all those above it that remain absolutely clear, contain enough trypsin to fully digest the 2.0 c.cm. of casein. Now comes the second part of the test. In each of eight to ten test-tubes are placed 2 c.cm. of the casein solution and 0.5 104 THE TOXINS OF THE HIGHER PLANTS AND ANIMALS c.cm. of a 2 per cent, dilution of the serum for examination; to these is next added the trypsin solution in successively increasing amounts, beginning with the smallest quan- tity which in the first part of the test was sufficient to completely digest the given amount of casein. Salt solution is then added to each of the test-tubes so that all con- tain an equal quantity of fluid, and the mixtures placed in an incubator at 37° C. for one-half hour. At the end of this time, several drops of the acid are added to each tube. Those tubes which become cloudy or show a precipitate designate the amounts of trypsin solution which have been neutralized by the 0.5 c.cm. of diluted serum. For example: In the first part of the test it was found that the tube containing 0.4 c.cm. of trypsin was the first to remain clear, in other words was sufficient to fully digest 2 c.cm. of the casein solution. In the second part of the test the lower limit of the added trypsin dilution was 0.4, and it was found that the tubes containing 0.4, 0.5, 0.6, and 0.7 c.cm. of trypsin, for example, now gave precipitates and only 0.8 remained clear. This indicates that part of the formerly sufficient amount of trypsin was now neutralized by the antitrypsin of the added serum so that digestion was interfered with. Thus the antitrypsin titer in this case is 0.8. Recently the above method of trypsin titration has been applied to the determina- tion of the presence of pancreatic ferment in the intestinal secretions, feces, and stomach contents. CHAPTER X. AGGLUTINATION. BACTERIAL AGGLUTININS. HEMAGGLUTININS. TRANSFUSION TESTS. If the serum from an immunized animal, or a patient convalesc- The Phe- mg after an infection, be mixed with a suspension of the bac- o°A ^lu ter*a wkicn were involved m the production of said conditions, tination a Peculiar phenomenon takes place. In the former dif- fusely cloudy liquid, small granules and clumps appear which sink to the bottom of the test-tube and leave a supernatant clear fluid. On microscopic examination, the sediment presents bacteria, (which have re- mained alive as is demonstrable by making cultures of same) . This same observation can be made with perhaps more flattering results when the experiment is performed in a hanging drop. The bacteria are seen to lose their motility, adhere to each other, finally gravitate toward larger groups and arrange themselves in clumps. The phenomenon thus described was discovered by Gruber and Durham, and is called agglutination; while sub- stances which cause this, agglutinins. If instead of the immune serum or that of the convalescent patient, normal serum is employed and the above test repeated, it will be seen that agglutination likewise occurs. The reaction is, however, somewhat incom- plete ; the clumps are smaller, and formed much more slowly. // a quantita- tive determination with different dilutions of both sera is made, the power of agglutination disappears with the normal serum at a low dilution, while the immune serum remains perfectly active at even much greater dilutions. Thus the main difference between the agglutinating normal and immune serum is a quantitative one depending upon the amount of agglutinins present. Whether any qualitative difference exists between the normal and immune agglutinins is doubtful. It is, however, of no practical significance. If instead of homologous bacteria, different (heterologous) bacteria are employed, e.g., cholera vibrio and typhoid serum, agglutination also takes place, if the typhoid serum is used in concentrated or only slightly diluted form; but in moderate or great dilutions, no agglutination occurs. Nor- mal serum will agglutinate the cholera vibrio in the same strength as the immune typhoid serum. In other words the typhoid serum contains more agglutinins for its homologous bacteria than a normal serum, but it has only the same titer of agglutination as a normal serum for heterologous bacteria. 105 106 AGGLUTINATION The agglutination reaction is specific in the respect that high dilutions of serum will agglutinate only its homologous bacteria and leave the heterologous ones uninfluenced. Agglutination becomes non-specific, when concentrated or low dilutions of serum are employed. The relative specificity just described is of great clinical Diagnostic diagnostic value. For example, given a serum suspicious of Value of Ag- typhoid, the question is to establish this absolutely. One glutination. immediately proceeds to make a suitable dilution of the unknown serum and mixes with it known typhoid bacilli. A similar dilution of normal serum is made as a control and mixed with the same amount of typhoid bacilli. If agglutination occurs with the unknown serum and not with the control serum, the former must have come from a typhoid patient. If the bacteria are not agglutinated, the serum was not of typhoid origin. In an equal manner can the identity of unknown bacteria be estab- lished by the use of known sera. Thus, when certain bacteria have been isolated and information is wanted as to whether they are typhoid, an emulsion of these is made and mixed with a typhoid serum in suitable dilution, and a similar amount of bacteria is mixed with a normal serum of like dilution. Agglutination occurring in the first of these mixtures and not in the second proves the typhoid character of the unknown bacteria. In this manner the agglutination test can be used for identification of any antigen. The practical application of agglutination has been greatly used in cases of typhoid fever. Here agglutinins are very easily stimu- lated in the course of the disease and generally they can be demon- strated in the serum seven to ten days after infection. The agglutinins remain not only during the active stage of the disease, but also during the convalescing period. Widal, the Parisian clinician, was the first to adopt this agglutination reaction for the serum diagnosis of typhoid. It is thus commonly known as* the Widal reaction. Technique of The technique of the reaction is as simple as its principle. Aggluti- This accounts for its wide adoption. It may either be per- nation. formed macroscopically or microscopically (orientation test). The Macroscopic Agglutination Reaction. For this reaction it is necessary to have 1. The immune serum and a normal control serum; 2. A homogeneous bacterial emulsion. The production of a homogeneous bacterial emulsion offers slight technical difficulties. It can be obtained in the following ways: AGGLUTINATION REACTION 107 a. Bouillon Culture. — Many bacteria, like typhoid, paratyphoid, dysen- tery, coli, etc., grow very easily in broth. Such afresh (twenty-four hours), diffusely turbid culture can be employed readily for agglutination purposes. In place of live bacteria, dead may also be used — a fact which has greatly added to the practical application of the test. For preparing the latter, 0.5 per cent, of phenol or i per cent, of formalin (40 per cent.) are added to the twenty-four hour bouillon cultures. The result is, that a sediment of bacteria is formed from which the supernatant fluid should be carefully poured off. The bacterial suspension is kept on ice and thoroughly shaken before use. Picker has in this way prepared standard emulsions of dead typhoid and paratyphoid bacilli which are sold by Merck under the name of "Ficker's Diagnosticum." For Widal's test, a small quantity of the patient's blood is collected in a capillary tube and the end closed with sealing-wax. The blood is allowed to clot, and the serum to separate off. The separation of the latter can be hastened by centrifugalization. In practice, the Widal test as performed with Ficker's diagnosticum, is arranged as follows: Bacillus suspension Dilution of serum Physiological salt solution Tube 1050 cm o 5 c cm. Tube 2 o 5 c cm o ' 5 c cm of i ° 10 Tube 3, o. 5 c.cm. 0.5 c.cm. of i 150 Tube 4 o 5 c cm o 5 c cm of i • 100 One of four results may be obtained, i. 2. Positive reaction Doubtful reaction Negative reaction Worthless reaction i. No agglutination No agglutination No agglutination Agglutination. 2. Marked agglutination. Marked agglutination Very slight agglutina- Agglutination. tion. 3. Marked agglutination. Very slight agglutina- No agglutination Agglutination. tion. r •:••'•-•' -' :%'.•-•• :.';..' -;: '• .-,;. -_ - 4. Slight agglutination.. . No agglutination No agglutination Agglutination. It is very advisable to make control tests with normal serum. After mixture of the various ingredients the tubes are placed in the incubator at 37° C. for two hours. Then the results are read off; the first tube must show absolutely no agglutina- tion, otherwise (as seen in Division No. 4 above) the entire test is of no signifi- cance. The cause for such spontaneous or pseudo-agglutination occurring in tube i may be found either in the bacterial emulsion or NaCl solution. The grade of agglutination is estimated by the size of the agglutinated clumps and the rapidity with which they are formed. The mild grades of agglutination are frequently over- looked by the beginner. For typhoid a positive reaction is one where agglutination 108 AGGLUTINATION takes place in the dilution of i :ioo; a positive result in the dilution of i : 50 can only be considered as probably positive. As has been said, broth cultures may be used for the agglutination test if the bacteria grow diffusely and regularly within the bouillon. This is not the case, how- ever, with all bacteria, as for example, the cholera vibrio which produces a thin pellicle upon the surface of the broth. b. A gar Cultures. — Kolle and Pfeiffer have advised instead of broth the use of agar cultures. The bacteria are washed off, and an even emul- sion made in physiological salt solution, or in a dilution of the serum for examination. The details of the procedure are as follows: Into a row of test-tubes is placed i c.cm. of various dilutions of the serum for exami- nation, e.g., i : 10, i : 50, i : 100, i : 200, i : 500. A normal serum is similarly diluted as a control. One other test-tube is to contain i c.cm. of saline only. A full loop of an eighteen to twenty-four hours old agar culture is evenly and finely rubbed up in each of the above test-tubes as follows: The test-tube is held almost horizontally in the left hand between the thumb and index finger; a platinum loop between the thumb and index finger of the right hand is filled with the bacteria from the agar culture, and placed in the tube containing the serum dilutions. The bacteria are then gently and thoroughly rubbed up on the moistened wall of the tube but not within the fluid. By rolling the test-tube slightly, a part of the rubbed up bacteria is washed into the fluid and the remaining bacterial mass is again triturated. This process is repeated until all the bacteria are washed into the fluid. Thus, a homogeneous suspension is obtained. The author has found this method of Peiffer and Kolle most accurate. It is worthy of note in this connection, that the controls show no clumps or granules. (Pseudo-agglutination). There are some bacteria which can be evenly emulsified only with great difficulty, while others are very easily agglutinated even by normal serum. In either case the test is not conclusive. For the hanging-drop method, blood is collected in a Wright capsule or a small test- tube-, 6 to 8 drops of blood suffice. The blood is allowed to clot or the serum is hastened by centrifugalization. Four loopfuls of broth or saline (or equal amounts as measured by a Wright pipette) are placed on each of two slides. To one of these one loopful of the serum (or one equal part as measured by the Wright pipette) is added and thoroughly mixed. From this mixture one loopful or equal measure is mixed with the broth or normal saline upon the second slide; thus making serum dilutions of i 15 on the first slide and i : 25 on the second slide. A loopful of typhoid cul- ture is placed on the center of each of two cover slips. To the first is added one loopful of the serum dilution 1 125, and to the second is added one loopful of the serum dilution i : 5 thus making a dilution of i : 50 and PRODUCTION OF AGGLUTINATING SERA I Op i :io respectively. Each cover slip is inverted over a hollow slide pro- tected by vaseline, and examined microscopically. A control with normal serum and also one with culture alone should be made. For the identification of bacteria only highly agglutinating Production animal sera can be employed. Rabbits, goats and horses, of Aggluti- are most suitable for such experiments. The best results nating Sera. are obtained when the animals are immunized intravenously by repeated injections with gradually increasing doses of dead bacteria (killed at 60° C.). Usually two to three injections of 1/4 to i agar culture of bacteria suspended in saline solution suffice to give an agglutinat- ing titer of i to 5000. The serum should be withdrawn eight to ten days after the last injection. With typhoid bacteria one may attain a strong agglutinating serum in a rabbit by the intraperitoneal injection of i c.cm. of a twenty-four hours' live broth culture. This is to be repeated in 7 to 10 days. As a matter of course, the titer of the serum should be tested from time to time, because the height of the antibody curve can only reach a certain point. When a sufficient strength is obtained the animal is bled. It is not possible to produce equally strong agglutinating sera for all bacteria. The agglutinins belong to the class of the more resistant serum sub- stances. By the addition of one drop of pure carbolic acid they can be preserved on ice for a long time. Heating variously affects the different bacterial agglutinins. The agglutinins for pest and tubercle bacilli are destroyed at 56° C. while other bacteria are not influenced by even higher temperatures. The animal from which the agglutinating serum has been obtained also influences to a great degree the resistance toward heat. Thus the typhoid agglutinating serum derived from the horse is much more resistant than that obtained from the rabbit. The Microscopic (Orientation) Agglutination Test. This method is especially of use, when only small amounts of culture or serum are obtainable. Also, if agglutination is employed for the quick recognition of bacteria, as for example, when it is desirable to know whether a blue colony on a Conradi-Drigalski-agar plate is typhoid or not. In such a case a drop of the immune serum in the dilution of i : 50 or i : 100 is placed upon a cover-glass held with a Cornet's forceps, and a small part of the bacterial colony for identification carefully mixed with this serum. As controls, a mixture is made with salt solution and with normal serum. If agglutination occurs, small granules or clumps can read- ily be seen with the naked eye by holding up the cover-glass against the light. The control glasses on the other hand should show only a homo- geneous turbidity. These changes are still more evident if the mixture is examined microscopically in the form of a hanging drop. (Described, p. 108.) 110 AGGLUTINATION Group Agglutination. On testing the titer of a strongly agglutinating typhoid serum, and a strongly agglutinating cholera serum, against typhoid, paratyphoid, colon and cholera bacteria, the results will be the following : Agglutination titer Of typhoid serum Of cholera serum Against typhoid . . . i : 2000 i : 10 Against paratyphoid i : zoo i : 10 Against bacter. coli Against cholera . . .... i 125 i : 10 i : 10 i : 3000 The cholera serum acts strictly in accordance with the rules stated above for specific agglutinins, i.e., marked agglutination with homologous bacteria; very weak, with heterologous. The typhoid serum on the other hand, although it fulfills the same requirements in the main, nevertheless it manifests some important differences when mixed with heterologous bacteria. It has practically no influence upon the cholera vibrio with which the typhoid bacillus is not at all related; agglutination of i :io can be attained even by a normal serum. The colon bacillus which closely re- sembles the typhoid, morphologically, but which has very different bio- chemical properties, is more strongly agglutinated, i 125; while the para- typhoid bacillus, very much like the typhoid bacillus both morphologically and biologically, is agglutinated even in larger dilations, i : 100. This entire phenomenon is an expression of the biological relationship of the various bacterial groups and is known as group reactions. An understanding of group reactions is to be found in a more Partial Ag- complete conception of specificity. From this source we have glutination. learned that the difference in antibodies is influenced by the dissimilarity of the injected antigen. For example, the differ- ence between the cholera and typhoid agglutination is caused by the differ- ence existing in the protoplasmic structure of the respective bacteria. As these bacteria, however, are not constituted of a distinct chemically de- fined substance, but made up of a mixture of various substances, there may be a number among them which can act as antigens. If, figuratively speaking, there are five different elements in the body of the typhoid bacillus which can act as agglutinogens, i.e., antigens, these should be able to form according to the law of specificity five different agglutinins. On mixing a typhoid serum with typhoid bacilli, one brings together five distinct antigen antibody combinations and consequently complete and thorough action results of this union. A biological relationship of bacteria implies the exist- CASTELLANIS TEST FOR MIXED INFECTIONS III ence of some common protoplasmic constituents. Expressed in the same figurative manner the colon bacillus can be said to have antigen number i in common with the typhoid and paratyphoid bacillus and the paratyphoid may have antigen numbers i and 2 in common with the typhoid bacillus. As a result, the typhoid serum will react with colon bacilli by virtue of their common agglutinin number i, and with paratyphoid bacilli through its agglutinins numbers i and 2. The other "partial agglutinins" remain inactive on account of the missing suitable agglutinogens. The existence of such partial antigens and partial antibodies is for some bacteria more than of mere theoretical importance. It is even possible that a strong colon serum will agglutinate no colon bacilli other than that par- ticular strain employed for the production of the serum. Such being the case with a number of micro-organisms, the sera made at present both for diagnostic and therapeutic purposes are polyvalent (multipartial) . By polyvalent serum is meant one which is produced either by immunizing animals with many different strains of the same bacterium, or a mixture of sera obtained from different animals immunized with various strains. The practical importance of partial agglutinins is recognized Castellani's in the diagnosis of mixed infections. Castellani found that by Test. the mixture of an immune serum with its corresponding bac- teria, the agglutinins for these as well as the partial agglu- tinins for the heterologous bacteria are absorbed. On the other hand, if the same serum be mixed with the heterologous bacteria, the agglutinins for the homologous group are quantitatively retained. A practical example will make this clearer. The serum of a patient agglutinates typhoid as well as paratyphoid bacilli, in a dilution of i : 100. This may indicate one of three possibilities: a. Patient is infected with typhoid, but has formed an exceptionally large number of partial-agglutinins for paratyphoid bacilli. b. Patient is infected with paratyphoid bacilli, but has formed at the same time many partial-agglutinins for typhoid. c. Patient has a mixed infection of typhoid and paratyphoid and therefore formed agglutinins for both. A decision in regard to the above may be reached according to the following method given by Castellani: Four rows of test-tubes are arranged, each row containing three tubes with i c.cm. of serum dilutions i : 10, i : 50, i : 100 respectively. In each of the first and second rows, i loopful of typhoid bacteria is emulsified. In each of the third and fourth rows, i loopful of paratyphoid B. bacilli is emulsified. The tubes are placed in the incubator for two hours, absence or presence of agglu- tination in each test-tube noted, and after centrifugalization (which may become unnecessary if the bacteria are strongly clumped or grouped at the bottom of the tube), the supernatant liquid is transferred to other test-tubes and kept in the same order. Then each of the first row receives i loopful of typhoid bacilli, each of the second row receives i loopful of paratyphoid B. bacilli, 112 AGGLUTINATION each of the third row receives i loopful of typhoid bacilli, each of the fourth row receives i loopful of paratyphoid B. bacilli, All are once more placed in the incubator for two hours. a. If typhoid exists the agglutination titer in the second part of the test will become weaker for the typhoid bacilli in the first row, and weaker for the paratyphoid B. bacilli in the second and fourth rows. The titer in the third row remains the same. b. If paratyphoid exists, the agglutination titer for typhoid in the first and third row becomes less, that for paratyphoid in the fourth row diminishes, while the titer in the second row for paratyphoid remains the same. c. If a mixed infection exists, the agglutination titer in the first and fourth row diminishes and in the second and third row remains the same. In this connection a few exceptions may be mentioned: A serum which is kept for a long time frequently loses part Agglutinoids. or even all of its agglutinating titer. Whereas it formerly agglutinated in the strength of 1:1000, it may now become inactive in dilutions even of i : 10. The first thought that arises in explana- tion of this is that the serum has perhaps degenerated and the agglutinins were destroyed. If, however, further dilutions are made, i : 100 may show mild, while i : 500 strong agglutination. This, first of all, demonstrates that agglutinins are still present, although diminished in amount, and sec- ond, that another substance has arisen which in the stronger concentrations interferes with agglutination. A simple experiment explains this. If the test-tube containing the serum dilution i : 10 and the non-agglutinated bacteria be centrifugalized, the serum removed and the bacteria mixed with a known strongly agglutinating serum, it will be found that the bacteria have become inagglu- tinable. Substances of certain kinds have combined with the bacteria and prevented them from undergoing agglutination. These substances are strongly specific, acting only upon homologous bacteria. Their origin can also be demonstrated. An agglutinating serum which is heated to 65° or 70° C. loses its agglutinating power but the substance interfering with the subsequent agglutination has remained. Ehrlich explains the situation as follows: He claims that agglutinins are built complexly; that they possess a binding (haptophore) group by means of which they unite with the bac- teria (agglutinogen) and a second group (ergophore or agglutinophore) by virtue of which agglutination results. If serum is kept for a long period of time, or exposed to high temperature, many of the ergophore groups are rendered inactive, while the haptophore groups being more resistant remain and unite with bacteria. Agglutinins possessing only their haptophore groups are known as agglutinoids. They combine with the bacteria, and still do not agglutinate them, but at the same time prevent other agglutinins from acting. If this old agglutinoid and agglutinin containing serum is diluted, so few of both of these substances remain that the bacteria can absorb both, allowing the relatively few agglutinins to manifest their activity. It is important to note in this respect that occasionally even a fresh, highly valent serum will present a tendency toward interfering with the agglutination processes. This is also explained by the existence of agglutinoids — a fact as yet not definitely proven. Another finding, only encountered in exceptional cases, is me existence of the so-called non-agglutinable strains of bacteria. These give all the AGGLUTININS IN TYPHOID FEVER 113 characteristics of the general class of bacteria to which they belong, but are not agglutinated by their respective serum; as, for example, a strain of typhoid bacilli, which are not agglutinated by any typhoid serum. The only positive proof that they are typhoid bacilli is the ability to produce by their injection into animals an active immunity against fully virulent typhoid bacteria. Non-agglutinable strains of bacteria can be isolated especially from the lower ani- mals. At times, however, they regain their agglutination property when they are grown in artificial media and frequently subplanted. Possibly, the reason that the bacteria become inagglutinable at all, is that they undergo immunization within the organism against the existing agglutinins. By growing bacteria in agglutinating serum for a certain time, one can obtain non-agglutinable strains. i. Agglutinins for typhoid and paratyphoid A and B, can, not Agglutina- infrequently, be demonstrated in the patient's serum as early T^h kl as t^ie ^i1^ ^ay> but as a rule, at about the beginning of the and Para- second week of the disease. Moreover, they remain within typhoid, the serum for several weeks after the illness and disappear only gradually. A positive agglutination test does not, how- ever, mean the existence of the corresponding disease. A healthy bacillus carrier can also have an agglutinating serum. Some cases of icterus catar- rhalis even give a positive Widal test. But in order to assign to this last a correct explanation, one must remember that typhoid bacilli may remain in the gall-bladder for years and thus lead to catarrhal inflammation and stone formation. Partial agglutinins from coli infections must always be considered. Some authorities mention a positive Widal, in connection with endocarditis maligna, sepsis, malaria, phthisis, and miliary tuberculosis. An absence of the agglutination test, especially at the early part of an illness, should not influence a negative diagnosis of typhoid too greatly, inasmuch as many cases are known where the reaction appeared for the first time during the period of convalescence. In the employment of this test as an aid for the differential diagnosis between several bacterial in- fections, it is best to titrate the serum to its limit, as the higher titer for one class of bacteria generally speaks in favor of the infection by the same. Paratyphoid serum agglutinates typhoid bacilli only slightly, while true typhoid agglutinates both typhoid and paratyphoid bacteria with almost equal strength. In severe and difficult cases, Castellani's test should be performed. Paratyphoid B . serum always gives the limit of its agglutinat- ing titer both with the pathogenic mouse typhoid and hog cholera bacillus. 2. Cholera. — Only rarely has the agglutination test been employed with the serum of patients infected with cholera. On the other hand, if cultures from a cholera stool are made upon a plate, the identification of the suspi- cious colonies grown here is regularly conducted by means of this test. For 8 1 14 AGGLUTINATION this purpose, it is very specific, as group reactions almost never take place. Strong agglutinating sera are easily obtained by immunization of animals. 3. Epidemic, Cerebrospinal Meningitis. — Agglutination in this disease serves mainly for the identification of suspicious meningococcus cultures. As has been shown by Wassermann and Kutscher, some strains are agglu- tinated only after a long period (twenty-four hours) and at higher tempera- tures as 56° C.1 4. Dysentery. — The agglutination property is employed both for testing the serum, and identifying cultures. The Flexner type of bacillus pro- duces agglutinins more readily than the Shiga-Kruse. It is also- agglutinated more readily. Only positive reactions in dilutions of i : 30 are of diagnostic consideration. Occasionally, partial agglutination takes place with heterologous dysentery strains, typhoid and colon bacteria. 5. Pest. — The reaction is very specific, but of slight significance, as it appears only on the ninth day; occurring with a serum dilution of 1:3, it is considered of positive diagnostic value. 6. Malta Fever. — In most instances the serum gives the agglutination reaction with the micrococcus melitensis. Normal serum may give the reaction in dilution i : 30, so that higher dilutions only are of aid in diagnosis. 7. Staphylo-, Strepto- and Pneumococci. — Clinically, the agglutination test is never employed in these cases. 8. Tuberculosis. — Here the agglutination test is associated with the difficulty of obtaining a homogeneous tubercle bacillus suspension. This, however, is overcome in one of two ways. a. Arloing-Courmont's Method (1898). — The tubercle bacilli are obtained in the so- called "homogeneous culture" form. S. Arloing first grows the bacteria on potatoes for a long time, and then transplants them in glycerin bouillon which is shaken daily for five minutes. After a number of subcultivations, a culture is obtained after several months. This strain grows rapidly in a few days and diffusely clouds the broth. Such a culture diluted with physiological saline solution is used for the test. Here small test-tubes are preferable and the ingredients should be mixed in the following proportions: 2 drops of serum -f 10 drops of culture (i : 5) i drop of serum -f-io drops of culture (i: 10) i drop of serum +15 drops of culture (i: 15), etc. The tubes are well shaken and placed in the incubator. According to Arloing and Courmont, a positive reaction even in the dilution of i : 5 speaks for tuberculosis. Best results are by this means obtained in incipient and mild tubercular cases; those which are farther advanced do not react. b. Method of Koch. — Koch niters the ordinary tubercle bacillus bouillon cultures, dries the remnants upon the filter, and rubs them up in an agate mortar with N/5o NaOH to a dilution of 1:100. The solution is centrifugalized and enough weak HC1 is added until the reaction is only slightly alkaline. The dilution is then brought 1 Frequently, during even the first days of the disease, the patient's serum in a dilution of i- 10 gives the agglutination test. This is rare with higher dilutions of the serum as 1-50. It usually takes some time before the agglutination becomes evident. \ HEMAGGLUTININS 115 up to 1:3000 by the addition of 0.5 per cent, phenol in normal saline, and kept for twenty-four hours in the incubator. A somewhat simpler procedure is to dilute new tuberculin B. E. to i: 100 with 0.5 per cent, of carbolic saline solution, centrifugalize this for six minutes and then dilute to i : 1000. The solution thus obtained can be preserved in the ice-box for fourteen days. Just before using, a still further dilution of i : 10 is made. The agglutination test has not been generally adopted as a method of diagnosis. The technique is rather difficult, and the results not absolutely reliable. The reason for this is that high agglutination values are rarely met with, and slight ones are found even in normal individuals. Then, too, the methods of tuberculin diagnosis are so much simpler that they have been given the preference. Koch himself advised the agglutination test, not as a means of diagnosis, but rather as an aid in tuberculin therapy. He found that during the treatment of tuber- culosis with new tuberculin the agglutinative power of the patient's serum increased. He therefore took this as an index of the acquired immunity. Further study, however, convinced him that the agglutination cannot thus be interpreted, so that at the present day tuberculosis agglutination has no practical application. 10. Glanders. — Highly valent sera can be obtained, according to Kleine, by intra- venous immunization of donkeys and goats. The serum serves for identification of the glanders bacilli. Kleine prepares a standard bacterial emulsion in the following manner : Four well grown glanders cultures are killed at 60° C. and the mass of bacteria tritu- rated in 2 c.cm. of 1/2 per cent, carbolic-saline solution. This is then diluted in a measuring glass so that 40 to 50 c.cm. of carbolic-saline solution are added for each culture. The entire mixture is filtered through paper and 3 c.cm. are used in each test. Normally, horses may have an agglutination titer up to 1 1400. Glanders in- fected animals react as high as i : 2000. Injections of mallein increase the agglutina- tion titer. Experiences in this respect with the human being are still scanty. Hemagglu- Just as injections of bacteria produce bacterial agglutinins, tinins, injections of erythrocytes stimulate the formation of hem- Immune and agglutinins which cause the red blood cells to congregate in Normal. clumps. At times the presence of hemagglutinins is masked by the simultaneous existence of hemolysins which dissolve the red blood corpus- cles. If, however, the immune serum is heated to 56*° C. the complement is destroyed, thus interfering with the action of the hemolysin and allowing the agglutinins to exhibit their action. In other instances as during the immunization of rabbits with dog's erythrocytes, hemagglu- tinins are formed in such great quantities that by mixing the immune rabbit's serum with the dog's erythrocytes so strong an agglutination occurs that the hemolysins can no longer attack the clumped erythrocytes. The hemolysis can be demonstrated only if clumping is prevented mechanically, by thorough shaking of the mixture. Also under normal circumstances, i. e., without immunization, the serum of one animal species can to a certain degree agglutinate (and hemolyze) the red blood cells of another species. The quantity of these normal hemagglutinins is comparatively small. 1 1 6 AGGLUTINATION Still more interesting is the observation that there are hem- Isohemag- agglutinins against the red cells of different animals even of glutinins. the same species, so-called Isohemagglutinins or Isoagglutinins. These have thus far been demonstrated in the bloods of dogs (Von Dungern) steers and rabbits (Ottenberg and Friedman). Isoagglut- inins in the human serum were discovered independently by Landsteiner and Shattock in 1900. At first the occurrence of isoagglutination was regarded of pathological significance, but soon it was shown that the phenomenon occurred with a large percentage of normal bloods. In fact all human bloods can be divided into four sharply defined groups according to the way in which they interagglutinate. The groupings can be explained by assuming the existence of two agglutinins of which the first group possessed both, the second one, the third one, and the fourth neither. In each case the cells are susceptible only to that agglutinin which does not exist in the individual's own serum. Thus: The serum of the first group, designated as group I, possesses the power of agglutinating the red cells of members of all the other groups but the red cells of members of group I are not agglutinated by any human serum. This group includes about 50 per cent, of all persons examined. The serum of the members of the second group agglutinates the red cells of members of the third and fourth groups. The cells of members of the second group are agglutinated by sera of individuals of groups I and III. The serum of group III agglutinates cells of persons belonging to members of the second and fourth groups; its cells are agglutinated by sera of the first and second groups. The fourth group, whose members are relatively rare, is characterized by possessing no agglutinin for human red cells and by its cells being agglutinable by the sera of all other groups. The group characteristics are permanent for each individual through- out his life. When concentrated, the agglutinins act almost instantane- ously; when diluted they act more slowly. Agglutination occurs in the cold as well as at high temperatures. The peculiar groupings are not only permanent with the individual but they are hereditary. Von Dun- gern and Hirschfeld have conclusively proved that agglutinins are hereditary and follow the Mendelian law. This observation had, however, been made long before this in a paper by Epstein and Otten- berg (1908). With the recently increasing popularity of blood transfusions, the phenomena of isoagglutination and hemolysis, the two being very closely related, have attained a more practical significance. In selecting donors for a transfusion, agglutination and hemolysis tests should always, when time permits, be made before operation. These tests in vitro are usually a safe guide as to conditions in vivo. That donor should be chosen who TRANSFUSION BLOOD TESTS 1 17 belongs to the same group as the patient, that is where no interagglutina- tion or hemolysis exists. This is advisable not only so as to get the best results from the^ transfusion, but also in order to avoid any untoward symptoms or intoxications that are associated with the intravascular agglutination or hemolysis (rise of temperature, dyspnea, edema, hemo- globinuria) . The more important of these two factors is not as yet clear. If for a given transfusion a donor belonging to the same class cannot be obtained, it is safer to use a person whose serum is agglutinative toward the patient's cells than one whose cells are agglutinated by the patient's serum. The materials necessary for the agglutination and hemolysis Transfusion tests are as follows: Tests. (i) Sterile syringes or needles for puncturing the vein; (2) i per cent, sodium citrate solution in 0.85 per cent, salt solu- tion; (3) 0.85 per cent, salt solution; (4) a test-tube rack having two narrow test-tubes (4X1/2 in.) for each donor (numbered); one partly filled with the sodium citrate solution, one empty. From each donor i to 2 c.cm. of blood are aspirated from a vein of the elbow; several drops of this blood are allowed to flow into his tube with the sodium citrate solution, the rest is collected in his dry test-tube. This is also done to the recipient, but from him 3 to 4 c.cm. of blood are necessary so as to have sufficient serum for a number of donors (10-15). The sodium citrate tubes are centrifugalized and a sediment of the red blood cells obtained; the supernatant fluid is pipetted off and the red cells made up approximately to a 10 per cent, suspension with the normal salt solution. The serum tube is also centrifugalized so that clear serum is separated off. The clot should not be disturbed too energetically as it is best to get absolutely clear yellowish serum not blood tinged. The following mixtures are then made with each donor's blood, pref- erably within 12-24 hours of the time of collecting the blood: (a) 3 parts or units of donor's serum and i part or unit cf recipient's red cell emulsion. (b) 3 parts or units of recipient's serum and i part or unit of donor's red cell emulsion. Controls: (c) 3 parts or units of donor's serum and i part or unit of donor's red cell emulsion. (d) 3 parts or units of recipient's serum, i part or unit of recipient's cell emulsion. (e) 3 parts of saline, i part of donor's red cells. (/) 3 parts of saline, i part of recipient's red cells. These mixtures are made in very small test-tubes (3X3/8 inch). The. Il8 , .AGGLUTINATION quantity comprised by each "part" or "unit" varies according to the amount of blood or serum obtained from each donor and recipient. If sufficient, it is best to work in drops (each drop being about 0.05 c.cm.), thus mixing 3 drops of serum (0.15 c.cm.) and i drop of the red-cell emulsion (0.05 c.cm.). If the serum is not sufficient, a smaller arbitrary unit may be used in the form of Wright's pipettes (4 to 5 millimeters caliber) fitted with rubber nipples, drawn out to a length of 2 to 3 inches, and marked off about i inch from the tip with a blue pencil, this distance comprising the unit. The tubes are placed in the incubator for three hours and then in the ice box for 24 hours. Agglutination when it occurs does so rather promptly, within 15 to 30 minutes after the mixtures have been made. It is recognized macroscopically by the clumping of the red blood cells into small floccules which later on appears like a distinct clot. A hanging drop preparation of such a mixture shows the same phenomenon; usually this is unnecessary as the macroscopical appearance is characteristic. If one is in doubt, however, microscopical examination should be made. Hemol- ysis, if pronounced, is observed in an hour or even less, but certainly after the three hours' incubation; the finer grades of hemolysis are detected after the tubes have remained in the ice box 12 to 24 hours. The control tubes must show no agglutination or hemolysis. In cases where for any reason the taking of blood from the vein is not allowed or impossible (as in infants), Wright's method of working with small quantities (Epstein and Ottenberg) should be resorted to. The skin at the bed of the finger nail or of the lobe of the ear is pricked deeply with a Hagedorn needle. For the red blood-cell suspension several drops of blood are allowed to flow or taken up with a dropper and expelled into a tube with sodium citrate solution. The red blood cells are washed and diluted as before. For the serum, three or four Wright's capsules (see under opsonins) are filled with blood which is allowed to clot, and centrifugalized. Each cap- sule is nicked with a file, allowing the capsule to be broken open and the serum pipetted off. For making the mixtures Wright's pipettes (four to five millimeters caliber) fitted with rubber nipples are used. With a blue pencil an arbi- trary point is marked off upon the drawn out extremity of the pipette. Three volumes of serum (a bubble of air between each volume), and one volume of cell suspension are drawn into the same pipette. These ingre- dients are mixed by running them gently out and then drawing up again into the pipette. The entire mixture is then drawn into the body of the pipette and the tip is sealed in a flame. Each pipette serves as a little tube. The time and manner of incubation, and the observation for hemolysis follow the rules mentioned above. TRANSFUSION BLOOD TESTS 1 19 This method is especially valuable when a very great number of donors are to be examined. If, however, a sufficient amount of serum is obtain- able the editor prefers the first method as it is a good deal simpler; then, too, the total amount in each test is the same, thus allowing better of comparison as to the intensity of hemolysis. CHAPTER XI. PRECIPITINS. In the former chapter, the phenomenon of agglutination was explained as a clumping of bacteria occurring when serum is mixed with its corre- sponding bacteria. In 1897 R- Kraus described a phenomenon, very closely allied to the one just mentioned. He found that when an immune serum, for example, of cholera, typhoid, or pest, is mixed with the clear, sterile nitrate of the respective bouillon cultures of their bacteria (instead of the bacteria themselves), the clear solution becomes turbid, and a precipitate forms. This reaction is known as precipitation; the elements within the immune serum, precipitins; while the substances (antigen) with which the precipitin reacts and which originally stimulated the pro- duction of the precipitin, precipitinogen. Like all biological reactions, the phenomenon of precipitation is not limited to bacterial immune sera and culture nitrates, but is observed when any animal, vegetable or bacterial soluble proteid substance is mixed with the serum of an animal which has been immunized against the particular proteid material in question. Tschistowitsch and Bordet were the first who called attention to these non-bacterial precipitins. Bordet (1899) found that the blood serum of rabbits treated with the serum of chickens gave a specific precipitate when mixed with chicken serum. Tschi- stowitsch demonstrated a similar reaction with the sera of rabbits treated with horse's and eel serum. The biological structure of the precipitins is strongly analogous to that of agglutinins. Many authorities, in fact, consider them identical. What- ever has been said in regard to the effects of heating and addition of acids or alkalies upon agglutinins, applies equally to precipitins. Moreover, they also are composed of two groups, a binding (haptophore) and a functionally active (ergophore) group. If the latter is missing, they are known as precipitinoids, and can interfere with precipitation just as ag- glutinoids do with agglutination. In speaking of precipitation, it has always been customary to differen- tiate between bacterial and proteid. For practical purposes this division is superfluous inasmuch as the bacterial precipitins are nothing more than precipitins of bacterial proteids. 120 PRECIPITIN REACTION 121 Bacterial Precipitin Reactions. For the production of precipitating sera, animals, preferably Production of rabbits, are injected either with broth cultures or salt solution Bacterial emulsions of agar cultures of the bacteria. Five to six in- Precipitat- jections of gradually increasing quantities are given intra- ing Antisera. peritoneally or intravenously, at intervals of from five to six days. The dose varies with the virulence of the bacteria. The intravenous method frequently giyes the stronger serum but the mode of administration depends also on the pathogenic properties of the micro- organism in question. The immunized animals should be bled about seven to twelve days after the last injection of bacteria. The filtrates of bouillon cultures and the various forms of bacterial extracts will also, when injected, produce precipitins. The serum from individuals undergoing an infection, or convalescing from one, contains precipitating bodies against the respective infective agent. Inasmuch as the precipitin reaction consists in the formation of a pre- cipitate, it is important that both of the ingredients (precipitin and precipi- tinogen) be absolutely clear and have no tendency to spontaneously become turbid, or form a precipitate. In order to get a clear serum one should avoid withdrawing Obtaining the blood during the period of digestion of the animal, because Clear Sera, it is chylous at such a time. In man the best occasion for obtaining the blood is in the morning before breakfast. As for animals, it is advisable to give them no solid food (or milk) for twenty- four hours previous to venesection. Then a very minute quantity of blood is withdrawn and immediately centrifugalized in order to ascertain whether the serum is clear or not. If it is satisfactory, larger amounts may be collected. The presence of erythrocytes and bacteria causes a serum to be turbid. Simple sedimentation or centrifugalization suffices to overcome this. If in spite of these precautions turbidity still persists, recourse may be had to filtration through paper or bacterial filters, preferably new ones. This method should, however, be used as a last resort, because filtration always tends to diminish the strength of a serum. Bacterial precipitinogens are prepared by filtration either of Bacterial bouillon cultures or bacterial extracts. The filtrates must be Precipitin- absolutely clear; also sterile, as frequently the precipitin ogens. reaction requires a long period of time. If bacteria are present they may grow quickly, and produce turbidity. After a time the precipitinogen loses its property of combining with precipitins and forming precipitates. In such a case the precipitinogen can still be employed for immunization purposes. 122 PRECIP1TINS A constant amount of precipitinogen is placed in each of a row Technique of °f test- tubes, and to these are added diminishing amounts of the Precipi- the immune serum, tin Reaction. A set quantity of serum and varying amounts of precipitinogen can also be employed. The result of the reaction depends to a very large extent upon the quantitative relationship of these in- gredients. // relatively too much precipitinogen exists, a precipitate will not form. An already formed precipitate will dissolve on the addition of more precipitinogen. The explanation of this peculiarity is unknown. Since colloidal sub- stances, however, at times give similar reactions, many authorities have classed the precipitins among them. The technique of a precipitation test is best seen in the following table: Cholera bouillon filtrate Cholera serum Physiological saline sol. Result . After 4 hours After 24 hours 5 . o c.cm. 5.oc.cm. 5 . o c cm. 5.0 c.cm. 5.0 c.cm. i.o c.cm. 0.5 c.cm. o. i c.cm. 0.05 c.cm. Very cloudy. Cloudy. Faint cloud. Clear. Clear. Clear. Clear. Clear. Clear. Clear; marked sediment at bottom. Clear with moderate sediment at bottom. Clear; slight sediment. Clear; no sediment. Clear; no sediment. Clear; no sediment. Clear; no sediment. Clear; no sediment. Clear; no sediment. 0.5 c.cm. 0.9 c.cm. 0.95 c.cm. i.oo c.cm. 5.0 c.cm. 0.5 c.cm. 5.9 c.cm. 5- 95 c.cm. i.o c.cm. 0.5 c.cm. o.i c.cm. 0.05 c.cm. A parallel row of tubes with normal serum should be included. If highly valent sera, such as are obtained by immunization with bacterial extracts, are employed, precipitation may result soon after mixing the two constituents. The precipitins are strongly specific, al- though it may be said that just as in agglutination, there exists in precipita- tion a certain degree of "group reactions." The precipitation test has no clinical diagnostic value. It Diagnostic demonstrates nothing more than the agglutination test, is Value of more difficult of execution and associated with greater sources Bacterial of error. Only occasionally is it of service to prove the Precipitation, presence of soluble bacterial substances within exudates or organ fluids. Porges and v. Eisler have employed the precipitation test as a means for the differ- entiation of capsule-bacteria where the method of agglutination is associated with certain difficulties. The precipitinogen was produced by filtration of four- weeks-old PRECIPITATION TESTS IN TYPHOID FEVER 123 bouillon cultures of pneumococci, rhinoscleroma, and ozoena bacilli. The immune serum was obtained from rabbits which had received four to five subcutaneous injec- tions of the respective bacterial suspensions. Fornet has recently advocated the precipitation test as an aid in the clinical diagnosis of typhoid fever. Although his attempts have not been attended with practical success, the principles of the reaction de- serve discussion on account of their originality. Fornet believed that it should be possible to demonstrate in the blood of typhoid patients the presence of the antigens (precipitinogens) which stimulate the antibodies, long before the latter themselves become evident. He actually was able to obtain turbid mixtures when he combined precipitating typhoid serum with the serum of typhoid patients. In many cases he obtained these results before the appearance of the Gruber-Widal reaction. The method which he has recently employed is known as the "ring test." Small test-tubes 8 cm. in height and 0.5 cm. wide, are placed in rows of Fornet's twenty each in a small black test-tube rack so arranged by the help of Ring Test, side stands that the tubes are inclined at an angle of about 45°. Across the back of the rack is attached a strip of dark cloth as a background to facilitate the detection of any precipitate. The immune (or convalescent) serum is placed in tubes in concentrated and diluted form 1:5 and 1:10 with normal saline, and then the serum for examination in concentrated and similar dilutions is carefully floated on top of the immune serum. The mixtures are allowed to stand undisturbed at room temperature for two hours, and if the reaction is positive a whitish ring at the point of contact of the two sera makes its appearance. A control test-tube of normal serum plus immune, and another of normal plus the unknown serum in the same dilutions as those employed in the test, must remain negative. Besides in typhoid fever, the ring test is also evident in scarlet fever, measles and syphilis. For the precipitation test in syphilis, the serum from patients Syphilis with manifest luetic symptoms is employed as precipitinogen, Precipitation, and the serum from individuals with general paresis acts as precipitating agent. The ring test must be carried out strictly in accordance with the rules given by Fornet, but even so, its diagnostic value for syphilis is still doubtful. Plaut claims that normal serum gives the reaction just as often as luetic serum; this is strongly denied by Fornet. Theoretically, it is questionable whether these precipitates and rings are similar in origin to bacterial precipitates, or whether physical-chemical causes are at the bottom of the former phenomenon. In accordance with the latter view several other reactions have been recently recommended for the serum diagnosis of syphilis. a. Forges and Meier noticed that luetic sera are capable of producing Forges' flocculent precipitates from lecithin solutions. Forges soon found the Reaction, same occurrence with solutions of bile salts. Many additions and modifications have been made to the Forges' 124 PREC1PITINS reaction since it was first recommended. According to the most recent publication, the reaction is carried out as follows: The requirements are: 1. One per cent, solution of sodium glycocholate (Merck) in distilled water. 2. The patient's serum which must be absolutely clear, and heated for one-half an hour at 56° C. Two-tenths of each of the above are placed into a narrow test-tube 6 to 7 mm. in diameter, and allowed to rest for sixteen to twenty hours at room temperature. A positive reaction consists of the appearance of distinct coarse flocculi, which as a rule, collect near the surface. Mere turbidity or faint precipitates are considered as negative. The original Forges method of employing lecithin was not at all specific, the reac- tion being present in tuberculosis, carcinoma, and other infectious diseases. As for the new modifications, nothing has been brought forward in their support. This reaction belongs to the same general class of precipita- Klausner's tion tests for lues, but is very much simpler than any of the Reaction, others. Two- tenths c.cm. of absolutely clear, fresh (at the most, two hours old), active serum is mixed with 0.6 c.cm. of distilled water, in a small test-tube 7X0.5 cm. Sera containing hemo- globin or lipoids are not suitable for this reaction. The mixtures are al- lowed to stand at room temperature. In several hours, at the latest fifteen, a thick flocculent precipitate 2 to 4 mm. high appears at the bottom of the tube. Kreibich's analysis showed it to consist of fibrin globulin. Apparently this substance is increased in luetic serum and precipitated by the dis- tilled water in which it is insoluble. Klausner's reaction is by no means specific for syphilis as it is in evidence in starvation, typhoid fever, measles, scarlet fever, pneu- monia, and other diseases, as well as during health. Nevertheless it must be said that it is found more frequently, earlier and much stronger in lues than in any other condition. Klausner states that in fresh cases of lues the best reaction is seen in about seven to nine hours, while in older cases a weak reaction appears in twelve hours. Mercury influences the test in that the interval until the precipitate becomes marked, is pro- longed and later on the precipitate becomes fainter. In spite of its simplicity, Klausner's reaction has not been generally adopted for clinical work, as the Wassermann reaction with its far greater accuracy has re- placed it. Proteid Precipitins. While bacterial precipitation is interesting from a biological standpoint but bears no practical significance, proteid precipitation represents one of the most important practical aids in forensic medicine. By this means the differentiation of various proteids can be easily and definitely determined, a problem which was left unsolved by chemistry. The phenomenon of protein precipitation is absolutely analogous to PRINCIPLES OF PROTEID PRECIPITATION 125 that of bacterial precipitation. If a clear proteid solution (a) is mixed with the clear serum (a') of an animal immunized against the above proteid (a), turbidity and precipitation will occur; while if a mixture of the serum (a') is made with a non-homologous proteid say (b), or a mixture of the proteid (a) with the serum (b') of an animal immunized against b, no precipitation takes place. Graphically expressed it looks thus: — a-\-a' = precipitation. b-\-af = no precipitation. a-\-bf = no precipitation. b + b' = precipitation. In other words, a precipitating immune serum reacts only with its homolo- gous proteid. The precipitin reaction is specific. It is greatly to the credit of Wassermann and his co-workers Forensic Use ^' Schiitze and Uhlenhuth, who recognized that this speci- of Albumin n°ity of precipitins was of great medico-legal value. Differentia- tion. For example, a bloody shirt is found in the home of a man charged with murder; the prosecution sees in that the proof of crime, while the defend- ant pleads that the stains belong to the blood of a sheep; the proof as to their source is of the utmost deciding evidence; and while chemical or microscopical examinations are of little or no use, serum diagnosis wins the day. The blood-stained clothing is extracted in water, part of the extract is mixed with a, the serum of a rabbit immunized against human serum and another part is mixed with b, the serum of a rabbit immunized against sheep's serum. If the mixture a shows a precipitate, it can be definitely stated that the blood stain contained serum derived from a human being; while if mixture a is clear and b shows the precipitate, it is strongly corroborative of the presence of sheep's serum. This example suffices to indicate the value of this biological Blood fact. In addition the reaction is made use of in the deter- Relationship, mination of the nature of meats (detection of horse meat sub- stitution for beef). Furthermore, this method has explained a number of scientifi- cally interesting problems. Just as group agglutination demonstrated the close relationship existing between various bacteria, so also serum precipi- tation proves a distinct relationship between the different species of animals (horse and donkey, dog and fox, hare and rabbit, ape and man, etc.). Thus the serum of a rabbit immunized against human serum precipitates not only human serum but also that of monkeys; the serum of a chicken immunized against rabbit's serum precipitates not only that, but also hare's serum. In order, however, to differentiate between rabbit's and hare's serum, Uhlenhuth advises the immunization of a rabbit with hare's serum. The serum of such an immunized rabbit precipitates 126 PRECIPITINS only hare's serum and not rabbit's, for the reason that "Isoprecipitins," i.e., precipitins against the same kind of animal, are, as a general rule, not developed. Similarly the differentiation between human and ape's serum can be accomplished by the immuniza- tion of apes with human serum. Attempts have also been made by means of the precipitation reaction to determine the origin of albumin in urine, and the foreign proteids circulating in the blood of artificially fed infants. The technique remains the same, independent of the purpose it is employed for. It consists in the mixing of the clear precipitating serum and the clear proteid, or albumin precipitinogen. Strongly precipitating antisera against proteid solutions are Production prepared by methods analogous to those employed for the of Proteid production of antibacterial sera. The use of rabbits is Precipitating generally advised. The sera or proteid solutions should be Antisera. sterile. Filtration through small porcelain niters may be necessary. The injections may be made subcutaneously, intraperitoneally, or intravenously. The subcutaneous route has no advantage unless the substances to be used are contaminated; but then, larger quantities are necessary. The intravenous path is the one of choice. A single injection of a large dose (15 to 20 c.cm.), or three injec- tions of moderately large doses (5 to 10 c.cm.) on three successive days may be given and the animal killed after 7 to 10 days. These methods have the advantage that a precipitating serum is obtained in a short period of time. They do not however yield as strong a precipitating serum as the following slower procedure. One c.cm. of the solution is injected four or five times at intervals of six days. The dosage may also be increased at each successive inoculation, for instance, beginning with 2 c.cm. of an animal serum and increasing gradually through 3, 5 and 8 c.cm. to possibly 15 c.cm. at the last injection. The animals should be weighed from time to time and if considerable loss of weight ensues during immunization, the intervals between injections should be increased. It is advisable to inject five or six animals at the same time, and by different methods, inasmuch as rabbits vary greatly in their indi- vidual power to produce precipitins and moreover, because some die after the third injection. Frequently only one serviceable serum is obtained, even though the immunization of five rabbits was under- taken. Beginning on the sixth day after the last injection, one should, at regular intervals of one or two days, remove a small quantity of blood from the vein of an ear and test the strength of the serum. As soon as' it is found to be satisfactory the animal should be bled and its serum preserved on ice, with precautions for sterility. The rules given above for obtaining 127 a clear serum should be kept in mind. If the precipitating value of the serum is insufficient, more injections may be given before the animal is finally bled. If the serum is not withdrawn at the proper time, its strength begins to diminish and further injections no longer stimulate new antibodies. It is even possible for the entire precipitin action of the serum to dis- appear. Titration ^^e ^°^ow^nS method of titration of the strength of a serum is the simplest. One c.cm. of various dilutions (i : 10, i : 100, i : 1000, i : 10000) of the proteid under examination (precipitin- ogen) is placed into different test-tubes and o.i c.cm. of the precipitating serum is added to each. The tubes should not be shaken, but it is occa- sionally necessary to place them in the incubator for one hour before any turbidity or precipitate appears. The least amount of proteid solution which still distinctly shows a precipitate is taken as the titer of the serum. For medico-legal purposes, Uhlenhuth advises the use of only Uhlenhuth's highly valent sera. Method of He considers an antiserum as efficient if o.i c.cm. of it, when Proteid Dif- mixed with its respective serum in the dilution of i : 1000, ferentiation. produces a distinct turbidity, either at once or in one to two minutes at the latest; three to five minutes is the limit for an indication of turbidity in the dilutions of i : 10000 and i : 20000. Like in all other biological reactions, control tests, here two in number, are of the utmost importance. One tube must contain o.i c.cm. of the precipitating serum mixed with i c.cm. of saline, another o.i c.cm. of the precipitating serum mixed with a heterologous serum in the dilution of i : 200 and i : 1000. Both of these tubes should show absolutely no pre- cipitate after twenty minutes. In this way the specificity of the pre- cipitin is determined; and it must be remembered that it is the quantitative specificity which counts. In the process of the determination of the nature of meats, it is es- pecially necessary to ascertain exactly the precipitating titer against bovine and pig's serum possessed by the rabbit's precipitating serum directed against horse's serum . When clear solutions are at hand the precipitin reaction is compara- tively simple. Frequently, however, the test must be performed with old and dirty blood stains, or all kinds of prepared sausage, so that the first and important task is to obtain a clear solution. In dealing with blood, milk, or seminal stains, the parts of the clothing involved are excised, divided into very minute shreds, and placed in a test-tube with a small amount of 0.85 per cent, of salt solution. If the material is not too old, extraction for one hour is usually sufficient, otherwise it may necessitate a period of twenty- four hours or more. Stains upon solid material such as steel, wood, stone, etc., 128 PRECIPITINS are carefully scraped off, and suspended in physiological salt solution. To obtain a clear solution the extract must be passed through filter paper or eventually the lilli- putian bacterial filter. In the examination of meats or other food stuffs, it is best to remove the material for examination from the center of its thickest part, as this portion has been least exposed to the methods of preservation, especially the high temperatures. Three hours' extraction is usually sufficient; the fresher the meat, the shorter this period. Very much salted meats are best washed with distilled water, previous to extraction. Inasmuch as a great deal of fat interferes with the reaction it is advisable to remove it beforehand by extraction with ether and chloroform for twenty-four hours (Miessner and Herbst). Before performing the actual test with the unknown blood stain, it is best to try out the entire reaction with a similar but known blood stain in order to make sure whether all the ingredients are in good working order. In laboratories equipped for medico-legal examinations, stains made upon linens from the blood of man, ox, pig, horse, etc., are always kept in readiness for such preliminary tests. Uhlenhuth indicates a set of rules to be observed whenever the reaction is undertaken. They are here cited in their original form, as practice has shown them to be of great service. "In order to obtain sufficient extract for the test, a small amount of the material is placed in a test-tube containing 5 c.cm. of normal salt solution. This must not be shaken. After one to two hours, 2 c.cm. are poured off into another tube and gently shaken. If a persisting froth appears upon the surface of the fluid, it can be taken as proof that sufficient extraction has occurred, and the rest of the fluid is thereupon also transferred to this tube. If no froth appears the 2 c.cm. should be returned into the first test-tube and the extraction continued until repeated tests finally show the pres- ence of froth. It is preferable not to disturb the sediment at the bottom of the test- tube. The extract eventually obtained may have to be filtered, if not absoutely clear. Such an extract is, as a rule, stronger than that required for the test, i.e., i : icoo. If one drop of a 25 per cent, nitric acid solution is added to i c.cm. of a i : 1000 serum dilution and then heated, a faint opalescence appears. Enough saline should there- fore be added to the final extract so that the nitric acid test corresponds to that given by a dilution of i : 1000. The following mixtures are then made: Precipitating Normal Result Test solution serum from rabbit rabbit's serum saline After five minutes After twenty minutes i c.cm. i : 1000 O I Opalescence. Turbidity and sedi- i c.c.m i : 1000 O I Clear. ment. Clear. O I i c.cm. Clear. Clear. PROTEID PRECIPITIN SPECIFICITY 1 29 The result should be read after twenty minutes at room temperature. As a further control a similar row of tubes should be made with the extract of the non-bloody part of the clothing in order to show that the latter alone does not give the reaction. Even putrid or otherwise chemically changed proteids may still give the precipitin reaction. The precipitation test only determines the animal species from which the proteid originates, but cannot prove whether it comes from the blood, semen, milk or other proteid body. In order therefore- to make a medico- legal diagnosis of " human blood stains," chemical evidences "Origin" must in addition be brought forward, that the stain really and "Con- consists of blood. Obermeyer and Pick have further shown stitutional" that fcesides animal specificity (" origin specificity"), pre- Specificity. dpitation also demonstrates the " constitutional specificity " of proteids. If instead of employing pure animal or plant proleids for the immunization of animals, variously changed albumins are used (heated albumins, acid albumins, formaldehyde albumin, etc.) the organism reacts by producing antibodies of a characteristic nature, different from those de- veloped after inoculation with the pure albumin. For example, the serum of a rabbit imm unized for a long time with horse's serum (normal immune pre- cipitin) will produce a precipitate when mixed in vitro with the pure horse's serum and not when added to the latter, heated, even if the normal immune serum is of very high titer. On the other hand, if a rabbit is injected with horse's serum which has been changed by being diluted and boiled for a short time, the immune serum thus obtained will react not only with normal horse's serum but also with heated serum and a group of its decomposition products with which the normal immune serum ordinarily never induces a precipitate. This fact is of practical application. In meat substitution, it is very popular to boil the sausage in order to make detection of the substituted meats more difficult. With the aid, however, of precipitins produced by immunization with heated proteids, this fabrication is more easily detected than if a normal immune serum were used. • While animal specificity is not destroyed when the albumins are modified in the above manner or changed by tryptic digestion or oxida- tion, Obermeyer and Pick have demonstrated that their specificity is lost when an iodin, nitro or diazo group is inserted in the proteid mole- cule. Immunization with such transformed proteid compounds, e.g., xanthoprotein, can produce a precipitating serum which will react with every xanthoprotein even in homologous animals. These authors conclude that species specificity is probably dependent upon a certain aromatic group of the proteid molecule. 9 130 PRECIPITINS It is interesting to note that the proteid contained in the lens of the eye belongs to this class of modified proteids which possess constitutional, but no species specificity. A serum produced by immunization with lens substance will react with the proteid derived from the lens of any animal, but with no other animal proteid. In conclusion, the origin of the precipitate formed during the Origin of precipitation reaction is of interest. When a very strong Precipitate, precipitating serum is employed, the precipitinogen is so greatly diluted that it no longer gives any of the chemical reactions for proteids, but nevertheless yields a heavy precipitate when the precipitating serum is added. This surely cannot come from the small trace of proteid in the precipitinogen. Furthermore, if the immune serum is diluted, the formed precipitate becomes comparatively weaker and disappears entirely if dilution is increased. It is, therefore, generally considered that the precipitate originates from the immune serum. CHAPTER XII. BACTERIOLYSINS AND HEMOLYSINS (CYTOLYSINS). If a guinea-pig is immunized with living or dead bacteria, for instance cholera or typhoid, and then to test its immunity is injected with a single fatal or many fatal doses of living bacilli, the animal remains alive; whereas a normal-control animal, not treated beforehand, succumbs to a similar inoculation. In order to determine the forces to which the im- munized animal owes its protection, Pfeiffer undertook the following experiment: Two guinea-pigs, one immunized and another normal, were simultaneously injected intraperitoneally with living cholera vibrios, and the peritoneal exudate was withdrawn from time to time and examined microscopically in hanging-drop preparations. (The method of withdraw- ing the peritoneal fluid with capillary pipettes and other technical details will be described below.) A very striking phenomenon occurred. While the cholera Pfeiffer's vibrios in the peritoneal exudate of the normal animal re- Phenomenon, tained their form and motility and increased in number con- tinuously until the animal succumbed to the infection, the bacteria in the peritoneal exudate of the immunized animal behaved quite differently; they first began to lose their power of locomotion, then their form changed, they broke up into evenly small shining masses, so- called "granula," and finally, after several minutes these also disappeared. Guinea-pigs injected with the peritoneal exudate from these infected immune animals remained healthy, and nutrient media inoculated with material from the same source remained sterile. The above experiment is named after its discoverer, Pfeiffer, and the phenomenon itself, "bacteriolysis." Bacteriolysis is a strictly specific process. If an animal which is immune to cholera is inoculated with typhoid bacilli, the bacteria markedly increase, as in a normal animal. The process by which this bacteriolytic force takes place is clearly demonstrated when a mixture of living cholera vibrios and blood serum of a guinea-pig, which has been actively immunized against cholera, is injected into the peritoneal cavity of a normal guinea- pig and, as a control, normal serum mixed with living cholera vibrios is inoculated into a second guinea-pig. Here the exudates on examination from time to time show that in the peritoneal cavity of the animal injected with the immune cholera serum, the same phenomena of bacteriolysis occur as described above, leading to the sterilization of the peritoneal cavity, and protection of the animal from illness. In the control animal , 131 132 BACTERIOLYSINS AND HEMOLYSINS however, the normal serum has no influence upon the bacteria, so that they increase rapidly and kill the animal. It is evident then, that the bacteriolytic power resides not only in the actively immunized animal, but that it may also be transmitted to other animals by means of the former's serum. Bacteriolysis, therefore, is not a property of the tissues of the actively immunized animal, but is to be traced to specific antibodies, "Bacteriolysins" which circulate in the blood serum and body fluids. From the above experiment it must be assumed that the phenomenon of bacteriolysis, like agglutination and precipitation, can be demonstrated also in vitro. The earlier investigations in this connection, however, were unsuccessful. Bordet was the first to obtain conclusive results and also to elucidate the cause of previous failures. While agglutination in vitro and bacteriolysis in 'vivo were Bacteriolysis readily produced by mixing living bacteria with old immune in Vitro, serum, bacteriolysis in vitro did not occur under similar cir- cumstances. But when freshly drawn blood serum or exudate of an immune animal was used, bacteriolysis took place in vitro also. (In fact, granule formation can be directly observed by the microscope.) When the serum becomes old — and twenty-four hours is sufficient to cause the change, it loses its bacteriolytic powers. It seems at first glance as if bacteriolysins may be active outside the body also, but that here they lead only an ephemeral existence. This view, however, is not quite correct; for " inactive" serum, which has become "ineffective" in vitro, can again produce bacteriolysis, if it is utilized to passively immu- nize healthy animals. Something must exist in the organism, which supplements the inactive bacteriolysins and restores their activity. This "reactivating substance" is independent of the immunizing process, since it is to be found in normal animals also. Furthermore, inasmuch as not only cholera and typhoid immune sera, but also all other immune sera and not only guinea-pig's serum but even rabbit's, horse's, and human serum may in like manner be reactivated, it is evident that the reactivat- ing agent lacks specificity. On account of this peculiar quality of supple- menting the inactive bacteriolytic serum so that it can develop its real effectiveness, Ehrlich called the reactivating substance "Complement." Accordingly, the complement is a normal non-specific substance which is found in the body fluids (particularly abundant in the blood serum) oj every organism; its existence is evidenced either by the activation or reactivation of bacteriolytic antibodies. Bordet demonstrated that the apparent ease with which the bac- teriolysins lose their activity is to be traced not to these bodies, but to the complement. If a small amount of fresh normal serum is added to bacteriolytic serum which has become inactive, reactivation occurs in PRINCIPLES OF BACTERIOLYSIS 133 vitro, that is to say, the bacteriolytic serum regains its ability to dissolve bacteria. The bacteriolytic power of fresh immune serum, depends, therefore, upon the fact that it contains not only bacteriolysins but also complement; the failure of old immune serum to produce bacteriolysis is accounted for by the lack of complement, while its capacity for reac- tivation is explained by the still present bacteriolysins. As the above-described experiments indicate, bacteriolysis is a complex process, which is produced by the interaction of two substances; one, the bacteriolysin, is formed through an immunizing process, and accordingly is a specific antibody of great stability, while the other, the complement, is a normal non-specific and very labile serum substance. The stability of the immune bacteriolysin is evident in its resistance to heat, whereas the complement is thermolabile. If freshly drawn immune serum is heated to 56° C. for one-half hour, the complement is, as a rule, rendered ineffective, while the bacteriolysin is not in any way in- jured; it retains its specificity, and the degree of its affinity to antigen remains unchanged. Bacteriolysins are interfered with by temperatures above 60° C. only. Concerning the finer mechanism of bacteriolysis there are two opposing views, that of Bordet and of Ehrlich. Without considering too closely the remarkable researches of these two investigators, the synonyms for bacteriolytic antibodies usually found in the literature will be reviewed. In attempting an explanation for bacteriolysis, Bordet has recourse to certain phe- nomena in staining technique. There are some substances which can be stained only when prepared in a definite way by means of another substance, a so-called mordant ("Beize") which itself is not a stain. According to Bordet, the specific substance pro- duced by immunization represents a kind of mordant which "sensitizes" the bacteria to the action of the second normal non-specific substance; the latter is really the active agent in causing the dissolution of bacteria and is called by Bordet "alexin"— an older term used by Buchner — in contradistinction to "substance sensibilitrice." Ehrlich, on the other hand, advocates a more chemical conception of the essential process of bacteriolysis. He believes that the substance formed by immunization, which, for the sake of brevity, is called the immune body, is characterized primarily by the fact that it has two binding groups. One of these has a chemical affinity for the bac- terial cell and is, therefore, known as the "cytophile group," the other is characterized by its binding affinity for complement and is, therefore, known as the " complemento- phile" group. Also because of its two binding groups (receptors) the immune body itself is called amboceptor, that is, double receptor. Thus, according to Ehrlich, bacteriolysis takes place in the following way : The cyto- phile group of the amboceptor, which is strictly specific for its antigen, attaches itself to the antigen, for instance the cholera vibrio; while the complementophile group binds the complement. The complement must be regarded as a sort of digesting (proteolytic) ferment. Although it is always present in normal serum, it is not effective, because bacteria have no affinity for it. Only through the medium of the amboceptor (Zwis- chen-Korper, intermediary body) , can complement bind itself to bacteria and dissolve them. 134 BACTERIOLYS1NS AND HEMOLYSINS The specificity of the bacteriolytic process depends, therefore, on the specificity of the cytophile group, while the complementophile group possesses no or, strictly speaking, only slight specificity; it adapts itself to the complements of very many though not quite all kinds of animals. Recent experiments have proven that the complement consists of two different parts, the middle piece and the end piece. Technique of Bacteriolytic Experiments. To determine the occurrence of bacteriolysis there are two methods of procedure : 1. Pfeiffer's experiment. 2. The bactericidal plate method. I. The Pfeiffer's Experiment. The essentials of Pfeiffer's experiment have been described at the beginning of this chapter. Briefly, it consists in injecting intraperito- neally, in a normal animal, bacteriolytic immune serum mixed with living bacteria. The resulting bacteriolysis is studied microscopically by with- drawing small amounts of peritoneal exudate from time to time. If this experiment is performed with various dilutions of immune serum, and if it be determined at what dilution bacteriolysis fails to occur, then the bacteriolytic titer is evident. The details can best be understood by taking a practical example. It is desired to find the bacteriolytic titer of the serum of a patient recovering from typhoid fever by means cf the Pfeiffer experiment. To accomplish this task the following ingredients are needed: 1. A strain of bacillus typhosus of known virulence for guinea-pigs. 2. Patient's serum, sterile, and free from complement. 3. Guinea-pigs of 250 gms. weight. A preliminary experiment must be performed in order to determine the virulence of the typhoid strain. TESTING THE -VIRULENCE OF STRAIN. . _ , '• ' • -'".••' '• Guinea-pig No. i. I i/II '09 One loopful of a typhoid agar culture suspended in j 2/II dead. i UV.U1. 01 uuumoii, mjecicu incrapenioneaiiy. Guinea-pig No. 2. i/II '09 One- half loopfui of same 2/II dead. Guinea-pig No. 3. i/II '09 One-fifth loopful of same ....... 2/II dead. Guinea-pig No. 4. i/II '09 One-eighth loopful of same , 2/II sick. 4/II dead. Guinea-pig No. 5. j i/II '09 One-tenth loopful of same, 2/II sick. 3/II well. INCREASE OF VIRULENCE OF BACTERIA 135 As far as the Pfeiffer experiment is concerned the titer of virulence in this case is 1/5 of a loopful of an agar culture because this dose is fatal within twenty-four hours. In order, however, to make sure of excluding all individual variations, which can and occasionally do occur, it is advisable to use not the titer dose, but its fifth or tenth multiple, that is, in this case, one loopful. Doses larger than one loopful should be avoided, so that if any par- ticular strain of typhoid bacilli is not sufficiently virulent, necessitating the use of larger doses, the virulence must first of all be increased. This is done by passing the organism through animals such as guinea-pigs. The method is as follows: A very large dose of the culture, for example To Increase the surface of an entire agar tube, is injected intraperitoneally. Every the Virulence, animal succumbs to this enormous dose. The bacteria-laden exudate from the abdominal cavity, which, of course, must be removed under sterile precautions, is then inoculated into a second guinea-pig and when it dies, into a third, and so on. As a rule, after passing through one or two animals the bacterial strain (which must be grown pure from the cadaver) becomes more virulent, as can be proven by titration. Very often the virulence is increased exclusively for the species of animal used and occasionally this is associated with a decrease in virulence for other species. After a series of passages through animals, the strain reaches a maximum strength beyond which it cannot be increased. The degree of virulence varies with the type of bacteria. Typhoid and cholera usually reach only a moderate virulence (i/io to 1/20 loopful); the bacteria of the hog cholera group can acquire a distinctly higher virulence; for instance, B. paratyphosus, i/ioo to i/iooo of a loopful, while the streptococcus and pneumococcus reach the highest figures, 1/10,000 to 1/1,000,000 of a loopful. For the Pfeiffer's experiment with cholera or typhoid, the most suitable strains are those of such a virulence that 1/5 to i/io of a loopful injected intraperitoneally kills in twenty-four hours. The serum to be investigated is freed of its complement by Technique of heating in a water-bath for one-half hour at 56° C. Then a Pfeiffer's series of dilutions are made in bouillon (not in salt solution) Experiments. for mstance i/io, i/ioo, i/iooo, etc. A c.cm. of each dilu- tion is put into a test-tube (a sterile pipette should be used) and rubbed up with a standard loopful of an 1 8 to 24 hour agar culture of typhoid bacteria. Finally the contents of each test-tube are injected intraperitoneally in a guinea-pig of 250 gms. weight. Inasmuch as small amounts are apt to be lost when aspirating the fluid with the syringe as well as when pouring the bacterial emulsion into a watch glass, it is better to rub up two loops of the culture in 2 c.cm. of bouillon instead of i loop in i c.cm., and then withdraw only i c.cm. for use in the experiment. The following controls should be prepared: i. Dilutions of the serum of a normal person (or animal of the same type) + typhoid culture. 136 BACTERIOLYSINS AND HEMOLYSINS 2. Dilutions of immune serum + a heterologous culture. 3. (a) Bouillon + typhoid-culture. (b) Bouillon + heterologous culture. The study of the bacteriolytic phenomena follows the inoculation. For this purpose capillary pipettes to withdraw the peritoneal exudate are prepared according to the directions of von Issaeff. A thin glass tube is heated in a Bunsen flame almost to the melting point, then removed from the flame and immediately drawn out with a sudden jerk. Very fine capillary pipettes can thus be made. The removal of the exudate is accomplished as follows: a small cut is made with scissors through the skin of the guinea-pig's abdomen; the capillary pipette, the large end of which is kept closed with the index finger, is forced into the abdominal cavity with a single push. The pressure of the finger is next relaxed and the tube slowly withdrawn. In order to avoid injuring the intestines, the precautions usually advised in intraperitoneal inoculations should be observed here. The author has found Friedberger's method of holding the animal very serviceable (see Fig. 5). The procedure is absolutely painless, moreover, the ordinarily sensitive guinea-pigs withstand the operation almost without uttering a sound. It is best to withdraw the exudate immediately after the injection and then at intervals of five to ten, twenty, and thirty minutes, etc. Observa- tions are made directly in hanging-drop preparations. Stained speci- mens are less reliable and instructive because, according to the investiga- tions of Radziewsky, the findings are dependent upon the kind of coloring matter used. Bacteria which are in the process of dissolution soon lose the power of being stained by methylene blue, while they retain their affinity for carbol-fuchsin and aqueous solution of gentian violet. Granules are demonstrated only incompletely in stained preparations. The prognosis for the animal quoad vitam is unfavorable, if bacterioly- sis does not occur; good, if it does. Yet there are exceptions to the latter rule, a subject to which reference will be made later on. Now that the most important technical details of the Pfeiffer phenomenon have been considered, the following protocol will more clearly illustrate the exact procedure. Titration of a bacteriolytic serum (after Pfeiffer). TITRATION OF A BACTERIOLYTIC SERUM 137 Guinea-pig 1-4-07 One loopful of -fo.i typhoid Beginning of bacterioly- Animal No. i. typhoid culture. serum. sis after 10 minutes; remained ^ _, after 30 minutes exu- alive. ~^~ in i c.cm. of bouillon intraperi- date was sterile. tone ally. \ Guinea-pig 1-4-07 One loopful of H-O.OI typhoid Beginning of bacterioly- Animal No. 2. typhoid culture. serum. sis after 15 minutes; remained c ' ^ after 20 minutes only alive. *" in i c.cm. .of bouillon intraperi- isolated, non-motile toneally. bacteria, many gran- ules; after 40 minutes sterile. Guinea-pig 1-4-07 One loopful of -f-o.ooi typhoid Beginning of bacterioly- 2/4 animal No. 3. typhoid culture. serum. sis after 15 minutes; slightly ill. ,^_ _, after 20 minutes nu- 3/4 animal ^ in i c.cm. of bouillon intraperi- merous granules and recovered toneally. also many non-motile and re- bacteria; after an hour, mained no bacteria at all. alive. Guinea-pig 1-4-07 One loopful of +0.0001 typhoid After 20 minutes gran- 2/4 animal No. 4. typhoid culture. serum. ules and many motile found ^ t bacteria; after i hour dead. in i c.cm. of bouillon intraperi- motile bacteria very tone ally. numerous. Guinea-pig 1-4-07 One-fifth loopful After 20 minutes gran- 2/4 animal No. 5. in i c.cm. of ules and many motile found bouillon intra- bacteria; after i hour dead. peritoneally. motile bacteria very numerous. Guinea-pig 1-4-07 One loopful of +o. i normal After 15 minutes many 2/4 animal No. 6. typhoid culture. serum. granules, also isolated found motile and non-motile dead. ^ in i c.cm. of bouillon. bacteria; after 20 min- utes motile bacteria; after 30 minutes motile bacteria very numer- ous. Guinea-pig 1-4-07 One loopful of +0.1 typhoid Complete similar find- 2/4 animal No. 7. bacteria, para- serum. ings. found typhosus. (Vir- dead. ulence I/ 10 loopful.) in i c.cm. of bouillon. The bacteriolytic titer of the tested serum in this case would lie between o.ooi c.cm. and o.oooi c.cm. and could be exactly determined by further tests which would take into consideration the intermediate doses. 138 BACTERIOLYSINS AND HEMOLYSINS On close study of the above experiment, it will be noted that even in those cases in which the animals died of the infection, bacteriolytic phe- nomena were not altogether absent. They occurred particularly in the beginning and were incomplete. This can be considered as evidence of the fact that even normal animals possess a certain supply of bacteriolysins which are, however, readily exhausted. This amount of normal bac- teriolysin in serum varies greatly with the species of animal; thus the sera of man and rabbit contain very little normal bacteriolysins for cholera and typhoid, while horse's serum is well supplied with the same. According to Kolle, a loopful of virulent cholera vibrios is destroyed in the peri- toneal cavity of a guinea-pig, by 0.005 to o.oi c.cm. of normal horse's serum, o.oi to 0.02 c.cm. of normal ass serum. 0.02 to 0.03 c.cm. of normal goat's serum, o.i to 0.3 c.cm. of normal rabbit's serum. The protective action of bacteriolytic sera differs very essentially from that of antitoxic sera. For the latter, the law of multiple proportions holds true; a stronger dose of toxin is neutralized by a proportionately larger amount of antitoxin; to bacteriolytic sera this rule does not Endotoxin. apply. If the bacteria are increased beyond a certain quantity, their dissolution can indeed be accomplished by the addition of sufficient amounts of bacteriolysins, but the animal dies nevertheless. Its peritoneal cavity examined during life or postmortem may be absolutely sterile. Pfeiffer's explanation for this phenomenon is that the endotoxins within the bacteria are liberated by bacteriolysis and kill the animal. Fatal results from endotoxin follow in a similar manner when dead instead of living bacteria are injected. Since endotoxins can continue their effective action in spite of the serum, it is evident that the usual bacteriolytic serum lacks the power to neutralize the poisons of the endotoxins. Many investigators have at- tempted to supply this deficiency. (This will be considered later.) • While bacteriolysis may take place without any resulting protective action, on the other hand a serum may be curative in spite of the absence of bacteriolysis. This is well demonstrated in MetchnikofFs experiment. A marked leucocytosis in the abdominal cavity of a guinea-pig is produced by the intraperitoneal injection twelve hours mkoff's . . Experiment Prevlously of 5 to 10 c.cm. of aleuronat solution or sterile bouillon. Pfeiffer's experiment is then performed. As a rule, bacteriolysis occurs up to a certain point, particularly when cholera vibrios are used; most of the bacteria, however, retain their form and are taken up by the leucocytes. Metchnikoff used this experiment to uphold his theory of the signifi- IDENTIFICATION OF BACTERIA BY PFEIFFER's TEST 139 cance of phagocytosis. Pfeiffer maintained that bacteriolysis was the most important protective weapon of the immune organism, against bacte- rial invasion. According to Metchnikoff and his followers among whom Bail in particular must be mentioned, bacteriolysis in the abdominal cavity is only an exceptional phenomenon (test-tube experiment in vivo); its occurrence is made possible by the circumstance that the abdominal cavity is as a rule almost free of wandering cells, and that the few which are present are so injured by the severity of the infection that they disinte- grate. If their number increases, bacteriolysis does not occur, or at least is only slight. Likewise, bacteriolysis is incomplete in the presence of cells, for instance in the blood, spleen, liver and subcutaneous tissue, etc. A detailed consideration of this much mooted problem does not fall within the compass of this book. It is sufficient to have pointed out the great questions of fundamental significance which hinge upon the discus- sion of the Pfeiffer experiment, questions which concern the essential features of antibacterial immunity. It can be readily understood, there- fore, why the phenomenon of bacteriolysis has been so much studied, although its practical significance is only limited. The Pfeiffer experiment can be used for the differentiation of The Practi- bacteria as well as for the demonstration of bacteriolysins in cal Applica- serum. It serves as a control for the agglutination reaction, tion of the Pfeiffer Ex- Pfeiffer and Kolle, Brieger and others, have used bacteriolysis as a periment. method of estimating the immunity obtained by active protective im- munization against cholera and typhoid in man. It must, however, be questioned whether it is admissible to draw conclusions as to the degree of active im- munity from the height of the bacteriolytic titer of the serum, inasmuch as animals are found which possess no active immunity and still have sera of high bacteriolytic powers. The most important practical use of the Pfeiffer experiment lies in the identification of suspected cholera cultures. In Germany, the Pfeiffer test made with the vibrios obtained in pure culture from the suspected patients is required for the official diagnosis of the first cases of cholera. The serum used for this purpose should be at least strong enough in amounts of 0.0002 c.cm. to cause the disintegration of the bacteria in one hour, when a mixture of one loopful of an eighteen-hour agar culture of cholera in i c.cm. of nutrient bouillon is injected into the peritoneal cavity of a guinea-pig. For this experiment four guinea-pigs of 250 gms. weight are used. 140 BACTERIOLYSINS AND HEMOLYSINS Animal Culture Serum Method of injection Result in cholera cases No. i One loopful of an 1 8 o.ooi c.cm. chol- Intraperitoneally. After 20 minutes or hours' growth of era serum=5 at the latest i hour, culture suspected times the titer bacteriolysis oc- to be cholera in i dose. curs; animal re- c.cm. of bouillo'n. mains alive. No. 2 One loopful of an 18 0.002 c.cm. chol- Intraperitoneally. After 20 minutes or hours' growth of era serum =10 at the latest i hour culture suspected times the titer bacteriolysis oc- to be cholera in i dose. curs; animal re c.cm. of bouillon. mains alive. No. 3 (con- One loopful of an 18 o 01 c.cm. of nor- Intraperitoneally. Increase in number trol). hours' growth of mal serum = 50 of bacteria; animal culture suspected times the titer dies. to be cholera in dose of the im- i c.cm. of bouillon. mune serum. No. 4 (con- One-fourth loopful Intraperitoneally. Increase in number trol of vir- of an 1 8 hours' of bacteria, animal ulence of growth suspected dies. culture) . of being cholera in i c.cm. of bouillon. In cases of subsiding cholera, the Pf eiffer experiment is performed with the serum of the patient in dilutions of i to 20, i to 100 and i to 500. Bacteriolysis with typhoid organisms is less typical than with cholera. For diagnostic purposes the test is resorted to, only when the agglutination reactions are doubtful. When bacteriolysis also gives uncertain results, an animal is immunized with the typhoid suspected bacteria and its serum tested for its power of agglutinating or destroying definitely known typhoid bacteria and eventually the immunized animal may be injected with virulent typhoid bacilli. Bacteriolysis is even more unsatisfactory with bacillus paratyphosus B. and the related hog cholera group of organisms. While with typhoid bacteria the onset of bacteriolysis offers a favorable prognosis for the animal, guinea-pigs inoculated with bacteria of the paratyphoid-hog-cholera group die in spite of complete bacteriolysis. Death always takes place late (from three to six days) , while the control animals succumb in about twenty-four hours. Bacteriolysis has also been observed with the bacillus of dysentery and with the tubercle bacillus; but thus far, these phenomena have gained no clinical significance. Bacteriolysis does not occur in anthrax, pest and the various diseases due to cocci. NEISSER AND WECHSBERG'S BACTERICIDAL TEST 141 II. Bactericidal Plate -culture -method. (Plattenverfahren) according to Neisser and Wechsberg. For the determination of the bactericidal titer of a serum, Neisser and Wechsberg recommended the so-called bactericidal plate-culture method. The principle of it is as follows: the serum to be tested is inactivated; different amounts of this inactivated serum are mixed with a definite constant quantity of bacteria, and a constant quantity of active normal serum is added as complement. This mixture is left in the thermostat sufficiently long to permit the occurrence of bacteriolysis. Now, to determine whether and to what degree death of bacteria resulted from the effect of the reactivated bacteriolysins (or of some bactericidal substance otherwise unknown), agar is added, the mixture plated, and the number of colonies counted. Stern and Korte recommend this procedure for clinical purposes, as a substitute for the Pf eiff er test in the diagnosis of typhoid fever. They point out the sparing of animals as one of its advantages. On the other hand, this method consumes much more time and its results are less trustworthy. It has not found a place, therefore, in clinical practice. The technique of Stern and Korte is the following: the serum The Tech- °f the patient, and that of a person not ill with typhoid as nique of control, are inactivated for one-half hour at 56° C. and i c.cm. the Method. of each in decreasing dilutions is poured into sterile test-tubes. To each is added 0.5 c.cm. of a twenty-four-hour typhoid bouillon culture diluted in bouillon to i : 5000 or i : 10000. For reactivation 0.5 c.cm. of fresh normal rabbit's serum in a dilution of i to 12 in physio- logical saline is added and the whole thoroughly shaken. The tubes are then placed in the thermostat for three hours. The entire contents of each mixture is plated on agar, and after eighteen to twenty-four hours the plates are to be examined. That particular plate is considered to indicate the extreme limit of the bacteriolytic action of the serum in which there is still evident a very great decrease in the number of colonies as compared with the innumerable colonies found on the control plates. Certain other controls are necessary: 1. One tube containing culture and complement. 2. One containing culture and inactivated immune serum in the highest concentration used. 3. The same with inactivated normal serum instead of immune serum. 4. Complement without culture and immune serum to test its sterility. 5. Immune serum without culture and complement to test its sterility. 6. One tube containing only culture, to be plated immediately. 7. One tube, containing only the culture, to be plated after standing in the thermostat for three hours. 142 BACTERIOLYSINS AND HEMOLYSINS Topfer and Jaffe pour a thin layer of agar into a petri dish and let it harden. Upon this the culture-serum- agar mixture is poured, and after hardening is covered with another thin layer of agar. In this way the formation of a film of culture in the water of condensation is avoided. A practical example is appended to illustrate the plate culture method. Result (poured after remaining 3 hours in the thermostat) Culture Serum Complement Normal serum Immune serum 0.5 c.cm. 1/5000 typh. i/ioo c.cm. 0.5 c.cm. : 12 rabbit's. o colonies. Many thousand. 0.5 c.cm. 1/5000 typh. 1/500 c.cm. 0.5 c.cm. : 12 rabbit's. 100 colonies. Many thousand. O.5 c.cm. 1/5000 typh. I/ 1000 c.cm. 0.5 c.cm. 12 rabbit's. Many thousand. Many thousand. 0.5 c.cm. 1/5000 typh. 1/5000 c.cm. 0.5 c.cm. 12 rabbit's. 00 Many thousand. O.5 c.cm. 1/5000 typh. i/ioooo c.cm. 0.5 c.cm. 12 rabbit's. CO 500 0.5 c.cm. 1/5000 typh. 1/20000 c.cm. 0.5 c.cm. 12 rabbit's. 00 200 . 0.5 c.cm. 1/5000 typh. 1/30000 c.cm. 0.5 c.cm. 12 rabbit's. 0 0.5 c.cm. 1/5000 typh. 1/40000 c.cm. 0.5 c.cm. 12 rabbit's. 5 0.5 c.cm. 1/5000 typh. 1/50000 c.cm. o. 5 c.cm. 12 rabbit's. 60 0.5 c.cm. 1/5000 typh. i/iooooo c.cm. o. 5 c.cm. 12 rabbit's. 800 0.5 c.cm. 1/5000 typh. 1/200000 c.cm. o . 5 c.cm. 12 rabbit's. CO Control I 0.5 c.cm. I : 12 rabbit's. Many thousand. 0.5 c.cm. i/soco typh. Control II and III 1 i/ioo c.cm. — CO 00 0.5 c.cm. 1/5000 typh. Control IV — 0.5 c.cm. i : 12 rabbit's. 0 Control V i/ioo c.cm. — 0 Control VI — — Many thousand. 0.5 c.cm. 1/5000 typh. immediately poured. Control VII — — 00 0.5 c.cm. 1/5000 typh. poured after 3 hours. ' ' In addition to the results which one would expect, this experiment shows one strik- ing point. With the normal serum the tube which contains the largest amount of normal bacteriolysins shows, on plating, the fewest germs. The greater the dilution of the serum the more prolific is the bacterial growth. The titer of the normal serum in this case lies between i/ioo and 1/500. The controls show that the serum and comple- ment are sterile, and that the inactive normal serum is ineffective. During the three hours in the thermostat the bacterial suspension has become stronger. The retarded growth in the complement culture tube can be traced probably to the presence of normal bacteriolysins. With the immune serum, on the other hand, results are quite different. Where the most concentrated serum is used, the bacterial growth is still rather profuse; only the moderate doses show a true bactericidal action and the small doses are altogether ineffective. The titer of this serum is between 1/30,000 and 1/40,000. Neisser and Wechsberg explain this phenomenon by the so-called ''deviation of the complement." They assume that in the serum of higher DEVIATION OF COMPLEMENT 143 concentration there are so many amboceptors that the bacteria cannot bind them all. The amboceptors remaining free attach themselves to the complement by means of their complementophile group just as the already bound amboceptors have done. Thus, a part of the complement is deviated from the bacteria and only an incomplete bacteriolysis takes place. The theory of complement deviation does not in the opinion of the author with- stand critical examination. Particularly the evidence brought forward by Bordet and Gengou that the affinity of complement for the bacterium -famboceptor complex (Sen- sitized bacterium) is considerably greater than for free amboceptor, militates against the view of Neisser and Wechsberg. It is possible that agglutination may account for the phenomenon of deviation of the complement in that the agglutinated masses of bacteria afford a more resistant barrier to the action of the bacteriolysins. The author has now and then observed an analo- gous phenomenon in hemolytic experiments; strong doses of hemolysin were less effect- ive than moderate ones, and in these cases the momentary hemagglutination was readily visible. Also, by titrating bactericidal sera in animal experiments, it has been found that moderate doses often afforded the greatest protective action. For the practical application of the plate culture method, knowledge of the following data is important, as it is necessary to consider the difference between the bactericidal titer of sera of normal and of typhoid patients. According to Korte and Steinberg the bactericidal titer was Of normal cases Of typhoid cases Under 100 in . . 74 per cent o o per cent Between 100 and 1000 in. . . 8 6 per cent. 3 3 per cent Between 1000 and 10 ooo in. 15 4 per cent. 15 i per cent Between 10,000 and 100,000 in Over 100,000 in 2.0 per cent, o o per cent. 23.3 per cent. 58 3 per cent In typhoid fever the bactericidal titer does not run strictly parallel either with agglutination or the Pfeiffer experiments. It falls toward the end of the disease and is low during convalescence. Besides its use in typhoid fever, the plate culture method has been employed for experimental purposes in cholera and dysentery; in these diseases, however, it possesses no clinical diagnostic significance. Concerning bacillus paratyphosus B, the views of different authorities are widely at variance. While some obtained very good results, similar to those found in typhoid fever, Topfer and Jaffe could demonstrate no bactericidal power whatever in vitro. This difference can be explained only by variations in sera. 144 BACTERIOLYSINS AND HEMOLYSINS Hemolysins. An animal that is injected with the red blood cells of a different species develops in its serum antibodies which are biologically analogous to bacteriolysins and differ from them only in that they cause disintegration of erythrocytes instead of bacteria. These antibodies are therefore called hemolysins, or to be more precise immune-hemolysins, since they arise through a process of immunization. The breaking up of the red blood corpuscle, hemolysis, is recognized by the naked eye. The hemo- globin passes from the erythrocytes into the surrounding fluid (serum or physiological salt solution) and colors it red. The previously opaque blood lakes and becomes transparent. Immune-hemolysins like bac- teriolysins belong to the class of amboceptors. They are relatively ther- mostabile in that they withstand a temperature of from 56° to 58° C. without being injured; they require complement for the develop- ment of their hemolytic action. Furthermore, immune-hemolysins, like all amboceptors, are specific, i.e., the serum of a rabbit immunized against horse's blood can dissolve only the blood of a horse and not that of a hen or cow. On the other hand, group reactions occur here also; for instance the immune-hemolysin produced in a rabbit against horse's blood is likewise active against donkey's blood. Just as various antitoxins, agglutinins, precipitins and bac- Normal teriolysins can be found in normal serum, so also can normal Hemolysin. hemolysins of amboceptor structure be discovered in the blood of many animals. While normal hemolysins come into play in only a few reactions, as in several modifications of the Wassermann test, the significance of immune- hemolysins is extraordinarily great. These antibodies, discovered by Bordet, and independently by von Dungern and Landsteiner, were care- fully studied by Ehrlich and Morgenroth and many others. Such re- searches have, first of all, greatly advanced the subject of immunity in its theoretical aspects, in that they have created the possibility for the dis- covery in minute detail the finer relationship which has explained some of the phenomena occurring in bacteriolysis. Furthermore, the studies of hemolysins led to the discovery of the complement fixation method, a procedure of exceptional practical value. As far as the technique for obtaining immune-hemolysins is Production concerned, the rules which hold for every process of immu- ofHemo- nization are naturally to be followed here also. It is not lytic Sera, possible, however, to immunize every kind of animal against every type of red blood corpuscle. Rabbits, goats, horses and chickens are the ones which are best adapted to supply hemolytic sera. An animal produces a better hemolysin the remoter its relationship to the PREPARATION OF HEMOLYSIN 145 animal from which the erythrocytes for injection are taken. The blood to be injected can be employed in just the condition in which it flows from the vein. Nevertheless it is as a rule defibrinated, to prevent coagulation. The simplest and most practical way of doing this is to place some glass beads into a bottle or Erlenmeyer flask and then sterilize it by dry heat. The blood coming from the vein is allowed to flow into one of these flasks and then it is repeatedly shaken for several minutes. This suffices to defibrinate the blood and thus prevent coagulation. The production of hemolysins depends entirely upon the red blood corpuscles. The presence of the serum is not only superfluous, but even harmful, as experience has shown that dangerous reactions may follow the injection of foreign serum. Before injecting, therefore, the erythrocytes are washed. For this Washing of purpose a few cubic centimeters of defibrinated blood are poured into Red Blood a centrifuge tube and the level of the fluid marked on the tube. An Corpuscles, equal or double this amount of 0.85 per cent, saline is added, and the tube rapidly centrifugalized. The erythrocytes fall to the bottom, while the upper layers of the tube consist of diluted serum more or less tinged with hemo- globin. The fluid is carefully decanted, fresh saline added, the tube gently shaken, and again centrifugalized. If this is done two to three times the erythrocytes can be freed of the last traces of serum; finally, by adding saline up to the mark made at the begin- ning of the experiment, the erythrocytes are obtained in the normal concentration, just as in the blood, but completely free of serum. The washed, defibrinated blood can be injected subcuta- Immuniza- neously, intravenously, or intraperitoneally. With the sub- don, cutaneous and intraperitoneal methods in a rabbit, injections of from 5 to 20 c.cm. are necessary at intervals of five to six days. Far larger quantities should be given to bigger animals, like goats and sheep. Subcutaneous injections often cause infiltrations and occasionally abscesses. The author therefore uses the intravenous method in rabbits. A suspension of washed blood corpuscles is diluted four to five times with physio- logical saline; o. 5 to i .o c.cm. of this fluid is slowly injected into the ear vein every five to six days. Three injections are almost always sufficient for procuring a good serum. The animals sustain the first two injections with ease, but the third and following ones are not altogether without danger. This is supposed to be akin to anaphylactic phe- nomena. It is therefore advisable to immunize several animals simultaneously, so that in case one dies there is another to replace it. Furthermore there are such marked individual variations in the ability to produce hemolysins that it is best to have several animals to choose from. Beginning on the sixth day after the third injection, blood should be withdrawn for the determination of the hemolytic strength and this process repeated daily until the titer has reached a satisfactory height and then the animal should be bled. If only a small amount of hemolysin is needed,ithe animal can be allowed to live; it will gradually lose its titer completely and will act apparently like a normal 146 BACTERIOLYSINS AND HEMOLYSINS animal. Nevertheless, an essential difference exists. For if the animal previously immunized is again injected, hemolysins reappear after a short incubation period, whereas in a normal animal a prolonged immunization is necessary. Hemolysins, therefore, exist to a certain extent in a preformed state in the cells of an immunized animal. If a stimulus to immunization occurs, the hemolytic substances are thrown off into the circulation, while in a normal animal the formation of hemolysins by the cells must first take place. If a great amount of hemolysin of the same titer is needed, it is The Preser- best to bleed the animal to death. For the preservation of vation of hemolysins the author recommends the following procedure Hemolysins. which he has found very trustworthy. One to 3 c.cm. of serum obtained sterile are poured into sterile tubes, which are closed with non-absorbent cotton. The tubes are placed into a water bath at 56° C. for one-half hour to inactivate the serum and are then cov- ered with sterile rubber caps. (These are sterilized by placing them in a i per cent, sublimate solution for forty-eight hours.) An immune hemolysin must answer both qualitative and quantitative determinations; qualitative, whereby is proven that the serum can hem- olyze only the red blood cells which serve as antigen or to a slight degree those of related animals, and that it has only the effect of a normal serum upon the erythrocytes of other animals. The quantitative estima- tion supplies the only means for the absolute differentiation between a normal and an immune serum. In complement fixation where hemolysis bears an active part, it is the quantitative use of the hemolysin which decides the result of the reaction. The immune serum must therefore be "titrated." If fresh active hemolytic immune serum is used, a constant quantity of blood serving as antigen is mixed with decreasing quantities of the immune serum and the mixtures placed in the thermostat. Results like the follow- ing will be obtained. Antigen blood Hemolytic serum of immune rabbit Result after 2 hours i c.cm. of 5% sheep's blood ! i c.cm. of active serum, i to 10 i c.cm. of 5% sheep's blood i c.cm. of active serum, i to 20 i c.cm. of 5% sheep's blood, i c.cm. of 5% sheep's blood. i c.cm. of active serum, i to 50 i c.cm. of active serum, i to 100 Hemolysis. Incomplete hemolysis. Incomplete hemolysis. No hemolysis. On the basis of this experiment the titer of the hemolytic serum for sheep's blood would lie between i/io and 1/20. But this is incorrect, as it was pointed out previously that by immunization only the amboceptors are increased and the complement remains unchanged. Each of the above TITRATION OF IMMUNE HEMOLYSIN 147 dilutions decreases therefore not only the amount of hemolysin, the quantitative estimation of which is the object of the experiment, but also the complement. Inasmuch as the latter was not at first increased, a point is soon reached where there is no complement at all in the diluted fluid; as a result hemolysis cannot occur, for only the combination of hemolysin + sufficient complement can exhibit any hemolytic action. Correct titration consists therefore in allowing varying quantities of hemolysin with a constant amount of complement to act upon a constant quantity of red blood cells. The simplest method of accomplishing this is first to destroy the complement by inactivation of the hemolytic serum, then to make the desired dilutions, and finally to add, to all, the same amount of normal serum as complement. The normal serum of an animal of the same species as that which provided the immune serum can under no circumstances serve as complement. On the contrary, foreign sera are much more suitable; and guinea-pig's serum is especially recommended as comple- ment when immune rabbit's serum is used. Not every complement serves equally well for any immune serum. A very good hemolytic system and one which is almost exclusively used for the complement fixation reaction, is sheep's blood as antigen, rabbit's immune hemolysin as amboceptor and normal guinea-pig's serum as complement. The preparation of these ingredients should be carried out as follows: i. Sheep's Blood. — This should be defibrinated and washed. Washing is necessary because fresh sheep's blood contains complement; if the blood is a few days old, washing is even more important. Although serum which is not fresh does not contain sufficient active Comple- complement to cause the danger of superfluous complement, it never- mentoids. theless contains substances which interfere with hemolysis. Probably the existence of " complementoids " is the disturbing factor. It must be assumed that complement is composed of two biologically different parts, as is the case with toxins and ferments. One is the haptophore group, which has affinity for the complement ophile group of the amboceptor and is the more stable of the two. The other corresponds to the energy group of the toxins (toxophore element) and of the ferments. Just as after the destruction of the toxophore group there remain only innocuous toxoids whose single perceptible activity consists in their ability to neutral- ize antitoxins, so also, after the destruction of the weakly resistant energy elements of the complement, there remain complementoids which lack the ability to activate a bacteriolytic o"r hemolytic amboceptor, although by virtue of their uninjured hapto- phore groups they bind the complementophile groups of the amboceptors. In this way they usurp the place of whatever active complement may still be present, rendering the latter inactive, and as a result hemolysis is absent or incomplete. Following the technique of Ehrlich and Morgenroth, a 5 per cent, sus- pension of washed red blood corpuscles is employed to test a hemolysin. 148 BACTERIOLYSINS AND HEMOLYSINS A pipette, closed at the top by pressure of the index finger, is thrust to the bottom of the washed erythrocytes contained in the centrifuge tube; a definite amount, for instance i c.cm., is withdrawn and allowed to flow into a graduate. For diluting purposes (in this case up to 20 c.cm.) only isotonic or weakly hypertonic NaCl solutions may be used. If water, hypotonic or strongly hypertonic salt solutions are employed, the red blood cells disintegrate. This is not a true biological hemolysis, but de- pends upon physical basis. 0.85 per cent, saline is most suitable for the majority of erythrocytes (man, rabbit, guinea-pig, ox, sheep). When, instead of an isotonic salt solution, an isotonic sugar solution is made, the red cells are retained in their proper form, but the addition of hemolysin and complement produces no hemolysis. The presence of salt is indis- pensable for hemolysis as well as agglutination Undiluted, unwashed, defibrinated blood if removed sterile can be kept several days in the ice-box. The "Frigo" apparatus is unsuited for this purpose, because the thawing of the frozen blood breaks the capsule of the red blood corpuscle. The deterioration of the preserved blood is recognized by the large hemoglobin content of the serum or the violet color of the blood. Occasionally blood left in an ice-box becomes dark. This is due to the lack of oxygen. When the 5 per cent, suspension is made and thoroughly shaken, the red color returns. Such blood of course is perfectly serviceable. Still, it is best not to keep blood longer than four days. Blood older than that, even if apparently unchanged, possesses a lowered resistance and can give a far higher' titer in hemolysin tests than fresh blood. 2 The rabbit's hemolysin must have been inactivated for one-half hour at 56° C. Dilutions are made with physiological saline. 3. Guinea-pig's complement is obtained by bleeding to death a healthy normal animal. The blood is allowed to flow directly into a centrifuge tube and then to clot; the clear serum is obtained by centrifugalization. For titration of hemolysin it is best to use a constant dose of complement as i c.cm. of a i/io dilution. Complement can be kept for twenty-four hours in the ice-box. When older than this it suffers a distinct decrease in efficiency as complementoid is produced. (See above.) In the "Frigo," comple- ment may be kept for weeks. Stern, however, does not recommend complement pre- served in "Frigo" for use in complement fixation tests, as its affinity for amboceptor is noticeably decreased. One c.cm. of each of the three reagents (each so diluted with saline that the desired dose is contained within i c.cm.) is mixed and 2 c.cm. of 0.85 salt solution is added to make the total volume up to 5 c.cm. The following controls are absolutely necessary. TITRATION OF HEMOLYSIN 149 1. A test showing that hemolysin in strong dosage but without com- plement is ineffective ; 2. A test indicating that without hemolysin complement in the dosage used is ineffective; 3. A test which shows that the NaCl solution is iso tonic. The three reagents must be thoroughly mixed by careful shaking of the tubes which are then placed in the thermostat at 37° C. and hemolysis watched for. The duration of the observation is a matter of personal preference. Only, the length of time must always be mentioned. One must say, for instance, that the titer of this hemolysin is i :8oo, using o.i c.cm. of complement under observation for one-half hour, or it is i : 500 with o. i complement under observation for two hours. The time in which hemolysins work is very different. While many hemolysins of the same titer act in a few moments, others require two hours. The author has made it a rule to read the result after two hours' observation, but he notes the progress of the reaction every one-half hour in order to determine whether it is a slowly or rapidly acting hemolysin. The following chart demonstrates the titration of a hemolysin as a preliminary experiment to the complement fixation method. Result of hemolysis Antigen Amboceptor Comple- ment 0.85 % Saline After i hr. After i hr. After 2 his. I. i c.cm. of 5 % i c.cm. dilution 1:10 i c.cm. dilu- 2 c.cm. complete complete complete sheep's blood tion i : 10 2. " I 100 ** 41 complete complete complete 3- i( i 250 11 " complete complete complete M l( M ,, Incom- Almost complete 4- i 500 plete complete 5- 11 i 750 " " Almost o Incom- complete plete 6. " I IOOO " " 0 Incom- complete plete 7- " I 1500 " " 0 0 Incom- plete 8. I 2000 o 0 o Control I .< I 10 3 c.cm. o o o Control II .< i c.cm. dilu- o o o tion i : 10 Control III 4 c.cm. o 0 0 Determining the end reaction is a source of difficulty for the beginner. Between the extreme "o" i.e., entire absence of hemolysis, where the appearance of the tube corresponds to that of control III representing a suspension of red blood cells diluted with isotonic saline, and the other extreme "complete/' i.e., complete hemolysis, where every trace of cor- puscular elements has disappeared and a fluid looking like dilute red-wine 150 BACTERIOLYSINS AND HEMOLYSINS remains, there are many intermediate stages. These intervening grades of reaction are represented by the terms almost o, incomplete, almost complete, and similar expressions. The meaning of the terms is self-evident. How any particular tube is to be designated is of course a subjective question since the so-called transitional stages are numerous. A few hours after the reaction is ended, a remarkable difference may be noted between the tubes in which hemolysis has occurred and those in which hemolysis has been incomplete or totally absent. In the last men- tioned, the red blood cells have sunk to the bottom and above them remains a clear fluid which consists of pure saline or diluted serum (complement + immune serum) and is colored accordingly. If the supernatant fluid is richer in hemoglobin than that of the corresponding control, it is evident that some of the erythrocytes were hemolyzed and their hemoglobin set free. If the erythrocytes have collected at the bottom apparently in the same quantity as in the control tube, and form there a large deposit, a trace of hemolys is or almost o would be the terms used in reporting the result. Tube 7 after two hours showed incomplete hemolysis, i.e., compared with control III it was noticeably clearer, but not completely transparent. After twenty-four hours there was a small mass of undissolved red blood cells at the bottom of the test-tube and above it a deep red fluid which was only slightly different from that in the tubes where the erythrocytes were completely dissolved. If this sediment should become so small that on shaking only a cloudy turbidity is produced, the result would correspond to the designations "very small sediment," "occasional erythrocytes at the bottom of the tube," or "almost complete hemolysis." In the tubes containing inactive hemolysin without complement (control I, and in complement fixation reactions) hemagglutination can occur because the agglutinins which also exist in the serum remain active. Hemagglutination is recognized by the shaking the sediment: the erythrocytes are not equally distributed, but remain in clumps or strings and soon sink to the bottom again. For many purposes it is desirable to titrate the complement Titration of content of a serum. The method is the same as that used the Comple- in hemolysin titration, with the difference that a fixed amount ment. of hemolysin and varying quantities of complement are employed. The titer of this complement when employed with a hemolysin of i/'iooo strength and allowed to stay in the incubator for two hours would be 0.04 c.cm. The complement content of the serum of a healthy guinea-pig is fairly constant. During illness the titer usually is decreased. Among healthy people the complement titer shows marked individual variations. For hemolysis a definite quantitative relationship between hemolysin and complement is necessary. TITRATION OF COMPLEMENT Antigen Complement Result Amboceptor NaCl after two solution hours i. i c.cm. 5% of sheep's blood. c.cm. hemolysin dil. : 1000. i c.cm. : 10 ( = 0.!). 2. o c.cm. Complete. 2. i c.cm f G7 of sheep's blood. ! c.cm. hemolysin dil. : 1000. 0.8 c.cm. : 10 ( = 0.08). 2.2 c.cm. Complete. 3. i c.cm. 5% of sheep's blood. . c.cm. hemolysin dil. : 1000. 0.6 c.cm. : 10 ( = 0.06). 2.4 c.cm. Complete. 4. i c.cm. 5% of sheep's blood. c.cm. hemolysin dil. : 1000. 0.4 c.cm. : 10 ( = 0.04). 2.6 c.cm. Complete. 5. i c.cm. S °7 of sheep's blood. c.cm. hemolysin dil. : 1000. 0.2 c.cm. : 10 ( = 0.02). 2.8 c.cm. Incomplete. 6. i c.cm. 5% of sheep's blood. i c.cm. hemolysin dil. : 1000. i.o c.cm. : ioo( = o.oi). 2.0 c.cm. o 7. i c.c.m 5% of sheep's blood. i c.cm. hemolysin dil. i : 1000. 3.0 c.cm. o 8. i c.cm. 5 /o of sheep's blood. i c.cm. i : 10 (= = 0.1). 3.0 c.cm. 0 9. i c.cm. 5% of sheep's blood. ii 4.0 c.cm. 0 On the basis of the two titrations outlined above, it is estimated that at least 0.04 c.cm. complement is necessary to activate o.ooi c.cm. of hemolysin. If less complement is used with the same amount of hemolysin, hemolysis does not occur or else it is incomplete. If the quantity of hemol- ysin is increased for instance threefold, then it will be found that 0.02 c.cm. of complement suffices to produce hemolysis. Vice versa with an excess of complement the hemolysin titer of o.ooi c.cm. may be reduced. How- ever, there are narrow limits to this mutual compensatory action. Cytotoxins, Cytolysins. The hemolysin bodies are characteristic and important members of a general class of substances known as cytotoxins, especially investigated by Metchnikoff and his co-workers. Just as the immunization with erythrocytes led to the production of lytic amboceptors which in connection with complement destroyed and dissolved their antigens, so in a similar manner, various substances more or less specific for their antigens have been produced through immunization with leucocytes "Leucocidin," with nerve tissue, "neurotoxin," with spermatozoa, "spermatoxin," and kidney tissue, "nephro toxin." The proof of their action, particularly of neuro-, nephro-, and hepatotoxin is not simple. As all these cytotoxic sera have at the same time a hemolytic action, it is not easy to decide to what extent the changes in the organs observed after the injection of the cytotoxic substances are dependent upon the action of hemolysins. It must further be taken into consideration that none of these sera are absolutely specific for the organ in question. This is not surprising, inasmuch as there are widespread common group charac- teristics (common receptors) among the different organs serving as antigens, and only very few groups of a specific nature. The hopes which, at the beginning, were placed upon the study of cytotoxins particularly with the expectation that they would tend to become diagnostic and therapeutic methods for the treatment of malignant tumors, have as yet been unreal- ized. The entire field of cytotoxins urgently requires further investigation. CHAPTER XIII. THE METHOD OF COMPLEMENT FIXATION. Its principle. Antituberculin. Ehrlich's side-chain theory. Serum diagnosis of syphilis, and diseases caused by animal parasites. It has already been demonstrated that neither bacteriolysis The Question nor hemolysis can take place without the presence of comple- of Multipli- ment. The question therefore arises whether this complement city of Com- is the same in both of these reactions or whether normal serum plements. possesses different complements. In order to solve this, a number of very complicated experiments have been carried out by Ehrlich and Morgenroth, Metchnikoff, and Bordet and Gengou. Ehrlich and Morgenroth endeavored to show that not only do the comple- ments of different animals of the same class vary, but that numerous com- plements exist within one individual serum (conception of the multiplicity of complements). Metchnikoff believed that each serum contained at least two complements, the microcytase and the macrocytase, thus enlist- ing the supporters of a dualistic theory. Bordet and his school, on the other hand, although agreeing with the idea that the complement varies in different animals, deny its multiplicity and contend that any given serum contains but one alexin, or complement — the theory of unity of complement. It would be superfluous to cite all the experimental data supporting these opinions, but nevertheless a review of the classical experiment of Bordet and Gengou which corroborated the existence of only one complement, thus offering the fundamental principle for the establishment of the most impor- tant method of serum diagnosis, namely, complement fixation, would not be out of place. Bordet and Gengou mixed in a test-tube typhoid bacteria The Principles (antigen), inactivated typhoid immune serum (amboceptor) ^enTrfcT anc* norma^ semm (complement). Union of the bacteria and ation. immune serm first took place followed by absorption of, and coalescence with, the bacteriolytic complement contained in the normal serum. As a result, bacteriolysis occurred and the bacteriolytic complement was used up during this process. Bordet and Gengou rea- soned that if the bacteriolytic and hemolytic complements were identical, then in the above mixture of typhoid bacteria, immune serum and normal serum, the hemolytic as well as bacteriolytic complement should be absent, while if the plurality of complement exists, the hemolytic complement should still be present. Accordingly, after a certain interval, washed erythrocytes and inactivated homologous immune serum were added and 152 BORDET AND GENGOU PHENOMENON 153 hemolysis looked for. No hemolysis took place, thereby attesting to the fact that the bacteria in the first part of the test had "fixed" ("held in check") not only the bacteriolytic but also the hemolytic complement. Bordet and Gengou thereupon named this test "complement fixation" or "complement binding" (La fixation d'alexine). With the aid of this experiment Bordet and Gengou were able to prove a number of theoretically important points. They demonstrated that absorption of complement was not necessarily accompanied by bacteriolysis. For example, the anthrax and pest bacteria when mixed with their respective homologous immune sera show no or only very incomplete bacteriolysis. The erroneous conclusion thus reached, to the effect that these sera contained no amboceptors, was disproved by Bordet and Gengou, who demonstrated that, i, these sera contained amboceptors in spite of the absence of bacteriolysis, 2, the complement was absorbed, although no bacteriolysis took place. During the process of immunization, amboceptors were found far more frequently than bacteriolysins. These two terms must not be considered as synonymous. Amboceptor signifies a more generic term, and one must differentiate between amboceptors of cytolytic and non-lytic properties. Whether the difference here really depends upon the different nature of the amboceptor, or upon the construction and constitution of the antigen, is not solved. The fixation of the complement precedes the act of bacteriolysis. The important requirement for the fixation is an antigen which has been sensitized by the attachment of the amboceptor* thus increasing the affinity toward the haptophore group of the complement. Antigen alone, or even amboceptor alone, binds the complement only very slightly or not at all. Whether the zymotoxic (energy) group of the comple- ment manifests its activity (bacteriolysis) or not (absence of bacteriolysis) is materially indifferent for the complement fixation. Through complement fixation, as introduced by Bordet and Complement Gengou, one is enabled to prove the presence of specific anti- Fixation as bodies when the antigen is known or reversely, an unknown a Method antigen provided the specific antibody is given. This method of Serum of serum diagnosis can be widely employed, as the majority of Diagnosis. Dac^eria an(j immune sera (with the exception of pure antitoxic sera)1 when mixed homologously, give a positive reaction — the absence of hemolysis, proving the absorption of complement by the union of the antigen and its specific amboceptor. This reaction is strongly specific. If bacteria are mixed with an inactive heterologous immune serum, or with a heated normal one, not in concentrated form (normal amboceptor), and complement is added, the latter will not be fixed but remains to be taken up by the subsequently added red blood cells, and its immune serum, causing hemolysis. Hemolysis indicates that the mixed bacteria and serum are not homologous, as the complement is left free, and given a chance to unite with the added erythrocytes and hemolytic ambo- ceptor. In the case where the bacteria are known, e.g., typhoid bacilli, the occurrence of hemolysis indicates that the examined serum contains no 1 Even antitoxic sera are said by Nicolle to give complement fixation reactions. 154 METHOD OF COMPLEMENT FIXATION typhoid amboceptors. If the serum is known (e.g., meningococcus serum) the occurrence of hemolysis proves that the bacteria under examination are not meningococci. The absence of hemolysis will in the first case point out that the unknown serum contains typhoid amboceptors, i.e., is a typhoid serum; while in the second case the absence of hemolysis would bear definite evidence in favor of meningococci. The accompanying figures, 15 and 16, represent schematically the positive and negative complement fixation test. Typhoid bacilli (antigen) Typhoid Inactive typhoid serum (typhoid am- boceptor) I Typhoid Complement amboceptor Hemolysin (hemolytic amboceptor) Sheep's red blood cells (2d antigen) Result: Due to the union of comple- Complement ment with the complex typhoid bacillus -f- the typhoid amboceptor, hemolysis does not take place. FIG. 15. p V7 Blood cell V Hemolytic A amboceptor A Typhoid bacilli Typhoid bacilli (antigen) Inactive cholera serum (cholera ambo- ceptor) Complement I Cholera Hemolysin amboceptor Sheep's red blood cells Result: The complement unites with the hemolysin and the sheep's red blood cells thus producing hemolysis. FIG. 16. V7 Red blood cell VI Hemolytic amboceptor Complement Gengou further showed that not only cellular antigens can stimulate the formation of amboceptors, but that during the course of immunization with proteids in solution (milk, serum, etc.), complement binding ambo- ceptors are also formed in addition to the precipitins. Citron has there- fore proposed the term " antigenophile," to designate the "cytophile" group of the amboceptor. Widal and Lesourd were the first to make practical application of the complement fixation property. They found that the Bordet-Gengou re- action could be obtained far more frequently and earlier with the serum of typhoid patients than the agglutination test. Nevertheless, this entire complement fixation method remained unheeded for several years. PRACTICAL APPLICATIONS OF THE COMPLEMENT-FIXATION TEST 155 Moreschi (at Pfeiffer's institute), while conducting some theoretical studies con- cerning the nature of anticomplements, i.e., such substances which tend to neutralize the action of complements, discovered anew, that by the mixture of a soluble proteid with its antiproteid serum the existing complement disappeared. This, as has been seen, can be explained by the presence within the immune serum of bodies similar to Gengou's amboceptors. Moreschi, however, stated that the complement disappeared because it was thrown to the bottom mechanically, by the occurrence of precipitation. Such a physical explanation for the complement fixation reaction led a number of authorities to the belief that the positive Bordet-Gengou reaction was in reality no amboceptor action, but a result of a similar precipitation process. Wassermann and Bruck, Liefmann, Wassermann and Citron, and later on Moreschi himself realized that this physical explanation was incorrect, inasmuch as complement fixation took place even if all precipitation was prevented by heat or other 'influences. Furthermore, complement binding of an unspecific nature can be produced by the mixture of glycogen or peptone with serum, a procedure wherein surely no precipitation plays any part. Finally Moreschi showed that there were strongly precipitating sera which nevertheless did not exhibit the Bordet-Gengou phenomenon. Thus was definitely established that the complement fixation was entirely independent of either bacteriolysis or precipitation. Following Moreschi's researches, Neisser and Sachs continued Gengou's studies and advised this demonstration of the proteid amboceptors as a control to the precipitation method for the differentiation of proteids. Its action is so much finer, and more delicate than the precipitin test that even the minutest traces of proteid can be recognized. With the encouraging results of Neisser and Sachs in mind, Wassermann attempted by the use of highly immune antibacterial serum to discover any soluble bacterial proteids which may exist in the blood, derived from the respective bacteria invading the organism at the onset of an infection. Practical application proved that not enough of these proteids existed free in the circulation, but that they were probably bound by the tissue cells. Wassermann and Bruck then employed the complement fixation test with the idea of demonstrating the existence of the respective antigens in the diseased organs. Tuberculous glands and lungs served as material for this experiment. They were able to obtain complement fixation when an extract of tuberculous organs as antigen was mixed with a tuberculous serum (manufactured by the Hochst Farbwerke) . If instead of the latter, the ' serum from tuberculous individuals was substituted, no positive complement fixation reaction was obtained. On the other hand, the reaction was given if the human tuberculous serum employed Antituber- came from an individual who had received therapeutic inocu- culin. lations of tuberculin. In other words, the serum of treated individuals contained, in contrast to the untreated ones, amboceptors against a soluble tuberculous substance also present in the extract of tuberculous glands. Wassermann and Bruck identified this substance as tuberculin, because the sera of the treated individuals gave the same positive results if a solution of old or new tuberculin was used in- 156 METHOD OF COMPLEMENT FIXATION stead of the extract of tuberculous organs. Thus, the latter contained tuberculin while the sera of the tuberculin- treated individuals contained amboceptors designated by Wassermann and Bruck as "antituberculin." The name antituberculin has not been a well chosen one, because it creates the impression among many as being an antitoxin. It is better to speak of it as antituberculin amboceptors. Since, according to Wassermann and Bruck these antituberculin ambo- ceptors were not supposed to be formed spontaneously in tuberculous indi- viduals, but only in those treated with tuberculin their demonstration could be of no apparent diagnostic value. On the other hand, their exist- ence greatly furthered the understanding of Koch's tuberculin reaction, as most tuberculous individuals who had antituberculin amboceptors in their serum did not respond to the subcutaneous injection of tuberculin. Wassermann and Bruck, moreover, showed that a mixture of tuberculin with an extract from tuberculous organs bound complement. From this they concluded that the extract likewise contains antituberculin ambo- ceptors. Thus reasoning they developed their tuberculin theory. The difference in the reaction observed in a normal and tuberculous Tuberculin individual after inoculation with tuberculin, can be fully explained by the Theory of presence of antituberculin amboceptors in the tuberculous focus. By Wassermann virtue of their specific affinity, the amboceptors attract the injected and Bruck. tuberculin toward them. The tuberculin and antituberculin unite, and absorb the complement from the circulating blood stream, since the complementophile group of the amboceptor is free and unbound. By virtue of the fresh complement which is an actively lytic ferment, and the attracted leucocytes, a partial destruction and casting off of the tuberculous focus results. Upon this depends the therapeutic effect of the tuberculin. During a prolonged treatment with tuber- culin, the body produces an excess of antituberculin amboceptors so that finally some appear free within the blood serum. When this is the case the tuberculous organism loses its power to react toward tuberculin, as the latter is neutralized in the blood- stream at a point away from the local focus. No therapeutic effect is any longer obtained from the tuberculin injections, so that they can, for a time, be suspended. The aim of tuberculin therapy should be to work with small doses so that only a focal reaction is obtained and the hyperproduction of antituberculin amboceptors be post- poned as long as possible. Numerous exceptions were at once taken to this theory and its experimental data, the most important of which can here be mentioned. " Summier- d" (E ^6^ an(* Nakayama disagreed with the proof of the existence of "anti- tuberculin" in the organ extracts, on the basis that Wassermann had overlooked the effect of a summation of antigen. This is best explained taken on .„„,.. as follows: Complement is bound not only by antigen + amboceptor, but also by large doses of antigen itself dependent upon the normally . present amboceptors existing in the serum employed for complement Antigen.) COMPLEMENT FIXATION DUE TO SUMMATION OF ANTIGEN 157 Old tuberculin Complement Erythrocytes Hemolysin Result 0.05 O. I 0.15 O. I O. I O.I i c.cm. 5% i c.cm. 5% i c.cm. 5% Twice the hemolytic titer. Twice the hemolytic titer. Twice the hemolytic titer. Hemolysis. Hemolysis. No hemolysis. 0.15 old tuberculin is thus sufficient of its own accord to bind complement. In their experiment, Wassermann and Bruck found that, for example, Tuber- culin Extract of tuberculous organs Comple- ment Erythrocytes Hemolysin Result O.I O I O.I O.I O I i c.cm. 5% i c.cm. 5% Twice the hemolytic dose. Twice the hemolytic dose. Complement fixation. Hemolysis. O I O I i c cm 5% Twice the hemolytic dose Hemolysis. This, however, in no way proves the existence of "antituberculin" in the extract of tuberculous organs, as it is perfectly possible and even probable that o.i of the organ extract contains 0.05 c.cm. at least of tubercle bacillus substance (tuberculin) which, when added to o.i of tuberculin used for antigen, is sufficient to give an amount of tuberculin perfectly capable, as has been seen, of binding complement by its own activity. In order to overcome this possibility one must work with such small but at the same time maximum amounts of antigen and antibodies, that at least double the quantity of each of these reagents does not, of its own accord, bind complement. For tuberculin this is estimated as follows: Tuberculin Complement Hemolysin Erythrocyte Result O. 2 o. 2-X Hemolytic dose. c.cm. 5% No hemolysis. 0.18 o. 2 X Hemolytic dose. c.cm. 5% No hemolysis. 0.15 o. 2 X Hemolytic dose c.cm. 5% No hemolysis. • d. 14 0. 2 X Hemolytic dose. c.cm. 5% Incomplete hemolysis. 0. 12 0. 2 X Hemolytic dose. c.cm. 5% Complete hemolysis. O. I o. 2 X Hemolytic dose. c.cm. 5% Complete hemolysis. Twelve-hundredths is the maximum non-binding or hemolytic dose. For the complement fixation test where the object is to demonstrate anti- tuberculin amboceptors, the maximum amount of antigen to be used is therefore 0.06 T. or one-half of the maximum non-binding dose. In the same way the hemolytic dose, and the dose of the antibody should be estimated. 158 METHOD OF COMPLEMENT FIXATION Organ extract Complement Hemolysin Erythrocyte Result 0. 2 o. 16 O. I O.I O. I O. I . 2 X Hemoly tic dose. !2XHemolytic dose. 2 X Hemoly tic dose. i c.cm. 5% i c.cm. 5% i c.cm. 5% o Complete hemolysis. Complete hemolysis. The non-binding dose is 0.16. The amount, however, to be employed in the complement fixation test must be o.c8 c.cm. of organ extract. If on mixing 0.06 T. and 0.08 extract, complement fixation still appears, then this summation of antigen can be disregarded and an antigen antibody reaction must be considered. For, even granting that 0.08 of extract does for its greater part, e.g., 0.06 at the most, consists of tuberculin, then this amount + 0.06 of the tuberculin in the antigen only makes 0.12 of tuber- culin, a quantity not sufficient to fix the complement. De facto, comple- ment fixation does occur when the above test is carried out with proper dosage, so that most probably it is occasioned by the biological antigen antibody reaction. As a general rule for all complement fixation tests, the dose of each ingredient employed should never be more than 1/2 of its maximum quantity that does not of itself bind complement. A second exception, taken by Weil and Nakayama as well as Other by Morgenroth and Rabinowitsch relates to the activity of the Exceptions, complement when it combines with tuberculin and anti- tuberculin. They claim that by this union the complement's lytic func- tion is entirely lost. Morgenroth and Rabinowitsch even go so far as to deny the existence of anti tuberculin in the blood of tuberculous individuals. The author also undertook a minute study of this question and came to the following definite conclusion. — There are some tuberculous individuals who spontaneously develop antituberculin amboceptors, a fact to be ex- pected because it has for a long time been known that on and off tuberculin can be liberated in the organism of tuberculous individuals. As a natural consequence antibodies will be formed, and most probably by those tissue cells in the neighborhood of the liberation of the tuberculin, i.e., the focus of infection. Before proceeding, however, to the author's conception of the tuberculin theory it is necessary to review Ehrlich's principles of immunity upon which the ideas of antibodies and their specificity are based. Ehrlich's idea of the biological structure of cells is that they Ehrlich's consist of two parts, a central functionating radicle ("Leist- Side Chain ungskern") upon which depends the specialized activities of Theory, the cells, as for example, a glandular or nerve cell, and a multiplicity of side chains or receptors (a term borrowed from the chemistry of the benzol group), by means of which the cell enters into chemical relation with food and other substances brought to it by the circu- lation. These receptors are exceedingly numerous, as the nutritive sub- stances upon which the cell depends for its maintenance are very varied. Besides these general receptors the special cells also have different and special side chains; then, too, there exist very great quantitative differ- ences among the latter; and finally it must be added that the selective activity of the cells depends upon the variability of these receptors. When an infection occurs, pathological material is brought to the cell bodies instead of physiological normal substances. Certain of these poi- sonous products find suitable receptors in all of the cell groups, others fit only into distinct groups of cells, while a third class are not taken up at all. The organism which possesses no receptors for any of the pathological agents cannot assimilate any deleterious substances and is therefore im- mune. Lack of amboceptors is therefore a natural form of immunity. The organism, having only a special group of cells for the reception of cer- tain pathological matter, will make use of these cells for the binding and assimilation of the toxic material. For example, the nerve cells alone have receptors for tetanospasmin ; no matter how or when the poison is intro- duced into the organism the nerve cells will absorb it. As this toxin is poisonous for the central atom group (Leistungskern) of the nerve cell, the latter is destroyed. The union between the nerve cell receptors and the tetanospasmin toxin is only the preliminary act for the cell destruction; the actual death of the cell being caused by the action of the toxophore group of the poison upon the functional radicle of the cell. If, however, such receptive side chains are possessed not only by the brain but also by other cells, e.g., connective-tissue cells, the tetanospasmin will in part be bound by the latter. The toxophore group of the toxin does not have any harmful effect upon the functional radicle of these cells, and thus no toxic effects will be incurred by the union, and the nerve cells remain unaffected. The number of receptors which cells possess for tetanospasmin, for example, are limited and after their junction with the tetanospasmin, are rendered useless and inactive. By the normal repa.rat.ive mechanism of the body, new receptors are generated. This reparative process does not as a rule stop at a simple replacement of lost elements, but according to the hypothesis of Weigert tends to overcompensation. The receptors eliminated by toxin absorption are reproduced in an excess of the simple physiological needs of the cell. Continuous and increasing dosage of the toxin soon leads to such excessive production of receptors that the latter find no more room to be attached to the cell, but are cast off and circulate free in the blood. They still, however, retain their property of being able to combine with tetanospasmin. l6o METHOD OF COMPLEMENT FIXATION If such, an organism is injected with tetanospasmin the latter toxin is bound by the free receptors in the serum, and thus the respective " sessile" receptors attached to the cells are prevented from coming in contact with the poison. Inasmuch as the free receptors possess no functional radicle which can be injured, the toxin remains entirely innocuous for the indi- vidual. Such protective bodies lend to the organism its attained immunity and are known as antitoxins. Their function can be compared to lightning rods. v. Behring well expresses their action when he states that the same elements, which attached to the cells render the body susceptible to toxic substances, when circulating freely in the blood serve to protect it. The antibodies against toxins and ferments are of the simplest form. They possess only a binding group which has an affinity toward the hapto- phore group of the toxins and ferments. They, therefore, belong to the class designated by Ehrlich as "haptines" of the first order. To the haptines of the second order belong the agglutinins and pre- cipitins. They possess besides a haptophore group also an agglutinophore or precipitinophore group by virtue of which agglutination or precipitation takes place. Belonging to the haptines of the third order are the class of amboceptors which have in addition to the haptophore group also a complementophile group for their union with the complement. These hypotheses of Ehrlich greatly simplify the explanation Citron's of many serum reactions as well as many of the phenomena Tuberculin associated with the action of tuberculin. In all probability Theory, the healthy cells which exist in the tuberculous focus and which are capable of reaction, produce the antituberculin. Christian and Rosenblatt offered experimental evidences for this state- ment. They demonstrated that tuberculous guinea-pigs, in whom anti- tuberculin was produced by tuberculin injections, showed a diminution of antituberculin in the blood when tuberculous glands were removed by operation. The antituberculin production by the cells is a transitory action arising only when tuberculin has spontaneously or artificially reached the circu- lation. Following this stage of activity there comes a period of quiescence during which no free antituberculin can be found in the serum. The cells, however, are supplied with a great many more sessile receptors than usually: they possess a higher affinity toward tuberculin and produce antituberculin much more readily than normal cells. This also explains why the smallest amounts of tuberculin produce a reaction in tuberculous and not in the normal individuals. In the former, the cells in the zone surrounding the tuberculous focus are abundantly supplied with receptors, so that on the injection of tuberculin, its action CITRON'S THEORY FOR THE TUBERCULIN REACTION 161 appears almost concentrated at this point. Occasionally the sessile recep- tors are* relatively scarce and the first injection excites no reaction. By the time of the second or third inoculation these sessile amboceptors have so increased that a positive reaction is apparent when the same or even a smaller dose is injected. This phenomenon of increased sessile receptors explains the reappearance of subsided subcutaneous, cutaneous, or oph- thalmo reactions after renewed injections of tuberculin. To recapitulate the biological phenomena associated with a positive tuberculin reaction, it may be said that the tubercle bacilli, or portions of their body substances existing in the infected focus, stimulate the adjacent cells to produce a great number of sessile receptors. When the tuberculin is injected for the first time, these sessile receptors at once take up the tuberculin and as a result, the production of an ti tuberculin in the focus is further stimulated. A part of the tuberculin has already been attracted by the receptors of the cutis or subcutis cells (intracutaneous reaction) or the cells of the mu- cous membrane (ophthalmo reaction) and here too has stimulated the production of antibodies (an ti tuber culm). It is only a quantitative differ- ence in the number of receptors which actually differentiates a normal from a tuberculous individual. Thus is explained that even in non-tuberculous individuals a local reaction may be obtained if the dose of tuberculin in- jected is large enough; a focal reaction, however, will be given only by a tuberculous subject. The greater the number of sessile antituberculin receptors that have been formed in the tuberculous focus, the greater becomes the affinity of these cells toward the tuberculin; so that with the second, third, and subsequent tuberculin injections, focal reactions (i.e., antituberculin productions) are more easily stimulated. As for the origin of the fever, it is probable that a pyrotoxic substance is formed by the union between tuberculin, antituberculin and comple- ment. This poison first isolated in vitro by Citron will again be referred to and belongs to the class of anaphylotoxins (Friedberger) or toxopeptids (M. Wassermann and Keyser). Finally the antituberculin receptors become so numerous that they are detached from the cells and become free receptors. This period, how- ever, is only transitory, as is corroborated by the difficulty connected with the demonstration of these antibodies in the focus. This free antituber- culin combines with the tuberculin (spontaneously formed or injected) and attracts the complement, or the complement producing phagocytes. Uncombined complement has no effect on the tissues. It is different, how- ever, with the phagocytes. These can without any additional help act directly upon the infected focus. If the tuberculin treatment is continued, * A sessile receptor is one which is still attached to its cell and not yet free in the blood. 1 62 METHOD OF COMPLEMENT FIXATION a period arises during which the antituberculin bodies are so greatly ac- cumulated in the local focus that they ultimately escape into the blood stream. This freely circulating antituberculin neutralizes any freshly injected tuberculin, so that such patients become refractory against even the largest amounts of it. (Tuberculin immunity.) Tuberculin immunity is not, however, in all cases to be identified with a strong antituberculin content in the serum as is demonstrated by the complement fixation method. For example, it is very difficult to produce antituberculin bodies by treatment with S. B. E., although by its use an immunity against B. E. is easily attained. The question as to how great a role the antituberculin bodies play, and their exact interpretation is very complicated. The present status of our knowledge may be expressed as follows. The administration of the various tuberculin preparations to tubercu- lous patients results in the formation of antibodies within their serum, which with the respective tuberculin as antigen will give the phenomenon of complement fixation in vitro. The different tuberculins are not equally efficient as antigens. Thus B. E. is the best stimulant of antibodies. Furthermore the antibodies obtained from the soluble tuberculin prepa- rations are not identical with those from the insoluble products. Some sera fix complement only with old tuberculin, others react only with new tuberculin. The tuberculous serum of Meyer and Ruppel contains be- sides these two antibodies another group which gives the complement fixation test with an alcoholic extract of tubercle bacilli as antigen (Citron and Klinkert). It may be assumed that with other antigens, other com- plement binding antibodies will be discovered. The presence of these antibodies is of itself no criterion for the existence of a tuberculin immunity. There are tuberculous subjects who are not susceptible to tuberculin and at the same time possess no in vitro demonstrable anti- bodies; reversely, there are very highly susceptible tuberculous individuals with antibodies in their serum. The explanation for the last class of cases has been furnished by Citron. The author demonstrated that the anti- tuberculin contained within the serum in certain instances raises the sus- ceptibility against tuberculin. Thus there is a hypersusceptibility of a humoral form analogous to serum anaphylaxis, besides the hypersensi- tiveness depending upon the increased number of sessile receptors. To offer an explanation for the first class of cases (i.e., tuberculous sub- jects with no antituberculin and not susceptible to tuberculin) it has been assumed that besides complement binding agents there are also directly neutralizing or antitoxin-like bodies within the serum. Pickert and Low- enstein could demonstrate that the serum of some tuberculous patients had the property when mixed with tuberculin to so neutralize the latter i63 that it could no longer be used for the cutaneous reaction. They named this antibody "antikutine," and proved its existence as follows: Twenty days after the last tuberculin injection the serum of the patient is withdrawn and mixed with tuberculin in i, 2, 5 and 10 per cent, tuber- culin serum mixtures. These are kept in the incubator for two hours, then in the ice-box for twenty hours and then used for the cutaneous reaction. The abrasions made with the v. Pirquet borer must be at least 3 cm. dis- tant from each other, intervened by control spots. By gently stretching the skin of the forearm one can apply ten reactions over this area. Only a distinct papule formation should count as a positive reaction and ob- servation should be continued for six days. Just as antituberculin amboceptors may be found in the serum of patients who have not had any specific treatment, so also can these antikutine bodies originate spontaneously. Marmorek described still another antibody, found in a horse immunized with living young tubercle bacilli. This serum when mixed with the urine from patients with active febrile tuberculosis will fix complement. Exam- inations by Citron and Klinkert showed that here one is probably dealing with an antibody brought about by immunization against altered tuber- culous tissue. In former times a negative tuberculin reaction after a prolonged treat- ment was stamped as a cure of the tuberculosis, a fact obviously incorrect; for no matter how successful the tuberculin therapy may be, it cannot always be considered as a complete curative procedure. In general, that method should be adopted which makes the individual non-susceptible to the largest doses of tuberculin. It was found in practice that those patients having the greatest amount of antituberculin in their serum usually offered a better prognosis. The experiences gained by the employment of the complement Serum fixation test in tuberculosis lead to its application in the study Diagnosis of of syphilis. The difficulties in this disease were greater, inas- Syphilis. much as there were no bacteria or preparations like tuberculin which could be used as antigen. Syphilitic human organ extracts were employed, with the idea that these would contain the specific virus. The serum of monkeys previously immunized with such extracts, when mixed in vitro with the latter, gave complement fixation. ' This experi- ment is not, however, conclusive; the positive reaction may be due to anti-human proteid amboceptors produced at the same time by the injection of the human serum contained in the organ extract. The experiment was changed and the syphilitic organ extracts from apes were used so as to exclude this error. Even in this way complement fixation was attained. Later on it was found unnecessary to inject the monkeys with the extracts since after ordinary infection their serum would give complement fixation. In this manner it was almost definitely established firstly, that these extracts contained 164 METHOD OF COMPLEMENT FIXATION a substance specific for syphilis which could with most probability be considered a luetic antigen, and secondly that infected apes possess antibodies against this antigen. The next step was to try the reaction in man. The first experiments of Wassermann, Neisser, Brack and Schucht did not give the hoped-for returns. Although the reaction was obtained with human serum, the percentage of positive results was so small (see next chart) that its practical value as a means of diagnosis offered no great help. Only in general paralysis did the expectation seem promising. In about 80 per cent, of all cases Wassermann and Plaut were able to demonstrate the luetic anti- bodies in the cerebrospinal fluid. Schiitze's experiments in tabes led him to the same findings. Citron has obtained a much smaller percentage of positive reacting cerebrospinal fluids in tabes. As it seemed that the means of diagnosis was not to be established by the demonstration of the syphilitic antibody, Neisser and Bruck believed that better results may possibly be achieved by the discovery of the luetic antigen in the serum through complement fixation. This attempt too was unsuccessful. No antigen could be found, but the extracts of red blood cells from syphilitic individuals when mixed with the serum of highly immunized monkeys gave a positive complement fixation. Neisser and his co-workers concluded therefrom that the erythrocyte extract contained the luetic antigen. Citron soon demonstrated that the extracts of normal individuals gave a similar reac- tion and what was more important, that this so-called blood antigen existed in the blood entirely uninfluenced by mercurial treatment. Since these experiments, not much importance has been attached to this reaction. Meanwhile the author working at the Kraus clinic proved by a large series of experiments that luetic antibodies were present in almost all cases of lues. The reaction is dependent upon two rules. The First. — The longer the syphilis virus has acted upon the organism and the more numerous its recurrent manifestations have been, the more frequently will a positive reaction be obtained and the stronger will the antibody content of the serum be. The Second. — The sooner a proper mercury therapy is instituted, the more often it is repeated, and the shorter the interval since the last treat- ment, the smaller will the antibody content of the serum be and the greater the possibility of a negative reaction. These points were soon corroborated by numerous other workers in the field, so that at the present day they can be taken as absolute facts. The following chart will explain some of the statements aforementioned. WASSERMANN REACTION IN LUETIC INFECTIONS First period Second period *,£ 'Si ^ T3 "-J3 T3 Citron d 3 £ 3 q ^9 e M 8 z £ ce c3 rrt CJ <-! £ Wassermann, ! ser, Bruck & Scl Wassermann ; Plaut J T3 § 4 G3 ^ Morgenroth ; Stertz Schiitze I I % % Incomplete 0.6 c.cm. luetic extract i c.cm. 5% o 0.4 c.cm. luetic extract i c cm 5% o 0.2 c.cm. luetic extract i c cm ^ % o The test proves that 0.4 c.cm. of extract is not able to bind o.i c.cm. of complement. That 0.8 c.cm. of lues extract causes only an incomplete hemolysis, while 0.6 c.cm. produces no hemolysis whatever, is explained not by its lessened tendency of binding complement, but by the greater amount of hemotoxin which 0.8 c.cm. possesses. Keeping in mind the rule of Weil and Nakayama about the summa- tion of antigen, this particular antigen will be employed in the quantities of 0.2 c.cm. and o.i c.cm. (0.2 being 1/2 of 0.4 c.cm. which is the largest amount that does not of itself fix complement) . b. Examination and titration of 4 luetic sera by Citron's method (see Plate II). The technical details of the test are as follows: Three test-tubes are assigned for each test and placed in a test-tube rack. The name of the patient is written upon the first of these tubes. Another rack contains one tube for each patient and labelled accordingly. In addition there is an " antigen tube," which is placed in the first rack at the end of all the other tubes; also a "normal extract," a "system," "complement," and "blood control tube," which are placed in the second rack. The amount of syphilitic antigen required for the entire work is calculated as follows. For each test 0.3 of antigen is required; for five cases (in- cluding controls) 1.5 c.cm. are needed + 0.4 c.cm. for the antigen control tube =1.9 c.cm. in all, or 2.0 c.cm. in round numbers. This amount is diluted with normal salt solution in the proportion of i : 5 so that 8 c.cm. of saline are added ( — 2 : 10) ; i c.cm. of this dilution contains 0.2 antigen and 1/2 c.cm. contains o.i antigen. The first tube (i, 4, 7, ic, etc., in diagram) of every test therefore receives i c.cm., the second i8o TECHNIQUE OF THE COMPLEMENT FIXATION METHOD tube 1/2 c.cm<, the third tube nothing, the antigen tube (tube 19) 2 c.cm. Physiolog- ical salt solution is added to make up i c.cm. in each tube; first tube nothing; second tube 1/2 c.cm.; third tube i c.cm. of saline. The normal extract required for the tubes in the second rack is similarly estimated, 0.2 c.cm. is needed for each test, 5 (tests) Xo.2 = 1.0+0.4 fc>r the antigen control tubes = 1.4 or 1.5 c.cm. in round numbers. For purposes of dilution i : 5, 6 c.cm. of salt solution are added and i c.cm. ( = 0.2 of extract) placed into each of the tubes (20 to 26) and 2 c.cm. into the normal extract control test-tube (tube 27). In this series also, salt solution is added to make up equal quantities of i c.cm. : first tube nothing, second tube 1/2 c.cm., third tube i c.cm.; antigen tube (27), nothing; system (28), complement (29) and blood (30) control tubes each i c.cm. The second ingredient of the test is next added, i.e., the respective serum. This is not diluted but added directly; 0.2 c.cm. into the first tube of each test; o.i c.cm. into the second; 0.2 c.cm. into the third tube; also 0.2 into the control series of tubes labelled with the patient's names in the second rack. Salt solution is again added to make up to the equal quantity of 2 c.cm. in each tube, thus: 0.8 c.cm. into first, 0.9 c.cm. into the second, 0.8 c.cm. into the third tube, and 0.8 c.cm. into the control series; nothing into the antigen tubes, i c.cm. into system, complement, and blood control tubes. The addition of complement follows next. Each tube, except the blood tube, receives o.i c.cm. of complement. Thus the tubes are counted and if, for example, nineteen tubes are present 19X0.1 c.cm. complement is taken, or in round numbers 2 c.cm. Complement is always diluted i : 10, or 2 c.cm. complement + 18 c.cm. saline, so that each tube except the blood tube (30) receives i c.cm. of this diluted complement. Tube (30) receives i c.cm. of saline instead. All tubes are then carefully shaken and the racks placed in the incubator for one hour. During this time, the hemolysin and washed red blood cells are properly diluted. The red blood cells are made up in a 5 per cent, suspension of which each tube will receive i c.cm. Thus in the present test there are thirty tubes, requiring 30 c.cm. of blood suspension; since i c.cm. of washed blood when diluted i : 20 will supply twenty tubes, for 30 c.cm. about i 1/2 c.cm. of blood will be required, or 2 c.cm. will make 40 c.cm. of a 5 per cent, blood suspension. Luetic extract Serum. Comple- ment Hemoly- sin;ic.cm. Sheep's blood Result of hemolysis I. O. 2 o. 2 Ser. I. Tabes un- 0. I i : looo i c.cm. 5% No hemol- treated; without luetic ysis. history. 2. O. I o. i As above. 0. I i : looo i c.cm. 5% No hemol- ' 4- + + + ysis. 3- 0.21 As above. O. I i : 1000 i c.cm. 5% Complete I hemolysis. 4. 0. 2 0.2 Ser II. Secondary O. I i : 1000 i c.cm. 5% No hemol- lues untreated. ysis. 5- o.i o. i As above. O. I i : looo i c.cm. 5% Incomplete hemolysis. ' ' 6. ... o. 21 As above. 0. I i : 1000 i c.cm. 5% Complete hemolysis. , TITRATION OF LUETIC SERA 181 Luetic extract Serum Comple- ment Hemoly- sin; ic.cm. Sheep's blood. Result of hemolysis 7. 0.2 0.2 Ser. Ill Tabes. O. I i : looo i c.cm. 5% No hemoly- Many inunction courses sis. 8. o.i o. i As above. 0. I i : looo i c.cm. 5% Complete • 4-4- hemolysis. \ \ 9. ... o. 2 As above. O. I i : looo ic. cm. 5% Complete hemolysis. IO. O.2 0.2 Ser. IV Gallstones. 0. I i : looo i c.cm. 5% Trace of bind- Lues seventeen years ago. ing; almost but Much treatment. No not quite com- symptoms for ten years. plete hemolysis. 4- II. O.I o.i As above. O.I i : looo i c.cm. 5% Complete hemolysis. 12. ... o. 2 l As above. O.I i : looo i c.cm. 5% Complete hemolysis. 13. 0.2 0.2 Negative control 0. I i : looo i c.cm. 5% Complete | serum (carcinoma hemolysis. hepatis). • — 14- 2.J 0.2 l As above. 0. I i : 1000 ic. cm. 5% Complete hemolysis. ' 15. o.i o. i Strongly positive 0. I i : 1000 i c.cm. 5% No hemol- control serum. (Lues ysis. maligna). • + + -h + 16. ... o. 2 As above. O. I i : 1000 i c.cm. 5% Complete hemolysis. 17. 0.2 o. 2 Weakly positive con- O. I i : looo i c.cm. 5% Incomplete trol serum. Primary hemolysis. lesion. \ j. 18. ... 0.21 Weakly positive O. I i : 1000 i c.cm. 5% Complete i control serum. Pri- hemolysis. mary lesion. 10. 0.4 O. I i : 1000 i c.cm. 5% Complete hemolysis. V • v* T^ Normal extract. 20. 0.2 o. 2 Serum I. O.I IOOO i c.cm. 5% Complete hemoiysis. 21. 0.2 o. 2 Serum II. O. I IOOO i c.cm. 5% Incomplete hemolysis, 22. O,2 o. 2 Serum III. O.I IOOO i c.cm. 5% Complete hemolysis. 23. 0.2 o. 2 Serum IV. O.I IOOO i c.cm. 5% Complete hemolysis. 24. 0.2 0.2 Negative control O. I IOOO i c.cm. 5% Complete hemolysis. serum. 25. 0.2 0.2 Strongly positive O.I I IOOO i c.cm. 5% Complete hemolys . control serum. 26. 0.2 0.2 Weakly positive O. I I IOOO i c.cm. 5% Complete hemolysis. control serum. 27. 0.4 O. I I 1 IOOO i c.cm. 5% Complete hemolysis. 28. ... O. I i : 1000 i c.cm. - . *--. * * - ";.-•• Complete. Complete. 0.2 c.cm. o. 5 c.cm. 0.5 c.cm. 0.5 c.cm. Complete. 0.5 c.cm. o. 5 c.cm. 0.5 c.cm. Complete. o. 5 c.cm. 0.5 c.cm. o 0.5 c.cm. 0.5 c.cm. 0 o. i c.cm. of a posi- tive serum. 0.5 c.cm. o. 5 c.cm. 0.5 c.cm. o. i c.cm. of a nega- o. 5 c.cm. tive serum. 0.5 c.cm. 0.5 c.crn. ! Complete. only recently. A negative reaction does not exclude gonococcus infection. If the disease is limited to the anterior urethra, a positive reaction is not obtained; probably the absorption of toxins is insufficient to stimulate the production of antibodies. The positive reaction does not entirely disappear until seven or eight weeks after the patient is apparently cured. If it still persists, one must be suspicious of an active focus somewhere. The complement fixation test should be of special usefulness in chronic cases where the bacteriological isolation of the gonococci is difficult, in gynecological conditions and in differential diagnosis of joint affections (Schwartz-McNeil, American Journal of Med. Sciences, May, 1911, Sept., 1912, Dec., 1912). That the complement fixation test can also be performed Typhoid in typhoid fever was first proven by Bordet and Gengou. Fever. They used as antigen a suspension of typhoid bacilli in normal saline, and the serum from a convalescent patient. Since then (1901) numerous contributions referring to this subject have been published, but the merit of the test is variously interpreted. (Widal and Lesourd, Ludke, Leuchs, Schone, etc.) The editor believes that to a great degree the variability in the results can be accounted for by the different antigens employed. Similar to the findings with the gonococcus complement fixation test by Schwartz and McNeil, Garbat has proven that a highly poly valent typhoid antigen (made from numerous COMPLEMENT FIXATION TEST IN TYPHOID FEVER 1 93 strains) is absolutely essential. The serum from a typhoid fever patient gave a strong complement fixation test with an antigen made up from his own typhoid bacteria isolated from the blood but did not react with a similar antigen made from seven other different strains. The antigen is prepared like the artificial aggressins of Citron, by growing the bacteria on agar for 24 hours. The growth is washed off in a very small quantity of sterile distilled water; kept at 6o°-jo° C. for 24 hours (Leuchs); after this the emulsion is shaken thoroughly with glass beads for 24 hours and then centrifugalized until the supernatant fluid is absolutely clear. The antigen is titrated as usual (see Wassermann reaction, gonococ- cus fixation test). The antisheep system is used. On testing the bloods of thirty-six patients in different stages of the infection with such an antigen prepared from twenty-eight different strains, the editor found a positive result in all but one instance. [This patient died before the test could be repeated.] In ten cases two or three examinations were necessary before the reaction became positive. For the sake of comparison, agglutination tests and blood cultures were made at the same time. No distinct relationship between the three could be dis- covered. In several instances the complement fixation test appeared earlier than the Widal or even before the blood culture became positive. (In one, confirmed by autopsy, as early as the end of the first week.) As a general rule, however, the complement fixatives appear later during the disease, at a time when the bacteria have disappeared from the circulation. Thus, this method becomes of importance as corroborative of the Widal test. The positive reaction becomes stronger during convalescence and persists for several months after. The exact clinical value of the reaction and its specificity require further statistics. One can assume, however, that almost all typhoid fever patients develop complement fixatives sooner or later, and that these can be detected if repeated examinations with a sufficiently poly- valent antigen are undertaken. VI. The Differentiation of Proteids by the Method of Neisser and Sachs. The technique employed here varies only in a few details from the method advanced later by Wassermann and Bruck for the diagnosis of bacterial infections. Neisser and Sachs do not employ a constant amount of complement (o.i), but first titrate the complement against the hemolysin in double its minimum hemolytic dose. For the test one and a half to two times the smallest amount of complement is necessary. The hemolysin consists of the serum of a rabbit immunized against ox's blood. This hemolysin acts both for ox's and sheep's erythrocytes. The amount of antiserum (for example antihuman serum) used for the test, is influenced by two factors. 13 194 TECHNIQUE OF THE COMPLEMENT FIXATION METHOD 1 . An excess of antiserum can interfere with the fixation of complement. 2. The antiserum if used in large quantities can bind complement of its own accord, without the addition of the human serum. It is therefore best to ascertain by titration the smallest quantities of antiserum which may satisfactorily be employed, as the complement fixation test must be sufficiently delicate to determine o.oooi c.cm. of the human serum. Diminishing amounts of antiserum are mixed with o.oooi c.cm. of human serum and o.i c.cm. of complement. A control series is made wherein the human serum is replaced by the same amounts of saline. (The quantity in all tubes should be made uniform by the addition of normal salt solution, but the total amount of fluid in each tube should not exceed 2.3 to 2.5 c.cm.) The tubes are incubated for i hour and the hemolytic amboceptor and red blood cells added. After two hours at 37° C. the results are read. The o.oooi c.cm. of the serum is added in the form of 0.2 c.cm. of a i 12000 dilution. TABLE III. Amounts of antiserum Series A contains the different Series B (control) contains anti- in cubic centimeters. quantities antiserum+o.ooo i c. serum+o. 2 c.cm. physiological cm. human serum (i : 2000.02) + saline-f-o. i of guinea-pig's serum o.i guinea-pig's serum. One hour at 37°. One hour at 37°. +0.001 c.cm. of amboceptor -f-i +0.001 c.cm. of amboceptor + i c.cm. of 5 per cent, ox's blood. c.cm. 5 per cent, of ox's blood. Hemolysis Hemolysis O. I Faint trace. 0.075 Faint trace. 0.05 0 0-035 o • Complete. 0.025 o 0.015 Trace. O.OI Slight. o Complete. The antiserum itself as seen in the control series (B) does not exhibit any tendency to interfere with hemolysis, even in the amount of o.i c.cm. (larger quantities never come into consideration). On the other hand, series (A) shows that the larger amounts of the antiserum do not bind complement as thoroughly as the medium doses. The zone of complete complement fixation lies between 0.05 and 0.025 c.cm. of the antiserum. It is advisable as a general rule to choose about one and one-half to two times this minimum quantity. Thus from Table III it can be noted that 0.2 c.cm. of a i :6 dilution could well be adopted as a test dose for com- DIFFERENTIATION OF PROTEIDS BY COMPLEMENT FIXATION 1 95 plement fixation. If it is required to know how delicate the complement fixation reaction can be with this dose of antiserum, the following experiment (Table IV) is performed. Diminishing amounts of human serum are mixed with a constant quantity of complement and with this constant dose of antiserum. At the same time a control series of tubes is used, in which the antiserum is substituted by salt solution. After one hour of incubation at 37° C. erythrocytes and hemolysin are added. TABLE IV. Amounts of human serum in cubic cen- timeters. Series A contains human serum + i : 6X0.2 c.cm.1 antiserum, +0.1 c.cm. of guinea-pig's serum. Series B (control) contains human serum+o.2 c.cm. physiological salt solution+o.i c.cm. guinea- pig's serum. One hour at 37° . One hour at 37°. +0.001 c.cm. of amboceptor+i c.cm. of 5 per cent, ox's blood. +0.001 c.cm. of amboceptor+i c.cm. of 5 per cent, ox's blood. Hemolysis. Hemolysis. O.I O.OI O.OOI O.OOOI O.OOOOI 0 0 o o o Slight. Complete. 1 Complete. It is seen from the above table that o.ooooi c.cm. of human serum still suffices to give a partial although incomplete fixation of the complement. The delicacy of the antiserum in this particular instance is not very great. In forensic practice, the reaction is carried out as shown in Table IV, but instead of the human serum, the solution of unknown blood stain in various dilutions is titrated. Control series B should not be omitted, because here, any foreign substance contained in the extract and which might interfere with the reaction can be detected. i :'6Xo. 2 c.cm. means o. 2 c.cm. of a 1:6 dilution. CHAPTER XV. PHAGOCYTOSIS. OPSONINS AND BACTERIOTROPINS. I. Phagocytosis. By phagocytosis is meant the taking up, or engulfing of foreign sub- stances by certain cells (digesting cells or phagocytes) for the purposes of digestion. As a mode of nutrition, this is well known to exist normally, in the lowest unicellular animals, for instance the amebae. Intracellular digestion, however, can be traced to organisms higher in the scale of the animal kingdom; even among mammals the function of cell ingestion is found, although limited to a definite group of cells, especially those derived from the mesoderm. The inspiration for the work on phagocytosis and the greater part of its theoretical considerations have emanated from Metchnikoff and his numerous pupils at the Pasteur Institute at Paris. Metchnikoff divides the phagocytes into two classes, the " sessile or fixed phagocytes," and the " wandering phagocytes." The first is the stationary endothelial lining of blood vessels, and lymph spaces, as well as the large cells of the spleen pulp and lymph glands; the second consists of the white blood cells of the circulation. From another standpoint the phagocytes are divided into ' ' microphages ' ' and ' ' macrophages. ' ' The former are practically identical with the neutro- and eosinophile polymorphonuclear leucocytes, while the latter present no distinct group, but include large lymphocytes, myelocytes, giant cells, etc. The cells designated as sessile phagocytes also belong to the class of macrophages. The size of the cell was considered by Metch- nikoff as the deciding feature; not all macrophages are mononuclear as generally believed. Thus macrophages appearing in the peritoneal fluid of guinea-pigs frequently possess, like the giant cells of the tubercle, numer- ous nuclei. According to Metchnikoff it is primarily the microphages to whom the function of bacterial phagocytosis is allotted, while the macro- phages serve for the purpose of ingesting dead or moribund tissue struc- ture. Still there are certain pathogenic micro-organisms, tubercle bacilli, lepra bacilli, actinomyces, which are favored in that they also are digested by the selective macrophages. The evidence of phagocytosis is established by mixing either in vitro or vivo the substance for phagocytosis, plus the phagocytes, and noting the changes which ensue; [either in a stained or unstained preparation]. The phagocytes of the guinea-pig's peritoneal cavity are especially well adapted for the study of phagocytosis in vivo. The following experiment of Metchnikoff may serve as a type. 196 PHAGOCYTOSIS 197 A guinea-pig receives an intraperitoneal injection of goose's blood. Immediately following this, the leucocytes disappear from the peritoneal fluid. This is due partly to a destruction of leucocytes (Phagolysis) and partly because the leucocytes are re- pulsed and settle upon the peritoneal wall. In one to two hours this so-called negative phase is overcome and there is an increase of the leucocytes, especially of the macro- phages in the exudate (Hyperleucocytosis). Now, the leucocytes can be seen sending forth short protoplasmic processes — pseudopodia, by means of which the erythrocytes are drawn into tHe phagocytes. After a short time the macrophages are filled with the erythrocytes. At first the ingested cells appear normal; gradually, however, they un- dergo changes, which are clearly visible in the unstained specimen, indicative of a disintegrating process, within the body of the phagocytes. ."'" .-•;• • The same phenomenon as described for goose's erythrocytes can also be observed with bacterial bodies. In order to exclude the possible bactericidal influences of the serum, it is advisable when one is working with bacteria which are readily destroyed, as cholera vibrios, to previously induce a hyperleucocytosis in the peritoneal cavity. The guinea-pig receives an intraperitoneal injection of 10 to 20 c.cm. of sterile bouillon or aleuronat solution; in about twelve hours hyperleucocytosis takes place, and a capillary pipette inserted into the peritoneal cavity will withdraw a thick and turbid exudate of leucocytes. If this animal is injected intraperitoneally with bacteria, and a smear of the peritoneal fluid made a short time after the inoculation, the bacteria will be seen lying within the microphages. This important fact has been variously interpreted. Pfeiffer and his pupils claim that the bacteria are first destroyed or their virulence greatly diminished by the bactericidal power of the serum and exudate, and that the phagocytes act only as receptacles for these already destroyed bacteria. Metchnikoff believes that the phagocytes take up the living bacteria and destroy them, thus representing these cells as the most important weapons of the organ- ism in its protection against infection. " Whenever an organism, that has lost its susceptibility to a par- ticular infection, either on account of a natural born immunity or one artificially attained, comes into conflict with the etiological agent, a struggle arises between the latter and the phagocytes of the threatened individual. The phagocytes appear as victors, since they take up the bacteria into their protoplasmic bodies and digest them, thus forever destroying the evil." (Metchnikoff cited by Levaditi.) Critically considered, there can be no doubt that the phagocytes by their very nature are capable of dealin gwith living virulent bacteria. At the same time one must observe that the opsonins and bacteriotropins of the serum, soon to be discussed, in most instances previously modify the living bacteria in a way at present still unknown. That, however, the phagocytes can ingest bacteria or protozoa which are alive and active, has been demon- strated by Metchnikoff 's school. Phagocytosis experiments were under- 1 98 PHAGOCYTOSIS. OPSONINS AND BACTERIOTROPINS taken with motile bacteria and spirilla. On microscopical examination it was seen that a phagocyte was in the act of taking up a spirillum, part of which was engulfed by the cell while the remainder was still outside of the cell and continuing its active motility. Not in all cases does phagocytosis of bacteria lead to destruction of the ingested microbes. More recent experiments also seem to prove that simple phagocytosis of bacteria must not be considered as identical with their death. Thus, the exudate from cases of anthrax in which the bacilli lie within the leucocytes, can still produce fatal anthrax when inoculated into animals. Vital Stain- A more exact understanding of the bio-chemical nature of ing with phagocytic digestion has been offered by the method of vital Neutral Red. staining with neutral red. Neutral red (used as a i per cent, solution in isotonic saline) is a chemical dye which stains only dead cells and not living ones. If live bacteria and phagocytes are mixed and hanging-drop preparations of these are made, and then a drop of the stain be added to different preparations at [successively increasing intervals of time, the first slide shows the extracellular living bacteria unstained, while of the intracellular bacteria, a part remains unstained and the other colored red. The later the mixtures are stained, the more numerous are the intracellular red stained bacteria, showing that the ingested micro-organisms remain alive for a short time, and then die. The intracellular bacteria retain their stain as long as the phago- cytes themselves remain alive. Later, when the phagocytes die, the formerly red bacteria lose their stain. MetchnikofFs explanation of the red staining process is that during the act of digestion by the phagocytes an acid ferment is liberated which gives the color reaction with the neutral red. For many years MetchnikofFs phagocytic theory opposed the concep- tion of Ehrlich and also Pfeiffer in relation to the importance of ambo- ceptor and complement in the mechanism of immunity. It would be out of place here to review the various experiments performed and offered on each side in explanation of its standpoint. Suffice it to say that Metch- nikoff denied the existence of free complement within the animal organism. He moreover claimed that the complement was found normally only in the phagocytes and hence called it "cytase," differentiating the two phagocyte groups as "micro- and macrocytase." The "cytase" is liberated when the phagocytes are broken up. The amboceptors are considered as split products of the phagocytes and known by Metchni- koff as "fixators." 2. Opsonins. In recent years the closer agreement which has arisen between the followers of phagocytic and humoral theories was made possible by the fact that Denys and Leclef, Leishmann, Wright and Douglas and others, PRINCIPLES OF THE OPSONIC INDEX 199 demonstrated that phagocytosis occurs in most cases only in the presence of serum. If the phagocytes are thoroughly washed, so that they are entirely serum-free, phagocytosis will not take place, or will do so imper- fectly. The belief of some authors that " spontaneous phagocytosis" without serum was altogether impossible was disproved, especially by Lohlein. The manner in which the serum acts, whether it stimulates the digestive activity of the leucocytes or whether it so changes the bacteria that they can more readily be taken up by the phagocytes, has been settled in favor of the latter view through researches, especially of Wright and his followers as well as of Neufeld. The substances within the serum which thus modify the bacteria have been designated by Wright as "opsonins." C'opsono" = I prepare food for.) Opsonins are demonstrated by mixing bacteria, serum and washed leucocytes, allowing this mixture to remain in the incubator for a short time, and then staining smear preparations. Wright then counts a certain number of leucocytes and the number of bacteria found within these leucocytes. The relation between the number of ingested bacteria and the counted number of phagocytes is designated as the The Opsonic phagocytic count. Wright compared the phagocytic counts Index. of infected individuals with those of normal persons and found that those of the former were much lower. The relation existant between the two he expressed in the form of a fraction and that is known as the opsonic index. Thus a smear made from a mixture of equal parts of an emulsion of staphylococci, leucocytes and the patient's serum showed for example 75 cocci to 100 leucocytes; while one made from a mixture of equal parts of the same bacterial emulsion and leucocytes, but a normal individual's serum, demonstrated 150 bacteria to 100 leucocytes. The opsonic index of the patient's serum was therefore one-half (0.5). According to Wright, the opsonic index expresses the animal's resist- ance to infection. He believes that a low opsonic index for a given bacterium indicates a susceptibility on the part of the individual for that particular infective agent. Furthermore, the opsonic index he claims can be used as an aid in the diagnosis of infectious diseases, inasmuch as opsonins are specific. Thus the opsonic index in a tuberculous individual is low only for the tubercle bacillus and not for other bacteria. When an animal is immunized, its opsonic index toward the respective bacterium is considerably increased. The question has been asked whether the immune opsonins formed during this process are identical with the normal opsonins. Wright and a number of the more recent authorities believe that they are different. Neufeld, who discovered these immune opsonins independently of Wright, named them Bacteriotropins, and pointed out that while the normal opsonins are destroyed when heated to 56°, the bacteriotropins remain unharmed. As yet the exact nature 200 PHAGOCYTOSIS. OPSONINS AND BACTERIOTROPINS of the immune as well as of the normal opsonins has not been clearly defined. It is still a matter for investigation whether in the case of opsonins one is dealing with entirely new substances or whether they are the old well-known bodies like the agglutinins, complements and amboceptors with a new action. The fact that the opsonic index is raised by immunization while it is usually found diminished during spontaneous infection in man, led Wright to believe that good results may be obtained by Increase of increasing the opsonic index of the already infected individual Opsonic by means of immunization. In this way he thought the Index by Im- patient's predisposition to the particular infection would munization. be overcome. Wright's experiments showed that the opsonic index could be increased by injection of extremely small doses of dead bacteria (Wright's vaccines.) If an individual suffering from an acne or furunculosis, and who has a low opsonic index for the staphylococcus, is injected with a very small number of staphylococci, his opsonic index sinks still more for a short period after the inoculation (negative phase). This is explained by the fact that the injected bacteria absorb the existing opsonins. New opsonins are, however, then produced, which immediately make up for the loss occasioned during the negative phase, with the result that after several days there is an increase of the opsonic index (positive phase) which lasts for a short time. Then the index again begins to fall, as the stimulus to the formation CHART 5. — Curve of the opsonic index following the £ . . inoculation of a small dose of staphylococcus vaccine. of °Psonms 1S transitory. It The arrow indicates the time of injection. usually sinks to below the nor~ mal level, only to rise again to a point slightly above the normal, where it remains stationary. This irregular curve represents the typical course of the opsonic content of the blood after a vaccine injec- tion; apart from this characteristic picture numerous exceptions exist. Thus by the use of very minute bacterial doses, the negative phase immediately following the injection is entirely absent. Reversely very large doses exhibit a prolonged nega- tive phase. Wright graphically represents these variations in the opsonic index by charts, an example of which is given here (Chart 5). In order that the therapeutic effect may persist, it is advisable to repeat the inoculation. A new injection should be given at the height of, or during the positive phase, as an inoculation repeated during a negative phase will result in further depression of the index to a very low level. It is even possible in this way to harm the patient. The Opsonic Index Normal A \ i \ i , t / \ / ^ s \ ^ / \ / \ f V V 9 10 »J1 12 13 m 15 16 17 18 19 20 21 22 AUTO-INOCULATION 201 poor results obtained during the first era of tuberculin treatment can, according to Wright, be attributed to the failure of this observation. It is the production of cumulative positive phases that is the aim of vaccine treatment. (Chart 6.) Wright and his co-workers have noticed that an increase in the opsonic index usually runs parallel with an improvement in the condition of the patient. Inasmuch as an increase in the opsonic index is occasioned by introduc- ing into the general system even a very small number of bacteria, it seems probable that such spontaneous inoculation will take place during the course of an infectious disease. In fact, a spontaneous rise in the opsonic index is observed during convalescence or after the crisis of an infection. A high index is, however, also noticed at other times; for example, tuber- Nov. Dec. 30 1 Z 3 4 12 13 14 1516 17 18 192021 222324252627282930 CHART 6. — Opsonic curve during treatment with New Tuberculin. culous individuals show a higher index than normal persons. Wright explains this by the so-called " auto-inoculation;" for example, after moderate exercise, or work, tuberculin is liberated from the tuberculous focus and in this way acts like a therapeutic injection of tuberculin, i.e., the index will be raised. Therefore, an excessively high opsonic index is of just as great diagnostic value as a low one. Wright furthermore believes that constant irregularities or variations in the height of the opsonic curve serve as plausible evidence for the existence of infection, because under normal circumstances the curve should remain at a level. Not infrequently, however, cases come under observation where in spite of a 202 PHAGOCYTOSIS. OPSON1NS AND BACTERIOTROPINS Dpsonii: Index. 2.4 2.2 2.0 16 14 7,2 Normal 1 0 as 0,6 Gono-Opsonic Index. • • -•- Tuberculo-Opsontc Index MINIMUM CHART 7. — Increase in the opsonic index for the gonococcus brought about by Bier's hyperemia. Feb. CHART 8. — Tuberculin auto-inoculation following physical examination and massage. (Tuberculous Lymphadenitis.) AUTO-INOCULATION 203 distinct evidence of the existence of an infection the opsonic index remains normal. In such instances, for some reason, the bacteria and their products do not reach the general circulation and therefore no occasion is offered for either an elevation or sinking of the opsonic index. Wright and Freeman were able to show that all active and passive motions of an infected joint, as well as any vascular changes which induce a flow of lymph toward the focus of infection, lead to auto-inoculations, which are manifested in a change of the opsonic index. Such artificial production of auto-inocula- tion can be employed in various forms as a means of diagnosis: thus, in articular rheumatism, massage; in pulmonary tuberculosis, breathing exercises; in laryngeal diseases, loud reading; and in tuberculosis of the lower extremities, active gymnastics will occasion changes in the opsonic curve. An example is given in Chart 7. The patient was a woman with a swollen wrist joint. In order to decide whether this was a gonorrheal' or tuberculous process, the opsonic index was taken and found to be 0.94 to 0.97 for the gonococcus and 1.03 to 1.35 for the tubercle bacillus. As these figures differed very slightly from the normal, the test was repeated, but this time after Bier's hyperemia had been applied and the forearm placed into warm water for one hour. The opsonic index for the tubercle bacillus remained the same, while that for the gonococcus had undergone marked variations. A similar experiment with a woman having tuberculous lymphadenitis is given in Chart 8. Wright makes use of these variations of index caused by auto-inocula- tion in determining the prognosis of a case. An infection is only then considered cured when artificial auto-inoculation is no longer possible. The Technique for the Determination of the Opsonic Index. For the determination of the opsonic index are required, 1. Serum of the patient. 2. Serum of the normal individual (as control). 3. Washed blood cells (Leucocytes). 4. Bacterial emulsion. The blood serum is obtained from the finger tip at the root of the nail. It is most efficacious to first produce a hyperemia of this part by constrict- ing the finger either with a narrow gauze bandage or a small soft rubber tube (the editor has found the latter much more convenient) . The prick is then made with a needle or finely drawn out glass tube. The blood flows spontaneously and is collected into one of Wright's capillary tubes (Fig. 19) by approximating the curved end of the latter to the blood (Fig. 18). The straight capillary end of the tube (away from the blood) is then warmed in a small flame and sealed. The tube is laid down flat, and 204 PHAGOCYTOSIS. OPSONINS AND BACTERIOTROPINS allowed to cool; by the cooling the blood is sucked back from the unsealed capillary end; this end may also be sealed in the tip of the flame. The blood then coagulates and the serum separates off. The separation of the latter may be hastened by centrifugalization for a short time. FIG. i 8. FIG. 19. FIG. 20. In order to obtain leucocytes, a small test-tube which holds 3 to 4 c.cm. is filled 2/3 with a 1.5 per cent, solution of sodium citrate, and about 6 to 7 drops of blood from a healthy individual" are collected in .this solution (Fig. 21). The tube is inverted several times to thoroughly m FIG. 21. FIG. 22. the blood so that the citrate, by precipitating the calcium salts of the blood, effectively prevents coagulation. The suspension is centrifugalized until the corpuscles are thrown down and a distinct white layer (leucocytes) is seen upon the surface of the red cells (Fig. 22). The clear supernatant citrate solution is pipetted off, care being taken not to disturb the white layer. Some 0.85 per cent, saline is added, mixed and the mixture again PREPARATION OF BACTERIAL EMULSION 205 centrif ugalized. The washing with normal saline solution is repeated once or twice and as much of the clear liquid as possible is finally removed; the remaining cells are thoroughly mixed and in this form are ready for use. The bacterial emulsions with the exception of the tubercle bacillus are made from agar cultures ; the growths of gram + cocci may be as old as twenty-four hours, while the coliform organisms and the gram — cocci are preferably used only four to ten hours old (the younger the better). A loopful of cul- ture from an agar tube is thoroughly rubbed up with several drops of salt solution in a watch-glass by means of a small glass pestle. The salt solution is best added very gradually, drop by drop, in order to make a more perfect emulsion. This may then be advantageously centrifugalized for a short period, to bring down the large clumps. The supernatant opalescent portion is Mar FIG. 24. taken off for use, thoroughly mixed, and if necessary diluted. Emul- sions of coliform organisms are the most easily made. Frequently it is sufficient to rub up with the platinum loop a loopful of such bacteria on the side of a small test-tube containing saline. The proper thickness of the resulting emulsion varies. As a rule, bacillary emulsions are re- 206 PHAGOCYTOSIS. OPSONINS AND BACTERIOTROPINS quired to have a thicker appearance to the naked eye than coccal ones. The latter should be only slightly opalescent. In order to make a satisfactory tubercle emulsion, a more elaborate method is necessary. The dead and dried tubercle bacilli are employed for this purpose. A portion of these bacilli is very thoroughly triturated in an agate mortar, or between two slides, or in a grinder devised for this purpose, at first alone and then with 1.5 per cent, salt solution added drop by drop. In this way a paste, and subsequently a comparatively thick emulsion is made. For use, a small portion of the resultant emul- sion is centrifugalized until the upper layers are fairly opalescent. These upper layers are pipetted off, and thoroughly mixed. A smear of this should be made and stained in order to observe that the emulsion FIG. 25. is free from clumps and not too thick. Such an emulsion sealed up in a glass tube and sterilized at 60° C. for i hour can be kept for about one week. Streptococci may similarly be rubbed up in a mortar with 0.85 per cent, salt solution and then centrifugalized. As a rule, however, vigorous pipetting into a watch glass with subsequent centrifugalization for a few minutes is sufficient to break up the chains and leave a satisfactory emulsion. If several specimens of blood are to be examined it is best to do a preliminary phagocytic count in order to test the strength and condition as regards clumping of the emulsion. The phagocytic count for tubercle should be between 1.5 to 2 per cell and for other organisms not less than 3 per cell. According to the preliminary finding further dilution or concentration of the emulsion is necessitated. The pipe tes employed for the opsonic index should b about 16 cm. long and made from glass tubing about 5/16 of an inch in diameter. They should all be approxi- PREPARATION OF SMEAR 207 mately of the same caliber and but slightly tapering toward the point. The rubber nipple should tightly fit the piece of tubing or bulb available. For use, the capillary end should be cut square and the pipettes marked with a paraffin pencil about 3/4 of an inch from their extremity. The content as far as this mark is the unit of volume in each case. The rubber nipple is now held between thumb and forefinger and gently compressed, the capillary end introduced into the well-mixed blood cells and the unit volume drawn up by slightly relaxing the pressure on the bulb. Next a tiny bubble is allowed to enter, then an equal volume of the emulsion, followed by another tiny bubble which latter is succeeded by an equal volume of serum. By gentle pressure on the bulb the several volumes are ejected upon a clean glass slide, and thoroughly mixed by alternately sucking the mixture into the pipette and squeezing it out again upon the slide. It is enough to repeat this action three times. Then the FIG. 26. mixture is drawn up into the pipette, the end sealed in a small pilot flame, the pipette placed into the opsonizer (Fig. 24) and the time noted. This operation is repeated with each serum. Coliform organisms and" the gram negative cocci should be incubated not longer than six to eight minutes. Tubercle bacilli and other organisms require fifteen minutes, more or less according to the strength of the emulsion. The pipettes are withdrawn in the same order in which they were placed in the opsonizer. The contents of each are blown out on to a slide, and very carefully mixed as before (Fig. 25). The entire quantity is divided between two or three slides and several smears are made, the best one being selected for counting. These slides should previously have been roughened with very fine (oo) emery paper and cleaned with a duster, and should rest on their concave surface so that the smear is made on the convex side. (It will be noticed that a slide can be made to rotate if resting on one surface (convex), but does not do so when resting on the concave surface.) The smears are best made by means of a broken slide with a slightly concave edge. This "spreader" (Fig. 26) is made by sharply breaking a glass slide at about its middle, this being 208 PHAGOCYTOSIS. OPSONINS AND BACTERIOTROPINS facilitated by scratching the edges of the slide with a glass cutter at the point where it is desired to break it. The editor has broken as many as twenty to thirty slides before a proper spreader was obtained. It pays to do this, because upon the sharpness of the fracture and cleanliness of the spreader depends the edge of the film, and secondarily the ease, rapidity, and accuracy of the count. If the film be well made, it will have a straight edge within which will be found practically all the leuco- cytes, as they are larger than the red blood cells, and therefore dragged to the end of the film. FIG. 27. — Phagocytosis of tubercle bacilli. The preparations are fixed in a saturated solution of corrosive subli- mate for two or three minutes, washed with water, and stained with methylene blue or carbol-thionin (1/4 per cent, thinonin, and i per cent, carbolic acid). Carbol thionin is preferable. It should be slightly diluted and warmed before being poured upon the slide. Here it is allowed to remain for several minutes, then washed off in water, and the slide dried with filter-paper. The tubercle films are best fixed with formalin vapor, stained with hot carbol or aniline fuchsin, decolorized in 2.5 per cent, of H2SO4, treated with 4 per cent, acetic acid to dissolve the erythrocytes and counter stained with 1/2 per cent, of methylene blue in 1/2 per cent, of sodium carbonate. It is most important that tubercle films be carefully stained because it is desirable to color every bacillus and yet not break up the leucocytes (Fig. 27). With a i/i 2 inch oil immersion lens a minimum number of one hundred DETERMINATION OF OPSONIC INDEX 2OQ polymorphonuclear leucocytes are now examined and the number of microbes they contain enumerated. Similar calculation is undertaken with the normal control serum. The fraction obtained by dividing the number of bacteria contained in 100 cells on the patient's slide, by the number in 100 cells on the normal slide, gives the opsonic index of the patient's serum. For example, the normal individual has 284 and the patient 262 bacteria in 100 cells, the fraction which gives the patient's opsonic index would be 262/284 °r 0.92. The principle of Wright's technique is simple, but it requires a great deal of practice before it is mastered. Only then are the results reliable. One must adopt the same principles when counting the control slide as when the patient's film is counted. If in the last case, for instance, the cocci situated on the edges of the cells are not included in the count, they should also be excluded in the first case. The absolute count is of no importance. It is the relative proportion which is significant. As a normal control, it is best to take the average of the phagocytic counts of a series (3 to 4) of normal sera or first equally mix the different sera, and take the phagocytic count of the pool. Normal sera should not differ from one another in a tubercular opsonic estimation by more than 10 per cent. Wright's Vaccine Treatment. As has been said, the principle of Wright's vaccine treatment depends upon the immunization with small doses of dead bacteria, so-called vac- cines, whereby the opsonic index of the individual is raised. This is usu- ally associated clinically, with improvement in the patient's condition. The effect of the immunization according to Wright depends upon: 1. Individual reaction of the patient. 2. Preparation of the vaccine. 3. Dosage and form of application. The individual reaction of the patient can be measured by the opsonic index. As far as the preparation of the vaccine is concerned, Pasteur's conten- tion that a vaccine must necessarily be made up of living cultures has not proved itself correct. Carefully killed cultures suffice in almost all cases. An example of the preparation of Wright's vaccine is here given. The Preparation of a Staphylococcus Vaccine. Agar cultures are grown for twenty-four hours, and about 3 c.cm. of sterile normal saline solution are added to each culture. The growth is washed off in the saline solution by means of a platinum needle or freshly 14 210 PHAGOCYTOSIS. OPSONINS AND BACTERIOTROPINS prepared capillary pipette. The suspension of bacteria is placed in a sterile test-tube, the end of this tube drawn out in the blow-pipe flame and sealed. The drawn-out portion should be about 2 inches in length and as strong as possible. The emulsion is now vigorously shaken for fifteen minutes. The extremity of the drawn-out tube is then cut and a few drops of the emulsion expelled into a clean watch glass, or a small part of the drawn end is cut off so that a portion of the emulsion is still contained within it. The tube is resealed, and then submerged in water kept at 60° C. for one hour. This usually suffices to kill the bacteria. The small amount placed in the watch glass or in the capillary test- tube serves for the standardization, which is carried out as follows: A pipette with rubber bulb, as prepared for the opsonic-index test, is also used here. A volume of freshly drawn blood of known corpuscular content, best taken from the worker's own finger, and an equal volume of bacterial emulsion are mixed thoroughly with six or seven volumes of i 1/2 per cent, citrate solution; several even films (which may be fairly thick), are then made by means of the ordinary edge of a slide, and stained with carbol-thionin, Leishmann's or Jenner's stain. The entire smear is divided up (with a blue skin pencil) into eight equal subdivisions, by one transverse line drawn parallel to the long diam- eter of the slide at its middle and five vertical lines, one at each edge of the smear, one in the center and one equally distant between the edge and the central line. It is also advantageous to employ an eye piece, the field of which has been divided or made very much smaller by the insertion of a paper screen with a small central opening representing the size of the desired field. Five or six fields are then counted in each of the eight sub- divided areas. The number of red blood cells seen in each field are enumer- ated in one vertical column, the number of organisms in the same field in another column. In this manner an average of the entire slide is obtained. By means of a simple proportional sum, the number of bacteria per cubic centimeter of emulsion is estimated, e.g., the number of red blood cells counted is 850 and the number of bacteria 1020. The red blood corpuscles used in the standardization are known to number 5,000,000 to a cubic millimeter or 5,000 million to a cubic centimeter; therefore the number of bacteria to a cubic centimeter of the unknown emulsion is expressed as follows. 850 : 1020 : : 5,000,000,000 : No. of bacteria per c.cm. of emulsion, .*. 6,000,000,000 = the number of bacteria per c.cm. of emulsion. After the emulsion has been heated for one hour, the tube is opened and a drop is expressed into an agar culture tube which is incubated for twenty-four hours to demonstrate whether the emulsion is sterile or not. At the end of this time, if a growth is observed, the emulsion must be heated again for one hour at 60° C/and its sterility again tested for. CONCENTRATIONS OF DIFFERENT VACCINES 211 Proper dilution of the emulsion is next undertaken. Small bottles containing 25 c.cm. of 1/2 per cent, carbolic acid in sterile saline are aseptic- ally closed with rubber caps; for example, it is desirable to make up these 25 c.cm. with staphylococcus vaccine so that each cubic centimeter contains 500 million bacteria, then (desired amt. to each c.cm.) 500,000,000 X 5 No. of c.cm. desired 6,000,000,000 (dose of original emulsion per c.cm.) 2.08 c.cm; or approximately 2 c.cm. of the original emulsion must be added to the 25 c.cm. (to be exact 23 c.cm.) to make up the desired dilution. The rubber cap is finally coated with melted paraffin wax. For stock vaccines it is best to make up the different vaccines in the fol- lowing concentrations: 1. Staphylococcus vaccine — prepared from various strains of staphy- lococcus, aureus, citreus, and albus, in three concentrations: 1000 million, 500 million and 100 million, to the c.cm. 2. Streptococcus vaccine in 20 mil., 10 mil. and 5 mil. concentrations. Since the streptococcus grows very sparingly, cultures of two or three days' growth may have to be employed for the preparation of a vaccine, and even then it may be necessary to use one broth culture instead of sterile salt solution to emulsify the agar cultures. On standardizing such thin vaccines it is frequently necessary to take one volume of blood to two, three, or even more volumes of emulsion and then calculate accordingly. 3. Acne vaccine in 20 mil., 10 mil. and 8 mil. 4. Mixed acne in 20 mil. acne and 500 mil. staphylococcus. 5. Gonococcus vaccine in 50 mil. and 5 mil. Gonococcus vaccines are best employed as autogenous vaccines. 6. Typhoid vaccine in 1000 mil. and 2000 mil. for prophylactic inocu- lation. 7. Colon vaccine in 25 mil., 10 mil., 5 mil. Vaccines of coliform organ- isms are very easily emulsified ; as a rule they should not be older than twelve hours and not be sterilized for more than three quarters of an hour. With the exception of the staphylococcus vaccines, it is advisable not to use stdck vaccines, but autogenous vaccines, i.e., vaccines made from the specific strain of bacteria causing the infection to be treated. It is very important to isolate the supposed pathogenic organism from the innocuous or less pathogenic bacteria contaminating or complicating the infection. In tuberculosis Wright employs a dilution of Koch tuberculin (T. R.). Recently he has prepared a tubercle bacillus vaccine in the same way as the other bacterial vaccines. The initial dosage varies with the different vaccines, but should in general be about 100 to 500 million of staphylococci; one may go as high as 2,500 or even 5,000 millions. 212 PHAGOCYTOSIS. OPSONINS AND BACTERIOTROPINS In colon, streptococcus, gonococcus and acne, doses of i to 3 million should be used at the beginning and then gradually increased. In tuberculosis Wright starts with the T. R. in dilution equivalent to about i/iooo mg. of the dry tuberculin substance and this is increased to about 1/600 mg. Wright cites two general rules to be observed in the therapy of infectious diseases. 1. In all cases where the normal antibacterial power of the blood has been lowered, immunization is indicated. 2. Whenever the blood possesses strongly active curative powers, an increase of the blood supply to the infected part should be attempted in order that the antibacterial elements of the blood and leucocytes may display their effect. In such cases the production of hyperemia is particularly of help. Similarly, massage and other such therapeutic measures can be useful. The therapeutic value of auto-inoculation is very slight and should not be encour- aged, as in this way the exact dosage cannot be followed. Wright has employed these vaccines in staphylo-, strepto-, and gonococcus infec- tions, as well as in coli infections, tuberculosis, malta fever and carcinoma (!) where injections of the bacillus neoformans Doyen were given. From a critical review of the cases published, which were treated with vaccines by Wright and his fellow workers, one certain conclusion can be reached; namely, that given an infection, inoculations with small doses of the respective dead or extracted homologous bacteria, will result in a thera- peutic immunization. Although Koch had advanced the same principle for the treatment of tuberculosis, it is Wright who first recognized the general application of this form of immunity. Furthermore, by means of his opsonic studies, he was able to prove that by the injection of even the minutest doses, for example 1/1,000,000 c.cm. of tuberculin, immune reactions are incited. In spite of this finding, investigators are still at variance over the ques- tion, and two camps exist: one of which believes that the ideal treatment of tuberculosis consists in the repetition of the small doses; the other, that the best results are obtained by gradually increasing the dose of tuberculin until very large doses are administered. Citron has found the latter course more satisfactory. Since, as is known, tuberculin is one of the harmful agents in tubercu- losis infections, it seems more advantageous to get the patient, if possible, into a condition where he is able to neutralize large doses of tuberculin rather than to have him at a stage where even moderate doses suffice to give a reaction. Other questions of importance in the vaccine therapy are: first, whether any parallelism exists between the increase in opsonic index and improvement in the clinical manifestations; second, whether the opsonic index must necessarily be used as a guide in vaccine treatment. As to the first, Wright has pointed out numerous cases where exact VACCINE THERAPY AND THE OPSONIC INDEX 213 study has proved that such parallelism exists. This fact is probably correct in the majority of instances, but it cannot be considered as an infal- lible rule, inasmuch as the formation of opsonins is only. one of a great number of factors in the complicated process of healing; consequently one should not be surprised when in some instances in spite of a rising opsonic index, the patient's clinical condition becomes worse, and reversely when in some cases improvement occurs although the opsonic index does not change. Accordingly, the opsonic index during the course of treatment becomes secondary in importance to the exact clinical observation of the case. Wright and his school have shown that certain bad effects may follow from the injection when performed during the negative phase. With the use of small doses the negative phase becomes short — only one day or even less; accordingly it is very probable that this state is entirely passed when an injection is repeated on the fifth to eighth day. The tuberculin therapy at the Kraus clinic is conducted on this principle, without estimation of the opsonic index. And yet, no harmful effects have ever been noted; while general improvement, as increase in weight, diminu- tion in temperature and cessation of cough, are constantly observed. It would be illogical to neglect these clinical data and give preference to the hypothetical action of opsonins as a guide in treatment. It seems that Wright himself does not insist as strongly as before upon the determination of the opsonic index. One of his assistants, Matthews, has made the statement that in a great number of cases the determina- tion of the opsonic index is entirely out of the question. If the choice between injections without estimation of the index and entire omission of inoculation should arise, therapeutic inoculation without the index is by all means indicated. There is a general tendency at present to omit the opsonic index in the treatment of staphylococcus infections, and this is at times also done in tubercle, gonococcus and streptococcus infections as well as in prophylactic typhoid inoculations. Neufeld's Method of Bacteriotropin Estimation. Neuf eld's technique varies from that of Wright in two points: (1) He uses serum free of complement. (2) He does not count the number of bacteria within the leucocytes; but makes various dilutions of the serum and notes in which dilution the bacteria are still ingested in great numbers, as compared with a normal serum in similar dilution as control. Neufeld usually obtains the leucocytes by injecting a guinea-pig intra- peritoneally 1 6 to 24 hours previously with 5 to 10 c.cm. of sterile aleuronat solution (i part aleuronat, 2 parts bouillon). It is best to kill the guinea- pig, and wash out the peritoneal exudate which is full of leucocytes with 214 PHAGOCYTOSIS. OPSONINS AND BACTERIOTROPINS 40 to 60 c.cm. of sterile salt solution. Occasionally it may be necessary to use i 1/2 per cent, sodium citrate solution instead, in order to prevent coagulation. The leucocytes must be washed free of any serum. They should not however be centrifugalized too rapidly, as this tends to clump them. If rabbit's leucocytes are preferred, 50 to 100 c.cm. of 3 per cent, to 10 per cent, peptone bouillon should be injected intraperitoneally. -For mice i c.cm. aleuronat bouillon is sufficient. Human leucocytes are obtained from abscesses or from the blood (Wright). The serum is inactivated by heating it at 50° to 60° C. for one-half hour. This may be omitted for old or carbolized sera as they are usually free of complement. Also tuberculous sera should not be heated as their bac- teriotropins are very susceptible to heat. The various serum dilutions (i : 10, i : 100, i : iooo,etc.) are prepared as usual, but small quantities suffice since only 0.5 to i.o c.cm. of each dilu- tion is necessary. The bacteria are best employed in the form of a 16 to 24 hours homo- geneous broth culture. If agar cultures are used, three loopfuls are rubbed up in i c.cm. of salt solution. For meningococci Neufeld advises agar cultures. Tubercle bacilli can either be triturated in an agate mortar or bought in the form of tuberculin residuum, T. R. (Hoechster Farbwerke). Equal parts (0.02 to o.i c.cm.) of each of the three ingredients (serum, bacteria, leucocytes) are transferred by means of a capillary pipette, to small tubes with flattened bottoms about 4 to 5 cm. long and i cm. wide. Double the quantity of leucocytes may be needed if they are in a weak suspension. The tubes are closed by small corks or non-absorbent cotton, gently but thoroughly shaken and placed in the incubator for definite periods of time, depending upon the micro-organism, from one-fourth to four hours. The supernatant fluid is then pipetted off, cover-glass preparations made of the sediment, fixed by heat and stained by i per cent, methylene-blue solution. A great number of fields are examined microscopically and note made of the weakest dilution which still favors phagocytosis. This is the bacterio- tropic titer of the serum. The necessary controls are: (i) Tube containing normal serum + bac- teria + leucocytes. Phagocytosis of less virulent bacteria frequently occurs even with normal serum. (2) Tube containing bacteria + leuco- cytes, without any serum; a virulent bacteria are sometimes taken up by leucocytes even without serum. This is never the case with virulent organisms. BACTERIOTROPIN ESTIMATION (NEUFELD) 215 If phagocytosis is entirely absent, one should not conclude that bac- teriotropins are not present. Errors in technique are possible: (I) Leucocytes may have been injured; this is especially prone to occur if one has worked with heterologous leucocytes; control examinations with homologous leucocytes (from same animal as the serum) should result in phagocytosis. (II) The serum concentration may be too high. The use of sensi- tized bacteria will obviate this. Bacteria are first mixed with the serum for a short time, the mixtures centrifugalized, and the serum pipetted off leaving a sediment of sensitized bacteria. (III) The time during which the tubes were in the incubator may have been too short or too long. Most micro-organisms require one-half to two hours; pneumococci usually need four hours; cholera vibrios 20 to 30 minutes, as they undergo intracellular digestion very readily. Neufeld's technique is considered by many simpler than Wright's method. In the determination of normal opsonins, however, concentrated or only slightly diluted sera are employed, thus encountering the difficulties mentioned above (II and III). Homologous leucocytes and sensitized bacteria will remove this interference. The editor has found it puzzling at times to decide upon the dilution injwhich bacteriotropins still exist. One may be helped in this decision by the presence or absence of a great number of extracellular organisms. CHAPTER XVI. IMMUNITY AND SERUM REACTIONS IN REFERENCE TO MALIGNANT TUMORS. EXPERIMENTAL TRANSPLANTATION OF TUMORS. IMMUNITY TOWARD TUMORS. SERUM REACTIONS. MEIOSTAGMINE REACTION. The etiology of malignant tumors is still unsolved. It has been defi- nitely proven, however, that tumors or at least some of them can be trans- planted. Naturally such experiments with human cancer are still limited and inconclusive, as attempts to transplant growths from one person to another are entirely out of question. On the other hand, the inoculation of human cancer into lower animals has been successful in the hands of only few reliable workers as Dagonet and C. Lewin. These failures are the less surprising when one considers that malignant tumors obtained from the lower animals, as rats, and transplanted into other animals of the same species, not infrequently cease to grow in their new surroundings. Successful implantations of spontaneous animal rumors vary very greatly, from 4.1 per cent. (Bashford) to 40 to 50 per cent. (Jensen). The percentage becomes still lower or even entirely negative if the transplantation is made upon animals not of the same but of very closely related species, for example, when the gray house mouse is employed instead of the white mouse. Once, however, the tumor continues to proliferate in its new host, it becomes more easily transplan table. Ehrlich has shown that the virulence of a tumor increases the more frequently it is successfully transplanted. Thus a growth can be obtained which may give 90 to 100 per cent, of posi- tive grafts. During this long- continued process, the histological structure of the tumor may change. Not only may a carcinoma be transformed into an adenoma, but even into a sarcoma, as was first observed by Ehrlich and Apolant, and later on by others (Liepmann, Bashford, C. Lewin). That an immunity toward carcinoma may possibly exist has been considered. This has been based upon the facts that some animals resist all attempts at transplantation (" natural immunity"), that in others the tumor grows only to a limited degree, and that in a few, comparatively large tumors re- cede and disappear spontaneously (" acquired immunity ") . The last class nowremains refractory to all tumors, even the most virulent, thus proving a non-specific immunity ("pan immunity"). Ehrlich attempted to produce an immunity against a highly virulent mouse carcinoma by using as vaccine the hemorrhagic mouse tumor which only rarely allows of transplantation. 216 FREUND KAMINER REACTION 21 J Further experiments proved that a certain degree of resistance in an animal may be attained by the injection of various cells, be they embryonal tissue elements or simply red blood cells. Naturally all the known bacteriological methods for the destruction of bacteria have been applied to tumor cells; for example the addition of dis- infectants, or the application of heat. Passive immunization with the serum from animals in whom the tumors disappeared spontaneously, or from rabbits treated for a long time with increasing quantities of an emulsion of carcinoma cells, also gave only doubtful results even though at times the experiments seemed encouraging. While it seems probable that the growth of neoplasms is attended by processes of immunity, one cannot directly compare this with the bacterial or proteid form of immunity and expect the same antibodies as given by the latter. In fact, all attempts at a serum diagnosis for carcinoma or sarcoma by the methods of precipita- tion or complement fixation have failed to withstand careful criticism. The reactions to be discussed will, however, tend to show that the sera of car- cinoma patients possess certain characteristics which may play an impor- tant role in the future development of this problem. "Brieger's cachexia reaction" has been reviewed in the chapter on anti- ferments. This test demonstrates that in a certain number of carcinoma patients the serum contains greater amounts of antitrypsin than normal sera. Similarly, many carcinoma sera have a stronger than normal hemolysin for the erythrocytes of the same or different animal species (iso- and heterolysins). Here too the results are inconstant and the sera of many noncarcinoma patients, especially tuberculous, give similar reac- tions. The same may be said of the test based upon the hemolysis of the carcinoma patients' red blood cells by cobra venom. The Freund Kaminer Reaction. Freund and Kaminer found that if normal serum is mixed with an emul- sion of carcinoma cells and allowed to remain at 40° C. for 24 hours, the latter are broken up and dissolved. This does not occur if the serum is derived from a carcinoma patient. The destruction of cells is deter- mined by counting them in the Zeiss Thoma blood chamber both before and after the 24 hours' incubation; or one may take the clearing up of the turbid cell emulsion as an evidence of the cell destruction. The technique of the reaction is as follows : Carcinoma tissue rich in cellular elements and not degenerated is excised as soon after death as possible, cut up into small pieces and placed into five times as much of a i per cent, solution of sodium diphosphate. The whole is pressed through gauze and allowed to stand until the cells have sunk to the bottom, after which they are washed in 0.6 per cent, salt solution. They are again allowed to settle and then mixed with 2l8 MALIGNANT TUMORS i per cent, solution of sodium fluoride. The sodium fluoride solution is first neutralized (alizarin as indicator) until the violet color is reduced to its minimum. The cell emul- sion may be thus preserved for several weeks. In the test, the carcinoma extract is diluted with 0.6 per cent, sodium chloride solu- tion until it becomes opalescent. To 3 c.cm. of this emulsion are added 10 drops of the patient's serum preferably fresh and active. They are allowed to remain at 40° C. for 24 hours and the test is positive if the turbidity persists. Control tubes must be made of the serum alone, carcinoma extract without serum, and carcinoma extract with normal serum. The last should become clear. Recently Freund and Kaminer proposed the following more delicate modification. The supernatant fluid from the emulsion of carcinoma cells is mixed with acetic acid (5 c.cm. of acid to 100 c.cm. of fluid) heated in the water bath for one-quarter of an hour at 80° C., filtered, and after cooling neutralized with sodium bicarbonate against litmus. Then it should again be heated and filtered. Boiling or heating over the free flame is to be avoided. An extract can also be made by heating the tumor itself (preserved in alcohol) in 0.25 per cent, of acetic acid, then filtering and neutralizing. Serum of cancerous individuals added to this extract produces a cloudiness; non-cancerous serum and extract remains clear. Freund and Kaminer advise that both the cell- counting method and the turbidity reaction should be applied to each serum, one acting as a control upon the other. Ranzi and Admiradzibi, Kraus and Graff have corroborated Freund and Kaminer's findings. Rosenberg working under Citron's guidance has found that although in the main the reaction is obtained as above quoted, some carcinoma sera give a negative reaction and some non-carcinoma patients (tuberculosis, pregnancy) give a positive result. The Meiostagmine Reaction. Weichard showed that by bringing together antigen and antibodies in certain dilutions, the rapidity of diffusion is increased (epiphanin reaction). M. Ascoli further demonstrated that the union between a specific antigen and its specific serum is associated with appreciable lowering in the surface tension, so that the number of drops to a definite quantity of fluid is dis- tinctly increased (Meiostagmine reaction). This term is of Greek deriva- tion, "paw," smaller, "o-ray/wx," drops. Traube's "Stalagmometer"1 measures the number of drops. The meiostagmine reaction has been tested by Izar and Vigano in typhoid, paratyphoid and lues and found to possess a certain degree of specificity. In tuberculosis, the test is positive only in active cases. Ascoli and Izar claim to get reliable results with their method also in car- cinoma. Their technique is as follows: 1 The Stalagmometer of Traube is merely a very finely and elaborately graduated pipette with a central bulbous reservoir. The dropping end of the instrument ends in a flattened ground base thus insuring uniformity in the size of the drops. The instrument is so graduated that a fraction of a drop can be estimated. THE MEIOSTAGMINE REACTION 219 i. Extract: a malignant tumor which has not undergone degeneration (or a sheep's pancreas) is cut into fine slices or small pieces. It is spread out on glass plates and dried in a temperature of 37° C. The dried residue is powdered and extracted with methyl alcohol in the proportion i : 4 for 24 hours at 50° C. The extract is filtered through a hard filter, at first hot and then cold. It should be protected from the light, but should not be placed in the ice-box. The dilutions cannot be preserved, but must be made fresh every time, by measuring out the requisite amount of extract with a dry pipette, placing it in a dry test-tube or flask, and adding the required quantity of distilled water and shaking thoroughly. The extract is first titrated: dilutions 1 150, i 175, i :ioo, i 1125, etc., are made, and to i c.cm. of each of these, 9 c.cm. of a i : 20 normal serum diluted with saline are added. The mixtures are placed for two hours at 37° C. The number of drops to each quantity is estimated. A control tube containing the same quantity of normal serum plus i c.cm. of distilled water is also counted. That dilution is chosen as the working dose which differs from the control tube by 3 to 5 fractions of a drop. The test is performed as follows: To 9 c.cm. of each serum diluted with saline i : 20 is added i c.cm. of distilled water and the number of drops falling from the stalagmometer is counted before and after incu- bating for two hours at 37° C. or one hour at 50° C. This is used as a control. To 9 c.cm. of each serum diluted i : 20 with normal saline is added i c.cm. of antigen emulsion of the desired strength, e.g., dilution i :i25. The number of drops in this mixture is counted only after incubation. As further controls it is advisable to have several known normal and tumor sera. The reaction is positive if the number of drops contained in the mixture of tumor extract and tumor serum is greater by at least 11/2 drops than in the controls after the incubation. 9 c.cm. serum i /2O Number of drops per c.cm. with i c.cm. tumor extract 1:125 (after incubation) Number of drops per c.cm. with i c.cm. of distilled water (after incubation) Normal serum 58+41 58+2 Normal serum . . . 59 58+7 Carcinoma serum 61+6 58+8 Carcinoma serum 60+4 58+5 The antigens spoil very easily. Recently, chemicals in the form of lip- oids are being substituted for the specific antigen. This simplifies matters greatly. In looking over the numerous publications by Ascoli and his 1S8+4 means 58 drops + 4 fractions of a drop. 220 MALIGNANT TUMORS followers, one meets with constant variations in the technique of the reaction. Thus it is very difficult to arrive at an exact description or^a conclusive opinion. Many non-cancerous sera give a positive meiostag- mine reaction. The test has not yet been adopted as a clinical aid. A review of the early literature of this subject is given by Bernstein and Simons, Amer. Jour. Med. Sciences, Dec., 1911. CHAPTER XVII. ANAPHYLAXIS. Introduction. Proteid Hy per susceptibility . Anaphylaxis. Passive Anaphy- laxis. Anaphylatoxin. Serum Disease. Hay Fever. The explanation for many phenomena discussed in the former chapters was based upon the observation that by overcoming an infectious disease the organism undergoes some transformation whereby it becomes refrac- tory or less susceptible to the same infection. This changed state, demon- strated neither by chemical nor microscopical methods, was termed "Immunity." The most constant evidence of immunity is the presence of antibodies. This, however, should by no means imply, as is so often done in literature, that the demonstration of antibodies and the exist- ence of immunity are identical. On the contrary, attention has already been called to the possibility of paradoxical reactions. Animals or individuals in whom a grade of immu- nity is expected, show instead a certain susceptibility toward the particu- lar antigen. This altered reaction of the organism was termed "Hyper- susceptibility" and "Anaphylaxis" (to distinguish it from prophylaxis- immunity). The "paradoxical reaction" was discussed under diphtheria where it was mentioned that horses immunized for a long period of time would suddenly become severely ill after the injection of small doses of toxin. The tuberculin reaction was another instance of such hypersensi- tiveness. Infection by the tubercle bacillus not only made the body highly sensitive toward the bacteria and their derivatives (tuberculin), but a severe and characteristic reaction always took place as the expression of this increased sensitiveness. In recent years, experimental studies have proven that this peculiar phenomenon is not based upon entirely new laws, but that anaphylaxis has as its governing influences principles that are closely related to those influencing the process of immunity. It is still to be determined whether the hypers usceptibility observed with various antigens like the pure toxins, the tuberculins, and the proteids all follow the same biological mechanism. The author is of the opinion that just as there are various forms of immu- nity so also must the existence of various forms of anaphylaxis be presumed. At least a cellular and humoral variety seem distinctly plausible; the tuberculin reaction is an example of the former, the proteid anaphylaxis of the latter. 221 222 ANAPHYLAXIS The parenteral introduction of a foreign proteid into an or- Proteid ganism results in the formation of antibodies against this Anaphylaxis. proteid: precipitins, cytolysins, complement fixation bodies, etc. At first it was assumed that a "proteid immunity" had taken place. Such conclusion proved erroneous. Richet and Portiers (1902) in working with actino-congestin (a substance extracted from the tentacles of Actinia) demonstrated that dogs injected with sublethal doses of this toxin would die acutely if the injection were repeated after an interval of three weeks. In the light of more recently established facts, the explanation of Richet's work was complicated because actino-congestin consists of two components, a true toxin against which an immunity can be stimulated, united with a proteid element which like all proteids produces anaphylaxis. It was therefore of extreme fundamental importance when Arthus and Theobald Smith showed that proteid substances, of themselves apparently non-toxic, can constantly produce the phenomena associated with hyper- sensitiveness. The Arthus Phenomenon. If a rabbit is injected subcutaneously with horse's serum at intervals of six days, a soft infiltrate which remains for two to three days appears at the site of injection after the fourth inoculation, a harder infiltration which continues for a longer period of time after the fifth inoculation, and gangrene after the sixth or seventh. A rabbit repeatedly treated intravenously with horse's serum may die with severe general symptoms several minutes after one of the later inoculations. The Theobald Smith Phenomenon. Theobald Smith observed that guinea-pigs injected with neutral mix- tures of diphtheria toxin and horse's antitoxic serum would be killed if after an interval of several weeks they were given a subcutaneous injection of normal horse's serum. Otto and others showed that both of these phenomena, above described, are identical in their principle; thus, that of Arthus can be likewise in- duced after a single injection of horse's serum if the first dose is small, and if the interval between the first and second inoculation is sufficiently long (about three weeks or more). Further study proved that anaphylaxis may be incited by repeated paren- teral introduction of practically any foreign proteid. The first inoculation prepares the animal in such a manner that after a definite incubation period the second injection of the same serum will PRINCIPLES OF ANAPHYLAXIS 223 bring on characteristic acute symptoms which may terminate fatally. The picture of hypersensitiveness or " serum sickness" in man, as described by v. Pirquet and Schick, is classical. Other investigators who deserve merit for work in this field are Arthus, Otto, Rosenau, Anderson, Kraus, Doerr, Besredka, Weichardt, Wolff-Eisner, Friedemann, Friedberger, H. Pfeiffer, Schittenhelm. It is through their efforts that the close relationship between hyper- sensitiveness and immunity is more clearly understood. Like the state of immunity, anaphylaxis is either spontaneous or acquired. It is also specific, that is a guinea-pig made sensitive toward horse's serum will react only when again treated with horse's serum but not when receiving rabbit's or human serum. A classical anaphylaxis experiment may be carried out as follows. A guinea-pig receives a subcutaneous injection of o.ooi to o.oi c.cm. of horse serum and after three weeks an intravenous injection of 3 to 5 c.cm. is repeated. This quantity, which under normal conditions has no influence upon the animal, will now produce very alarming symptoms or even death in a couple of minutes. There are several factors upon which the occurrence of the anaphylactic phenomena strictly depends. (a) The first, preparatory, or sensitizing dose must enter the system in some way other than through the gastro-intestinal tract. Only ex- ceptionally does hypersensitiveness arise if the antigens are given per os. (b) The quantity of antigen is of the utmost importance. The smallest amounts of proteid suffice, e.g., o.oooooi c.cm. serum. As aruleo.ooi to o.oi c.cm. are employed. (c) An incubation period (the preanaphylactic state) is always neces- sary. This period varies in the widest degree, but depends primarily upon the amount of antigen first injected. With guinea-pigs, it is never less than seven days. If o.oi to o.i c.cm. serum is injected, symptoms may usually be stimulated after ten days. Very large doses as well as very minute ones increase the length of the preanaphylactic state very mark- edly (as long as three months). (d) The susceptibility of the various animal species differs greatly. The most suitable is the guinea-pig, the rabbit far less so. (e) The actual anaphylactic shock is dependent upon the quantity of the second dose injected and the mode of injection. The intravenous path of a massive dose is the most reliable; then comes the intraperitoneal and then the subcutaneous method. While with the intravenous procedure the final result is usually death, this is almost never so with the subcuta- neous injection. Here the anaphylaxis expresses itself in a more marked local inflammation, edema and eventually necrosis. 224 ANAPHYLAXIS An animal that recovers from the second injection becomes Anana- immune to further administration of the same proteid. This phylaxis. non-susceptible condition has been wrongly termed antiana- phylaxis. It is much more reasonable to speak of "anana- phylaxis," since the absence of hypersensitiveness is not due to the neu- tralization of the bodies necessary for bringing about the hypersuscepti- bility. The ananaphylactic state sets in as early as two hours after the anaphylactic outburst. In order to prevent the shock of anaphylaxis, it has been suggested by Besredka and Steinhardt to give the second injection during the period of incubation, i.e.) about the eighth day, or give a very minute dose of serum at the regular time of the second inoculation and give the larger dose in 24 hours. Just as an immunity may be transmitted by injecting the serum Passive obtained from an immune animal, so also can the tendency Anaphylaxis. to hypersusceptibility be transmitted by introducing into a normal animal the serum from a sensitized one, i.e., one that has been injected with a foreign proteid. This is best demonstrated by injecting the anaphylactic serum subcutaneously, followed in 24 hours by the intravenous inoculation of the respective antigen. A fully satisfactory explanation for all the phenomena of Theories anaphylaxis has not as yet been advanced. Certain it is of Anaphy- that a number of underlying factors exist which bring ana- laxis. phylaxis and immunity into close relationship. Since the term immunization usually implies a beneficial process, while anaphylaxis in most instances represents a situation of an injurious nature, v. Pirquet recommended the term " allergic" to designate the reactive changes which an organism generally exhibits after infection or injection of an antigen. The allergic phenomena are divided into those associated with diminished susceptibility, i.e., prophylaxis, and those with increased sensitiveness, i.e., anaphylaxis. Besredka adheres to the view that the anaphylactic syndrome expresses an insult to the central nervous system. He was able to show that susceptible guinea-pigs, when etherized, will bear the second inoculation of the serum perfectly well. V. Pirquet and Schick consider the precipitin action as the basis for the anaphylactic phenomena. In the main, however, there are two theories, a cellular and a humoral one. The former suggests that the hypersusceptibility is due to the stimulation of new specific receptors which remain sessile, i.e., attached to the body cells instead of being thrown off into the blood stream. When more antigen is injected, these cells, due to their greater affinity, absorb more of the toxic substance of the antigen than they do normally, and thus the anaphylactic shock is incited. The humoral theory represents the main activity within the serum. This hy- pothesis was adopted by Wolff-Eisner, a pupil of Pf eiffer, and was based upon Pfeiffer's endotoxin principle. It takes for granted that all antigens, cells and proteids, contain ANAPHYLATOXIN 225 within their bodies a toxic substance that does not form antibodies when liberated in animals. The first injection of antigen produces bacteriolytic or cytolytic antibodies possessing the power of liberating the endotoxic poisons from the proteid molecule. When the second injection is given, these bacteriolytic antibodies at once cause a rapid liberation of the intracellular toxic fraction, and injury to the animal results. Wolff- Eisner's theory can apply only in a certain number of instances, because the essential factor of cytolysis is not always present; very many bacteria, especially tubercle bacilli, are not thus broken up. This view has, however, been the fundamental thought for the later very important work of Friedemann and Friedberger. By these supporters of the humoral theory the main influence is placed upon the union which takes place in vivo between the anti- gen -f- amboceptor + complement. As a result, anaphylaxis may be incited in one of two ways. The mere absorption of the complement may bring about the anaphylactic shock. This diminution in the complement content of the serum is always demon- strable. Or, as is more probable, the union of the above three elements causes a destruc- tion of the antigen and a liberation of a toxic agent. This theory is strengthened by the demonstration that not only in vivo but also in vitro can such a toxic substance be obtained by mixing antigen -f- amboceptor + complement. In 1902 Weichardt immunized rabbits with a proteid derived from syncytial cells. By mixing the fresh immune serum with an emulsion of the syncytial cells and filtering, he derived a very toxic fluid which was named syncytio toxin. In 1909 Friedemann and Citron simultaneously isolated in vitro toxic elements which could bring about anaphylaxis. This was confirmed by Fried- berger several months later. Friedemann got his poisonous agent by mixing erythrocytes, inactive hemolytic serum and complement. Three c.cm. sheep's blood + i c.cm. inactive hemolytic serum are allowed to remain at room temperature for one-half hour, then centrifugalized. The sensitized erythrocytes are mixed with 4 c.cm. fresh rabbit's serum (complement) and placed into the water bath at 37.5° C. for exactly five minutes; then cooled down by ice and centrifugalized. The red blood cells remain almost all unhemolyzed. The superna- tant fluid is only slightly reddened and when injected intravenously into a rabbit (1310 gms.) causes general weakness, diarrhea and death on the following day. It is to be observed that the anaphylatoxic substance was formed even though' no hemolysis took place. Citron accomplished the same end by a mixture of tuberculin, serum from a tuberculous individual who spontaneously produced antituber- culin amboceptors, and normal guinea-pig's serum (complement). o.i to 0.2 c.cm. T.+ 0.2 to 0.4 c.cm. antituberculin + 0.3 to 0.5 c.cm. comple- ment are mixed and incubated for one hour. The entire quantity is injected intra- peritoneally into tuberculous guinea-pigs; the animals die in four to five hours. Tuberculous animals receiving o.i to 0.2 c.cm. T. alone or o.i to 0.2 c.cm. T. -f 0.2 to 0.4 c.cm. antituberculin, remain alive. 226 ANAPHYLAXIS Friedberger demonstrated that a toxic product which he named "Ana- phylatoxin" can be obtained in vitro from every proteid, be it animal or bacterial in nature. His technique is almost identical with that of Friede- mann described above. Accordingly, Friedberger proposed a theory explaining the course and nature of all infectious diseases upon the basis of such a common anaphy- latoxin, independent of the character and virulence of the bacteria. While this hypothesis takes into consideration the older belief that disease is due to a struggle between the infectious agent and the invaded organism, it does so in a somewhat broader sense. From Friedberger's standpoint it is the unspecific anaphylatoxin which is the important factor; its vari- able quantity, which stands in direct association with the variable pro- portions of antigen, amboceptor and complement, accounts for the various pictures of the different bacterial infections. This explanation hardly suffices for all the characteristic symptoms of the infectious diseases. It does, however, account for one, the fever. That some relationship exists between hypersusceptibility and the thermal center was first observed by H. Pfeiffer and Mita. The ana- phylactic shock is always associated with a fall in temperature, and at times this may be its only manifestation. A different effect upon the temperature results according to the dif- ferent quantity of anaphylatoxin injected (Schittenhelm and Weichardt, Friedberger and Mita). Well-defined limits have been established. Aside from the effect upon the temperature, other characteristic phenomena of the anaphylactic intoxication are the incoagulability of the blood and the marked distention of the lungs as found by post mortem examination. The symptoms of active anaphylaxis differ in the various ani- Serum mals. In man very exact studies of the results of the repeated Disease, injections of foreign serum have been made by V. Pirquet and Schick. The evidences of serum sickness are numerous. Those which are pres- ent most frequently are fever, skin eruptions, swelling of the joints, glan- dular enlargement and edema. These symptoms may follow even the very first injection of serum. As a rule they develop after an incubation period of eight to ten days. Slight reddening at the point of injection accompanied by moderate swelling of the regional lymph glands appear as prodromal manifestations. The general condition of the patient is generally only very little dis- turbed in spite of the frequently associated high fever. Still there are instances, especially after the introduction of large amounts of serum, where the symptoms continue for about four to five weeks and then lead to severe disturbances. SYMPTOMS OF SERUM SICKNESS 227 The associated skin eruptions are usually of the type of an urticaria; although Hartung describes rashes simulating scarlet fever and measles. V. Pirquet and Schick consider the following as the most positive symptoms of serum sickness: 1. The occurrence of the exanthema seven to fourteen days after injection. 2. First appearance of the rash around the point of injection. 3. Regional enlargement of the lymph glands. 4. Complete absence of any changes in the mucous membrane. Measles is excluded by the absence of Koplik spots, coryza, and con- junctivitis. In scarlet fever the following symptoms help to exclude serum sickness: 1. Initial vomiting. 2. Occurrence of angina before or at the same time as the exanthema. 3. High fever. 4. The simultaneous existence of the infection among others in the hospital or neighborhood. If the serum disease does not arise after the first, but after a later in- jection, it is characterized by the absence of, or very marked diminution in, the length of the period of incubation, and in addition by increased severity of the symptoms. In the dog these phenomena have been carefully observed by Anaphylaxis Biedl and Kraus and Arthus: 3 to 5 c.cm. of horse serum is in the Dog. administered for the first injection and after three weeks 10 c.cm. are given intravenously. In about one-half a minute the dog becomes restless, begins to vomit, and has involuntary evacuation of urine and feces. This is followed by a period of excessive prostration during which the dog lies with his limbs outstretched and almost motionless as if paralyzed. After several hours the animal either begins to recover or dies. Biedle and Kraus noticed that about 15 to 30 seconds after the intra- venous reinjection, the arterial blood pressure begins to sink rapidly. This has been ascribed to the wide dilatation of the peripheral blood vessels due to a paralysis of the peripheral vasomotor system. In addition there is a. leucopenia and a diminution in the coagulation power of the bloodr symptoms which have been observed also in man. The animal par ex- cellence for anaphylaxis experiments is the guinea-pig. The rabbit is the next choice. If guinea-pigs receive as their second inoculation a large dose of serum in- Anaphylaxis travenously, they die acutely (Th. Smith) ; (animals that have not been in the sensitized bear this same quantity of serum without any disturbance- Guinea-pig whatsoever). The blood pressure first rises, then sinks rapidly. At and Rabbit, postmortem the lungs are firmly inflated (Gay and Southardt)> Death is probably caused by respiratory paralysis, Auer and Lewis having described a tetanic spasm of the bronchial muscles. 228 ANAPHYLAXIS If for the second injection only a small dose of serum is administered, or it is intro- duced intraperitoneally instead of intravenously, acute death does not result and the picture is similar to that given by the anaphylactic dog (Pfeiffer and Mita). The guinea-pigs become very restless, move about continuously and are very easily fright- ened. Their hair becomes raised, and isolated clonic muscular contractions may be observed. Continuous hiccough sets in and evacuation of urine and f eces, first formed and later fluid occurs. The abdominal muscles become spastically rigid and at times a severe pruritus of the skin probably exists. Following this transient period of excita- tion the animals enter into a stage of depression. They stagger about or fall to one side and remain for hours as if paralyzed; breathing is slow and superficial. In the midst of these symptoms the temperature falls abruptly, often as many as 7° to 13° C. Death is due to a paralysis of the peripheral blood-vessels and takes place usually in one to two hours, sometimes four to eight hours, being always preceded by Cheyne-Stokes respiration. The symptoms of the anaphylactic shock simulate very closely Relation to those occasioned by peptone poisoning, as seen after intra- Peptone venous injection of Witte's peptone, for example. Biedl and img' Kraus thus considered that anaphylaxis was a manifestation of peptone poisoning. They believed that through immunization with proteids, antibodies of a ferment nature were stimulated, and that these split up the proteids into peptones. According to these authors, animals exposed to peptone poisoning become hypersensitive, and vice versa, animals that recover from an anaphylactic attack withstand poisoning by peptones. The production of anaphylatoxin in vitro, and the demonstra- tion by Pfeiffer that peptone is formed in the test-tube during this process, further added to the support of their theory. This hypothesis cannot as yet be definitely accepted. In Toxopeptid. the first place Manwaring was unable to confirm the experi- ment that recovery of an animal from an anaphylactic shock renders it refractory toward future peptone poisoning. Furthermore, M. Wassermann and Keysser mixed kaolin + inactive immune serum + complement and obtained a poison which caused the same disturbances as anaphylatoxin. Since kaolin is no proteid and cannot be split up, the anaphylactic symptoms in this instance cannot be due to a peptone as a split product of the antigen. M. Wassermann and Keysser have a different conception of the nature of anaphy- laxis. The antigen serves in a physical chemical capacity to fix the amboceptor. By the addition of complement the amboceptor is broken up with the formation of "Toxo- peptids," and it is these which stimulate the anaphylactic phenomena. Passive anaphylaxis is explained by the passive transmission of the specific amboceptors; the antigen itself plays no chemical role; it is not split up. The difference between ana- phylaxis and immunity lies in that in the former the complement under cover of the antigen digests the amboceptor alone, while in the state of immunity the strength and number of amboceptors are very much greater and the activity of the complement extends not only to the amboceptor but also to the secondarily attached antigen. HAY-FEVER 229 It would be impractical to discuss all the theories enlisted for anaphy- laxis. Exact facts are stil] insufficient. Experimental work is constantly disclosing new ideas. More detailed reviews may be found in the following articles: H. Pfeiffer: Problem of Pro teid Anaphyl axis. G. Fischer, Jena, 1910. A. Schittenhelm: Anaphylaxis from the standpoint of pathological physiology and the clinic. Jahresbericht iiber die Ergebnisse der Immun- itats forschung. Enke, Stuttgart, 1910. J. Citron: Critical review of the problem of anaphylaxis. Fol. sero- logica, 1911, Bd. vii, H. 3. Friedberger: Anaphylaxis. Fortschritte der deutschen Klinik. Bd. ii. Urban u. Schwarzenberg, Berlin- Wien, 1911. H. Bold: Bacterial Anaphylatoxin and its Importance in Infections. G. Fischer, Jena, 1912. Besides bacterial infections, recent teaching places urticaria, Hay-fever, eclampsia, bronchial asthma and hay-fever into the class of diseases with an anaphylactic basis. This is especially applicable to hay fever, formerly considered a pure intoxication, the pollen toxin having been described by Dunbar as the etiological factor. In Germany the disease seems to come chiefly from pollen of the grasses and grains (rye pollen being most active); whereas in America, ap- parently, the most important pollen springs from the ambrosia (rag weed), solidago (golden rod), and other members of the family of the compositae. The toxin is isolated by mixing for ten hours the ground pollen with 5 per cent. NaCl solution ard 0.5 per cent, phenol at 37° C. Then, in the form of a proteid it is precipitated by the addition of eight to ten volumes of 96 per cent, alcohol and the resultant white precipitate dis- solved in physiological salt solution. Susceptibility to the pollen toxin is limited only to certain individuals. Some are influenced by the rye pollen only, others by the golden rod alone, while a third class is affected by all. The cause for this peculiar idiosyn- crasy is unknown. The majority of observers are, however, now agreed that one is dealing here with a reaction of hypersusceptibility, as was first pointed out by Weichardt and Wolff- Eisner. Only by means of antibodies does this non-toxic pollen proteid become a poison. All those who suffer from hay-fever develop a marked conjunctivitis whenever even the slightest amount of pollen proteid (i/iooo mg.) is dropped into the conjunctival sac. In addition, all the symptoms of hay- fever or asthma may be incited. Sim^ar effects are in evidence when sub- cutaneous injections are resorted to. For purposes of immunization horses are most suitable, but only those which after an injection of pollen extracts manifest a local and general re- 230 ANAPHYLAXIS action. This is found in one-third of the animals. Their serum thus rendered immune is capable of neutralizing all effects of the pollen toxin. As regards the standardization of this serum, it is effected by mixing the dosis minima certe efficax of the toxin with various dilutions of the serum and instilling the mixture into the conjunctival sac of individuals with a tendency for hay-fever. That amount of the serum which suffices to neutralize the toxic action is taken as the unit of measure. Sera of at least thirty times the unit strength are selected for therapeutic application. The immune serum is manufactured in fluid and powder form, and is placed on the market under the name of "Pollantin." Its use is mainly local, by spraying a small quantity of the pollen powder upon the nasal mucosa several times daily and by placing several granules into the conjunctival sac with a cameFs-hair brush. The serum can also be em- ployed as a prophylactic. If the eyes are especially reddened, it is best to deposit some fluid serum into the conjunctival sac every day. Prausnitz advises the injection of i to 2 c.cm. of the serum subcutaneously when asthmatic attacks occur, when the above local treatment has failed. In America a special Pollantin is made against the frequent form of hay-fever known as "autumn catarrh" by immunization with the pollen of the golden rod and rag weed. The pollantin therapy and prophylaxis has been quite satisfactory, inasmuch as two-thirds of the patients remain either entirely free from attacks or are so greatly benefited that their general duties are not inter- fered with. The specificity of anaphylaxis has been proposed for diagnostic aid. As yet the methods have not been perfected sufficiently to be of clinical value. CHAPTER XVIII. PASSIVE IMMUNIZATION (SERUM THERAPY). BACTERIOLYTIC SERA. SPECIAL SERUM THERAPY. In the former chapters it was proven that during active immunization of an animal specific protective bodies were formed which circulate in the blood and can by means of the serum be transferred to another organism. It was further found that such bodies exert this protection against fatal intoxication or infection in various ways; thus, as antitoxins and anti- aggressins they neutralize toxic poisons and aggressins; as bacteriolysins they bring about lysis of the bacteria; while as bacteriotropins they prepare the bacteria for phagocytosis. The defensive qualities of such a transferred serum are evident not only if the infection is incited at the same time as, or a short period after the serum is given, but in numerous instances curative effects are observed if the serum is given even after infection has already taken place. Of all sera, those with antitoxic properties have met the greatest suc- cess in therapeutic application. They have already been referred to in their respective chapters. The efficiency of the pure bacteriolytic sera on the other hand has been dis- appointing. One reason given for this lack of curative action is, the in- ability of bacteriolytic serum to neutralize the endotoxins. Pfeiffer's experiment revealed that if the number of bacteria exceeded a certain limit, then in spite of bacteriolysis, death of the animal takes place. This was explained by the existence of endotoxins. By bacteriolysis the endotoxins normally confined within the bacteria are liberated and thus get a chance to show their toxicity. The aim, therefore, was to produce antiendotoxic sera. The accomplishment of this was prevented by the erroneous view of Wolff-Eisner who claimed that it was impossible to immunize against endotoxin. Numerous methods have been advocated for the liberation of these endotoxins: maceration of bacteria, exposure to very low temperature, admixture with chemical substances which would dissolve the outer capsule, ferment digestion, growth upon certain culture media, etc. At the present day, there is absolutely no doubt that the bacterial bodies contain poisonous substances against which it is difficult and to a cer- tain degree impossible to attain an immunity. Whether one should adhere to the old idea and apply to these the term endotoxin, or include them in the class of true toxins with the only difference that they are not secreted but contained within the bacterial body and therefore more difficult to isolate, is purely a question of theoretical importance. Another cause for the therapeutic failure of bacteriolytic serum, as ad- 231 232 PASSIVE IMMUNIZATION vocated by Bail and his school, is the lack of its antiaggressin action. This applies only to the cases in which the bacteriolytic serum was produced by immunization with dead bacteria. When live bacteria are used, this objection is not to be considered, as according to the experiments of Wassermann and Citron, "aggressin" is nothing more than the immunizing substance of the living bacteria. As far as the structure of the antiaggressins is concerned, the author was able to show that like the bacteriolysins, they are amboceptors which bind complement. Artificial aqueous extracts of living bacteria, belonging to the class of half parasites made according to the method of Wassermann and Citron, contain the endotoxin as well as the aggressin. Such artificial aggressins, therefore, represent ideal antigens. The sera produced by their injection contain but few bacteriolytic bodies and a very large number of ambo- ceptors, easily demonstrable by the Bordet-Gengoa reaction. Wassermann explains the lack of therapeutic efficiency on the part of the bacteriolytic sera by the absence of complement of the organisms, as well as by the inability of human complement invariably to fit animal amboceptors. As is known, amboceptors increase during immunization while the complement content remains the same. But since amboceptors without complement remain inactive, even a very strong serum may only be slightly effective, depending upon the amount of existing complement. If too many amboceptors are injected, the serum may become entirely powerless due to a phenomenon similar to Neisser and Wechsberg's complement deviation. Wassermann advises therefore the addition of complement to a serum before its injection, in order to activate it. This suggestion has not been widely adopted in practice. It is for a similar reason, that the classical experiment of bacteriolysis is so beautifully demonstrable in the guinea-pig's peritoneal cavity, an area relatively poor in cells, while this phenomenon is incomplete and replaced by phagocytosis when occurring in the blood, inner organs, and subcutaneous connective tissue. It is in this connection that Metch- nikoff and his followers see the main reason for the failure of the therapeutic activity of bacteriolytic sera. An additional impediment is offered by the wide differences which exist among the numerous strains of the same bacterium. This may be so marked that an immune serum produced with one strain will offer no protection against a different strain of the same bacterium. It is now overcome to a certain extent by immunization with as many different strains of the same bacterium as possible (polyvalent sera). Cultures grown upon artificial media for a very long time adapt them- selves to their new surroundings and frequently lose some of their bio- logical characteristics, e.g., virulence. If the culture is then inoculated POSSIBILITIES OF SERUM THERAPY 233 into an animal, the virulence is increased usually only for that animal species, but may remain the same or even lowered for man. Many authors, therefore, employ for the production of immune sera only virulent strains of bacteria freshly isolated from a human being. In spite of all the above considerations, the fact still remains that most immune sera excepting those of the cholera, typhoid, and paratyphoid bacteria, show no bacteriolytic tendencies even under the most favorable circumstances; but by means of their amboceptors they fix free comple- ment and with the aid of bacteriotropins, stimulate phagocytosis. Whether complement fixation is at all to be considered as a protective phenomenon, cannot with the evidence existing at present be definitely decided. Conditions are much more favorable as far as the bacteriotropins are concerned. Active phagocytosis is always an expression of good resistance. It is not necessary for the leucocytes to digest the bacteria; it is amply sufficient if a protective wall of these cells is formed (Ribbert, Citron, Gruber) ; moreover they can neutralize the bacterial poisons. In this connection it must always be borne in mind that phagocytosis by no means necessitates the death of bacteria. Granting, however, that all the above requirements have been fulfilled and a suitable serum has actually been produced, will such a serum always be effective, or are there any other causes which may interfere with its good results? In order to answer this, the infectious diseases must be divided into acute and chronic. With the first class, success is quite assured as long as it is possible to bring sufficient amounts of the active serum substances into direct contact with the bacteria. In meningeal infections, subdural injections may have to be adopted. It is difficult, however, in cases of this nature to judge definitely whether the serum therapy was really the effective agent, inasmuch as diseases like erysipelas, meningitis, pneumonia, etc., are self limited, lasting for a period of time and then subsiding of their own accord. With the chronic infections, on the other hand (especially tuberculosis), serum therapy has a new difficulty to overcome. As a result of the long course of the disease, it is naturally impossible by means of a single injection to introduce sufficient curative bodies, as can be done in diph- theria, for example. It is necessary, therefore, frequently to repeat the injections. Under such conditions the human organism produces anti- bodies against the foreign proteid (anaphylaxis) , perhaps even against the curative substances in the serum (antiamboceptors) . In both instances the desired effect of the serum is lost. A further impediment lies in the possibility that bacteria remaining in a system for a long period immunize themselves, and thus resist the action of the antibodies directed against them. Such bacterial strains are known as " serum fast." 234 PASSIVE IMMUNIZATION Special Serum Therapy. i. Meningococcus Serum. — Numerous investigators have attempted the Meningococ- production of an immune serum for man, among these Jochmann, the cus Immune Berlin Institute for infectious diseases, Ruppel, Kraus, Flexner and Serum. Jobling, and others. The sera of Jochmann (Merck) and Ruppel (Hochst) are produced by immunization of horses with meningococci which are at first employed in dead, and later in live form. The other sera mentioned are obtained by immunization with bacterial extracts or bacterial extracts plus full bacteria, and therefore contain agglutinins, precipitins, bacteriotropins, amboceptors and antiendotoxins. It is difficult to test the efficiency of these sera in animals, as the meningococci vary greatly in their virulence toward them. Jochmann and Ruppel assert that they have been successful in growing cultures extremely virulent for ani- mals which they employed for the titration of the therapeutic value of the serum. In the institute for infectious diseases, the method of complement fixation is taken as the index of the therapeutic value of the serum. This procedure is very unreliable. The protection of the serum in mice against the meningococcus endotoxin as well as the demonstration of the bacteriotropic action of the serum is far more significant. In man, the immune serum is injected subdurally, after a quantity of spinal fluid has been withdrawn to relieve the pressure. In adults 20 to 40 c.cm. and in children 10 to 20 c.cm. are daily injected until there is clinical improvement or a fatal prognosis becomes inevitable. It is ad- visable to precede the serum inoculation by a morphine injection, and to elevate the pelvis for eight to twelve hours after the inoculation. The earlier the serum therapy is instituted, the more favorable are its results. Subcutaneous applications of the serum or employment of a serum more than three months old is absolutely of no use. Both in the United States and in foreign countries the value of the serum as a thera- peutic agent seems fairly established. In Germany, the serum is obtained gratis at the institute for infectious diseases at Berlin. The serum in Switzerland is distributed by the serum institute of Bern (Kolle). In the United States, the Rockefeller Institute of New York first conducted its dispensation, but now it is under the supervision of the New York Board of Health. Numerous statistics can be cited exemplifying the good results of the serum. The following figures given by Levy describing the experiences in the Essen epidemic are especially instructive : From the first of January until the first of November, 1907, the total number of epidemic meningitis cases which occurred in Essen were: 55 cases with 29 deaths = 5 2.72% mortality, of these, treatment was given outside of the barracks to 15 cases with 12 deaths = 80% mortality, inside the barracks were treated 40 cases with 17 deaths =4 2. 5% mortality, of these 14 cases were not treated with serum with n deaths as a result = 78.6% mortality, SERUM THERAPY IN CEREBRO SPINAL MENINGITIS 235 those treated with serum were 23 cases with 5 deaths =2 1.7% mortality, of these, those which were treated only incompletely (subcutaneously) and with insuffi- cient doses, numbered 6 cases with 3 deaths as the outcome =50% mortality, systematic subdural treatment with large doses. 17 cases with 2 (i) deaths=n.8 (6.3)% mortality. The figures in parentheses represent the moribund cases coming under treatment and the percentage which would result if these were not included in the calculation. The experiences with the serum of Flexner and Jobling are similarly encouraging. In a report of 400 cases (1909) the mortality is reported as lowered from 80 per cent, to 20 per cent. In a more recent report (1913), 1,294 patients treated with serum show the following results: recovered 894; died 400 (30.9 per cent.). Flexner's tables which are here reproduced are instructive. MORTALITY ACCORDING TO THE PERIOD OF INJECTION OF THE SERUM. . Period of Injection No. of cases Recovered Died Per cent, recovered Per cent, died ist to 3d day 100 163 36 81.9 18.1 4th to 7th day 346 252 94 72.8 27.2 Later than 7th day. . . 666 423 243 63.5 36.5 Totals. . I 211 8^8 •27-7 60.2 30.8 030 MORTALITY ACCORDING TO AGE. Age No. of cases Recovered Died Per cent, recovered Per cent. died Under' i year 1 20 6s 64. SO 4 49. 6 i to 2 years 2 to 5 years 5 to 10 years 10 to 20 years. . . . 87 194 218 360 60 i39 185 2C4 27 55 33 106 69.0 71.6 84.9 70. 6 31.0 28.4 i5-i 29.4 Over 20 years Age not given 288 18 180 ii 108 7 62.5 61.1 37.5 38.9 Totals I 2Q4. 804. 400 . .• 69.1 30.9 236 PASSIVE IMMUNIZATION MORTALITY ACCORDING TO AGE AND PERIOD OF INJECTION. Injected ist to Injected 4th to Injected later than 3d day 7th day 7th day T3 -a — Age £ .- •3 « .22 8 . 1 o 8 1 M 6 8 | . Q j! ,_, 1 tt Q *, « Q CL, 5 g Q fS Under 2 years . . . 13 12 i 7-7 37 28 9 24-3 159 81 78 49.1 2 to 5 years . . . 3° 24 6 20. o 66 49 17 25-8 93 63 30 37-3 5 to 10 years 55 49 6 10.9 69 61 8 n. 6 77 61 16 20.8 10 to 20 years 67 58 9 13-4 106 73 33 3I.I 171 115 56 32.7 Over 20 years .... 34 20 14 41.2 65 39 26 40.0 164 101 63 ! 38.4 l Totals . . 199 163 36 18.1 343 250 93 27.1 664 421 243 36.6 2. Streptococcus Immune Sera. — The role of the streptococcus in some diseases, for example, scarlet fever, is imperfectly understood. Moreover it has only been in- definitely established whether there are various groups or only one kind of strepto- coccus; even the significance of their virulence or hemolysin formation is not clear. These difficulties account for the great number of methods advocated for the pro- duction of an immune streptococcus serum. The oldest, serum is that of Marmorek. It was produced by immunization with a strain made highly virulent by passage through animals. The other sera on the market are: a. Serum of Aronson (Schering). — This is a polyvalent serum produced by immuni- zation of horses with cultures pathogenic for man; some strains having previously been passed through animals, others not. The strength of the serum is tested in mice infected with the latter strains. b. Serum of Meyer-Ruppel (Hochst Farbwerke). — Horses are first immunized with a strain of streptococcus whose virulence has been raised by passage through horses and mice; each horse is then injected with a different strain of human streptococcus. When the serum of each animal is of such a strength that doses of o.oi to 0.0005 c.cm. protect mice infected with its own particular strain, the sera of the different horses are mixed. Thus a polyvalent serum is obtained. c. Serum of Menzer (Merck) is monovalent and produced by immunization with a culture which is pathogenic for man and not passed through animals. d. Serum of Moser is polyvalent, produced by injections of streptococci from scarlet fever. The sera of Menzer and Moser are not tested by injections of white mice. The others are. One cannot strictly rely upon this method of serum titration for its employment in man. The virulence of streptococci against mice and human beings bears no definite relation. A serum may be perfectly efficient in mice both for prophy- lactic and therapeutic purposes, and be entirely inactive in man; also vice versa. The action of the serum should be in the main of bacteriotropic nature. Antistreptococcus serum has been tried in scarlet fever, puerperal sep- sis, erysipelas, and articular rheumatism. Complement fixation experiments (Foix and Mallein, Schleissner) STREPTOCOCCUS AND PNEUMOCOCCUS IMMUNE SERA 237 have shown that the streptococci of scarlet fever can be definitely sepa- rated from the other varieties of these bacteria. This work could not be confirmed. In this disease favorable results have been observed by the use of Moser's serum. Escherich states that of 112 scarlet fever cases injected, those receiving the serum on the first and second days of their illness all recovered, while of those injected later on, there was a high percentage of mortality. Other authorities have seen no, or only very slight effect from the serum treatment. Two hundred cubic centimeters of Moser's serum must be given subcutaneously. The treatment of puerperal fever has been favorably influenced by Aronson's and Meyer-Ruppel's serum, of which 50 c.cm. are injected on several successive days. Menzer's serum is said to serve its purpose best in acute and chronic rheumatism as well as in tuberculous mixed infections. In erysipelas all the well known sera have been tried. On account of the very variable course of the disease it is difficult to judge the exact value of the serum. In fact, thus far one cannot with certainty depend upon any serum treatment of a streptococcus infection, but the serious nature of such infections makes every possible therapeutic measure strongly justifiable. 3. The pneumococcic sera most frequently used are those of Pane, Pneumococ- Romer and Merck. Pane immunizes donkeys with highly virulent pneu- cus Immune mococci and uses the serum for the treatment of pneumonia. Several Sera. Italian investigators record favorable results. Romer prepares a polyvalent serum by injecting horses with different strains of pneumococcus obtained directly from man; the strength of the serum is tested in mice. The serum is mainly employed both for the protection and cure of ulcus cornea serpens. The result according to Romer depends upon the very variable virulence of the pneumococci. The severity of the infection in man is said to run parallel with the virulence in mice. Romer, therefore, ascertains in every case of ulcus serpens whether his serum has any protective bodies for that particular strain of pneumococcus, and tests the virulence of the same. The serum can be injected intravenously and subcutaneously, and in pneumococcus meningitis, subdurally. It is manufactured by the Hochst Farbwerke, in vials of 10 and 20 c.cm. A similar serum is manufactured by Merck. It is obtained from horses and stand- ardized at the Institute for experimental therapy in Frankfort so that o.oi c.cm. injected subcutaneously protects a mouse inoculated intraperitoneally 24 hours later with 10 to 100 times the lethal dose of a living pneumococcus culture. This is known as a normal serum and i c.cm. contains one immunity unit (I. E.). The serum on the market contains 20 to 40 units per c.cm. In pneumonia 200 to 400 units are given subcutaneously and repeated in three to four days, if the fever does not subside. As a prophylactic inoculation, 200 to 400 units are given to old people where a "hypostatic" pneumonia is feared. In ulcus serpens of the cornea 200 to 400 units are employed and if no improvement sets in, the dose is repeated upon the third day. In addition, several drops of the serum are 238 PASSIVE IMMUNIZATION instilled into the conjunctival sac every two hours. As a prophylactic dose in this disease, 100 units suffice. Merck also prepares a vaccine of dead pneumococci in doses of i c.cm. which further aid in the treatment of pneumococci infections. One cubic centimeter of such dead pneumococci can be administered for the prophylaxis of ulcus serpens. 4. Pest Sera. — A large number of pest sera are in use. a. The Paris serum (Yersin) produced at Pasteur Institute by immunization of horses with dead and later on living bacilli. b. The Bern serum of Tavel employs the same principles. c. Lustig's Serum. — For this serum, horses are immunized with the pest-nucleo- proteids. Pest cultures are broken up by i per cent, of potassium hydroxide and from this, by the addition of acetic acid, the nucleoproteid is precipitated and then suspended in salt solution to serve as antigen. d. Serum of Terni-Bandi is prepared by the immunization of donkeys and sheep with natural pest aggressins. e. Serum of Markl is supposedly an antitoxic serum prepared by immunization with nitrates of old pest bouillon cultures. All the above sera contain agglutinins, precipitins, bacteriotropins and ambocep- tors; the serum of Terni-Bandi contains aggressin amboceptors, that of Markl, anti- endotoxins. The sera are tested for their anti-infectious properties in animals such as guinea- pigs, rats, mice. Markl also estimates the toxin neutralization power of his serum. The Paris serum comes either in dry form or in bottles containing 20 c.cm. without any preservatives. Ten to 20 c.cm. should suffice as a prophylactic injection, although Martini advises 100 c.cm. at least. The period of protection is short, averaging about fourteen days. Prophylactic injection is advisable in those instances where an immediate protection is necessary, like the inoculation of physicians and nurses attending pest patients. Under all other circumstances either active immunization or the simultaneous method of Shiga should receive the preference. For the treatment of pest infections, Calmette and Salimbeni advise intravenous administration of 20 c.cm. and two subcutaneous injections of 40 c.cm. each — all to be given on the first day; on the second day two similar subcutaneous injections; and if the case is of a severe nature, the dose may be doubled. The results are variable. From comparative studies, it seems that Lustig's serum is somewhat weaker than the Paris serum. The sera of Terni-Bandi and Markl have not been sufficiently employed, so that opinion is reserved. 5. Tuberculosis Sera. — The best known and most studied are those of Maragliano and Jarmorek. a. Serum of Maragliano is prepared by Maragliano's institute in Genoa from horses which are immunized for about six months with the soluble substances of tubercle ba- cilli. The favorable action of the serum is reported on, especially by Italian authorities. b. Serum of Marmorek is prepared in the laboratory of Marmorek, at Paris-Neuilly, by the immunization of horses with the so-called "primitive" tubercle bacilli, i.e., young tubercle bacilli whose acid-fast character is still very slight or entirely absent. When the horses have attained a high grade of immunity, they receive injections of various strains of pure cultures of streptococci obtained from the sputum of tubercu- lous patients. The serum of these animals is, therefore, antituberculous and at the same time polyvalent antistreptococcic (a double serum), serving against the mixed infections. This serum is administered daily, either subcutaneously 5 to 10 c.cm. or per rectum SPECIFIC SERUM THERAPY 239 20 c.cm. The latter form is more advisable for the sake of preventing anaphylaxis. Citron has found the serum entirely harmless, the bad effects described by some being probably due to the idiosyncrasy of patients against foreign sera. The most favorable results have been claimed in localized bone and joint tuberculosis and in the incipient stages of pulmonary tuberculosis. Especial consideration of the serum should be given in those patients who evince persistent temperature or the very severe but not hopeless cases, where the tuberculin therapy cannot be undertaken. In some of these instances very encouraging results have been noted. Occasionally the author started with the serum treatment, and then combined with it the tuberculin administration and finally left the serum entirely. 6. Anthrax Sera. — Sclavo, Deutsh, Sobernheim and others have produced immune sera by the immunization of donkeys, sheep and horses. These have been mainly employed in veterinary practice. In man the serum has been tried only by Sclavo. He injects 30 to 40 c.cm. sub- cutaneously for several successive days; in severe infections 10 c.cm. are administered intravenously. Two cases described by Bandi received 150 c.cm. intravenously. 7. Typhoid Immune Sera. — The ordinary bacteriolytic sera (Tavel) have not met with the desired success in the therapy of typhoid fever. Attempts have, therefore, been made to produce antiendotoxic sera. Chantemesse treats horses for several years with bouillon nitrates; Besredka injects first dead and then living typhoid bac- teria from agar cultures, Mac Fadyen breaks up the bacteria at very low temperatures and thus liberates the endotoxin for purposes of immunization. Kraus and von Sten- itzer use bouillon filtrates and aqueous bacterial extracts as is likewise done by Meyer- Bergell and Aronson. Garbat and Meyer employ sensitized typhoid bacilli, i.e., bacteria united with their bacteriolytic amboceptors. Chantemesse injects several drops of his serum subcutaneously. Its effect lasts ten days. Only occasionally is a second inoculation necessary; if so, it must be much smaller. His results have been good and have mainly depended upon an increase in the opsonic index. Meyer and Bergell as well as Kraus give 20 to 50 c.cm. subcutaneously. 8. Cholera Serum.- — Similar attempts for the production of a cholera antiendotoxic serum have been made. "Kraus has succeeded in obtaining an antitoxin against some El-Tor vibrios which have all the characteristics of true cholera vibrios. The experiments with Kraus' serum, and Kolle's serum (Bern Institute), at present being conducted in Russia, seem to be favorable. The serum therapy of infectious diseases is still in its prime. The contradictory results of many authors are to be attributed not only to the variable efficiency of the sera, but also to the method, the time, and the dose chosen for administration. The same serum in the hands of different physicians may yield opposite results. These subjective sources of error must be overcome or minimized by making a complete and thorough study of the effects which a certain serum may have and actually does have; here all the clinical and laboratory guides must be made use of. Employed in this manner, serum therapy will even at the present stage lead to beneficial results. Wright's motto at the beginning of his book on vaccines "The physi- cian of the future will be an immunizator," can justly be reversed to read, "the immunizator of the future will be the physician in the true sense." CHAPTER XIX. CHEMOTHERAPY. DEFINITION. METHOD. ATOXYL. SALVARSAN. CHEMOTHERAPY or MALIGNANT TUMORS. CONCLUSION. Serum therapy proved the fundamental fact that it is possible by the injection of specific sera selectively to destroy or counteract the poisonous effects of certain micro-organisms without in any way injuring the infected host. That similar results are attainable by chemical means is demonstrated by the empirical use of quinine in malaria. For a long time this chemical specific stood in a class by itself. In recent years, the progress made in the study of infections by the protozoa, especially the trypanosomes and spirochetes, and the possibility of transmitting these diseases to the lower animals, stimulated a renewed effort in quest of chemical substances analogous to quinine. Paul Ehrlich led the way in this new direction and termed this study " Chemotherapy " in contra- distinction to "Pharmacotherapy." Only those agents can be employed chemotherapeutically in which the " organatrope" and "parasitatrope" relation is favorable, that is, primarily when the curative dose is only a very small fraction of the toxic dose (Ehrlich). In infectious diseases it is the causative parasite that is aimed at, in malignant tumors the tumor cells are the objects for de- struction. In both, however, it is absolutely essential that the normal tissues of the body remain entirely uninjured. Thus far the most favorable results of chemotherapy have been ac- complished in trypanosome and spirochete infections. Researches under Ehrlich's direction clearly defined that there are three different types of substances which can destroy the trypanosomes : (a) The group of basic silk dyes (Fuchsin). (b) The group of cotton dyes (Benzopurpurin series) of which trypan red and try- pan blue have proved most efficient. (c) The group of arsenical products (atoxyl and its derivatives). The basis for differentiating these three classes is seen from the following experiments. If a mouse suffering from trypanosomiasis receives an injection of an active fuchsin preparation, the trypanosomes disappear from the blood. They remain away permanently (" Sterilisatio magna") provided that the injected dose is large enough. The animal is thus cured by a single injection. If the dose is not sufficient, many 240 PRINCIPLES OF CHEMOTHERAPY 241 of the parasites are destroyed, but some remain alive. This number may be so small that for several days the blood may seem sterile, but then the remaining trypanosomes multiply and a relapse occurs. The same fuchsin preparation is again injected and now, one may attain a complete cure if the dose this time is sufficiently large; but if not a relapse sets in. Thus the treatment is continued. If, however, the relapses are numerous and the same curative agent is repeatedly employed, a state is reached in which the fuchsin no longer has any beneficial influence whatever. The trypanosomes have become immunized or acclimated to fuchsin, in other words, have become "fuchsin-fast." To prove that it is the trypanosomes that have acquired a new characteristic and not that the organism of the mouse had become altered in its susceptibility, one may infect for the first time another mouse with this " f uchsin-f ast " strain, and here again this dye will have no effect at all. The artificially attained resistance toward fuchsin remains as a permanent characteristic of this particular strain of trypanosomes, even though transplanted from mouse to mouse for years. It disappears, however, if the parasites are allowed to increase by a sexual cycle of development which is possible for example in the rat louse Hematopinus spinolosus. This fuchsin-fast character has not at the same time altered the susceptibility of the trypanosomes toward any of the trypanocidal agents of the second or third group as trypan red or atoxyl. Just as guinea-pigs immunized against cholera may succumb to infection by the typhoid bacillus, so also may trypanosomes, which have become inert to fuchsin, be attacked by trypan red or atoxyl. Furthermore the acquired resistance of the protozoa to the action of these various chemicals is just as specific as is the immunity of an animal toward bacteria. Group reactions play a role in chemotherapy just as they do in im- munity; the individual vaccinated against cow pox becomes immune toward small-pox; the strain of trypanosomes which has become resistant toward trypan red can no longer be affected by trypan blue. Ehrlich's conception that it is necessary for substances to be taken up by the parasites before they can be acted upon destructively is concisely expressed by him: " Corpora non agunt nisi fixata." In terms of the side chain theory, the protoplasm groups which have a specific affinity for the chemical radicles are known as chemocep- tors. Accordingly, the trypanosomes may be said to possess three distinctly different chemoceptors, one directed against fuchsin, one against benzopurpurin and another against atoxyl. While the early chemotherapeutic studies with the aniline dyes were mainly of theoretical interest, the arsenic derivatives soon claimed their position of practical importance. In 1902 Laveran and Mesnil discovered that arsenous acid could 16 242 CHEMOTHERAPY destroy the trypanosomes in the blood of infected mice; but at the same time it acted as a severe general poison. Atoxyl was then Atoxyl. introduced by Ferd. Blumenthal and was found to be less poisonous and just as efficient. Extensive experiments were undertaken with this drug and its effects both in infected mice and human beings (sleeping sickness) carefully studied. Most of this work was carried out in the infected districts of Africa by Robert Koch and his co-workers. While the trypanocidal action of atoxyl was distinct, its effect was not permanent and relapses occurred. Uhlenhuth tried this arsenic compound in animal syphilis, basing his application upon the apparently close relationship existing between the spirochete and trypanosome as pointed out by Schaudinn. The results were very encouraging, so that for a time atoxyl was employed in the treatment of human syphilis with beneficial effect upon its clinical manifestations. Its use had to be abandoned, however, on account of its very severe asso- ciated toxic actions, of which optic atrophy was the most important. Up to that time the chemical compositioji of atoxyl had been repre- sented as meta-amino-phenylarsenic acid. Ehrlich and Bertheim found that this was incorrect and that in reality it was p-amino-phenylarsenic acid, called by Ehrlich arsanilic acid. It was thus made possible by systematic substitution to attain specifically active but less poisonous products. For a long time too it was unexplained why the trypanocidal action of atoxyl was so very marked in vivo and yet no effect could be noticed in vitro. Ehrlich presumed that when atoxyl enters the system it undergoes the chemical change of reduction; from a quinquivalent arsenic combina- tion a trivalent one resulted; and it is this reduction product to which atoxyl owed its activity. He proved this hypothesis by reducing atoxyl in the test-tube; he thus obtained the phenylarsenious acid which even in the dilution of i to i ,000,000 was capable of destroying the trypanosomes, whereas atoxyl in the dilution of i to 100 was inactive. " The arsenoceptor cannot unite with quinquivalent arsenic products, but requires ones with trivalent arsenic; this lack of saturation of the arsenic molecule is essential for the chemical union with the arsenic receptor" (Ehrlich). The numerous trivalent arsenic combinations are by no Arseno- means equally potent. Many are entirely unable to destroy phenyl- the protozoa, others are slightly active, while only few show glycm. a distinct specific action. Belonging to the last class is arseno- phenylglycin which contains an acetic-acid radical. This com- pound is peculiar in that it can attack trypanosomes which have become resistant to atoxyl. If, however, these " atoxyl-f ast " trypanosomes are treated for a long time with arsenophenylglycin, they became resistant also to the latter compound. According to Ehrlich's explanation for this VARIOUS ARSENIC COMPOUNDS 243 phenomenon, the trypanosomes possess in addition a chemoceptor with an affinity toward acetic acid, and therefore any strain which has become resistant toward arsenophenylglycin has become not only arsenic fast but has become immune to all combinations which like arsenophenylglycin contain an acetic acid radicle. The action of an arsenical compound depends upon its being anchored not only by means of one of its groups but by many groups, "Just as a butterfly that is to be mounted must be fixed by more than one attach- ment. First a needle must be passed through its body and then successive fixation through the animal's wings is necessary." In this sense Ehrlich speaks of primary and secondary haptophores. In arsenophenylglycin the primary anchoring group is the receptor of the acetic acid radical which delivers the arsenic element to the cell. Whereas for the try- panosomes it is this acid group which plays the main role, for the spiro- chete, it is the hydroxyl group, placed in the para-position of the trivalent arsenic molecule (arsenophenol). As = As \/ \/ OH OH Arsenophenol. By the introduction of different elements or combinations into the above formula (as iodin or an amido group) the toxic action of the com- pound can be further diminished, whereas the spirillocidal property is greatly enhanced. In this manner, Ehrlich and Bertheim developed the preparation 606 (salvarsan) which is the dihydrochloride of dioxy- diamidoarsenobenzol. As = As NH NH, OH OH As =•• As C1HNH; NH2C1H OH OH Dioxydiamidoarsenobenzol. Dioxydiamido-arseno-benzene dihydrochlor- ide (Salvarsan). Dioxydiamidoarsenobenzol is a canary yellow powder undergoing oxidation very readily so that it is kept in small vacuum tubes. It is insoluble in water but soluble if sodium hydroxide is added. The hydrochloric acid salt, salvarsan, is soluble in water giving a clear yellowish solution. It is acid in reaction and absorbed by the system with difficulty (Hata). On the addition of sodium hydrate first a monochloride and then a neutral salt in the form of a flocculent precipitate results. More of the hy- droxide redissolves, the precipitate giving a clear solution of the alkali salt; its reaction is strongly alkaline. It makes no difference whether the dioxydiamidoarsenobenzol is dissolved in sodium hydrate or whether the hydrochloric acid salt is made alkaline 244 CHEMOTHERAPY by adding the hydroxide: in both instances the sodium salt of dioxydiamidoarseno- benzol is formed. As = As fH2 ONa ONa The toxicity of salvarsan is comparatively mild. The following table of Hata shows this in detail. Animal Mode of application Dose tolerated Mouse Subcutaneous i : 300 pro 20 g. Mouse Intravenous. . i : 3f , 175-177 Mercurial therapy (biological) 168, 169 combined with salvarsan, 245, 246 Metchnikoff's resistance test, 138, 196-197 Microcytase, 152, 198 Mouse typhoid bacilli, 31, 70, 113 Much and Holzmann reaction, 99 Multipartial sera, in, 112 X Nastin, 72 Negative phase, 78, 200 Neosalvarsan, 254-255 Nephrotoxin, 151 Neurorecurrences, 247-248 ("neurorezidive") Neurotoxin, 86, 87, 97, 99, 151 Neutral red for vital stain, 198 New tuberculin, 62-63, 69-71 Nonagglutinable strains of bacteria, 112, 113 Nonbinding doses, 156-158, 174 Normal bacteriolysins, 138 Normal curative serum, 79 Normal hemolysin, 144 INDEX OF SUBJECTS 26l Normal loop, 10, 19, 20 Normal toxins, 79 O Obtaining blood, 12-14 Ointment reaction, Moro, 50 Oleic acid, 98-99 Ophthalmo reaction, 52-54, 160 Opsonic index, 199-209, 212, 213 definition, 199 diagnostic importance of, 201 increase by immunization, 200-201 relation between index and clinical con- dition of patient, 201, 212, 213 technique for determination of, 203-209 variations by auto-inoculation, 201, 202 Opsonins, 198-213 Opsonizer, 205 Organotrope, 240 Original tuberculin, old, 61 Ozena bacilli, 123 Paradoxical reaction, 240 Paralysis, progressive, 164, 165, 167 Parasites, 23, 41, 232 half, 23, 41, 232 total, 23, 41 Parasitotrope, 240 1 Parasyphilitic infections, 246 Paratyphoid bacilli, 30, 110-113, 140, 143 Paroxysmal hemoglobinuria, 100, 101 Partial agglutinins, no Partial aggressins, 62 Partial immunization, 63, no Pathogenicity, 23 Peptone poisoning in anaphylaxis, 228 Peritoneal exudate, 136 method for removal, 12, 136 Pernicious anemia, 100 Pest, 114, 153, 172, 238 sera, 114, 131, 238 Pfeiffer's phenomenon, 131 technique and practical application, 134- 140 in cholera, 139 Phagocytosis, 139, 196-215 cells engaged in phagocytosis, 196-197 definition, 196 demonstration of, 197, 198 object of phagocytosis, 196-197 during artificial immunity, 196-213 ' during natural immunity, 4 Phagolysis, 197 Phrynolysin, 96 Phytotoxin, 95-96 Pipettes for removal of peritoneal exudates, 136 v. Pirquet's reaction, 48-50, 55 Pneumococci, 32, 199, 237-238 Pneumococcic sera, 119, 237-238 ' ., Pneumonia, 102 chemotherapy of, 255 bacilli, 123 Pollantin, 230 Pollen poison, 229-230 Polyvalent sera, in, 232 Porges reaction, 123 Positive phase, 78, 200 Preservation of serum, 15 in "Frigo," 16 Precipitation, 120-130 Precipitate in precipitin reactions, 130 Precipitins, 120-130 biological structure of, 120, 160 phenomenon of precipitation, 120 bacterial, 121-124 clinical value of in typhoid, 122, 123 in syphilis, 123, 124 Fornet's ring test, 123 group reactions of, 122 Klausner's reaction, 124 Porges' reaction, 123 precipitating antisera for, production of, 121 precipitinogens, 121 proteid, 124-130 action of, 124, 125 determining nature of meat, by, 125, 127, 128, 129 differentiation of proteids by Uhlen- huth's method, 127-128 distinguishing blood of animal species, 125, 126 production of precipitating antisera, 126, 127 Prognostic employment of autoinoculation, 201 of the lues reaction, 167, 169, 185 of the tuberculin reaction, 58-59 Prophylactic inoculations in cholera, 35 hog cholera, 31 hydrophobia, 28 small-pox, 27 typhoid, 33-35, 44 . Prophylaxis in diphtheria, 84 in dysentery, 90-92 against hay fever, 230 in pest, 238 in tetanus, 88 in ulcus corneae, 237 Proteid differentiation by complement fixa- tion, 193-195 by precipitins, 124-130 Proteid anaphylaxis, 221-230; see "anaphy- laxis" Pseudo-agglutination, 108 Psycho-reaction, 99 Puerperal sepsis, 237 Pukal filter, 17 R Rabbit sepsis, 21 Rabies, 26-30 simultaneous treatment of, 30 vaccination against, 28-30 Rat trypanosomiasis, 22, 240, 241 "Reagine," 166, etc. Receptors, 133, 147, 151, 158-160, 231, 232- 233 Reichel filter, 17 Relapsing fever in mice, salvarsan therapy, 244 Resistance test of Metchnikoff, 138 Rheumatic fever, 236 Rhinoscleroma bacilli, 123 Ricin, 95 Ring test, 123 262 INDEX OF SUBJECTS Saprophytes, 23 Salvarsan, 243-255 chemical formula and properties, 243- 244 therapy in hen spirillosis, 244 therapy in relapsing fever, 244 therapy in syphilis of man, 245-255 therapy in other diseases, 245-246 Salvarsan therapy in human syphilis, 245-255 action, 245, 246 combined with mercurial therapy, 245, 246 contraindications, 248 harmful effects, 246, 248 mode of application, 248-255 intramuscular, 253-254 intravenous, 248, 249-253 subcutaneous, 253-254 neuro-recurrences in, 247 value in parasyphilitic diseases, 246 Scarlet fever, 124, 166, 173, 185, 246 Scorpion poison, 96, 100 Scrofulous reaction to tuberculin, 50 Seiden peptone, 155, 228 Selenium, 255-256 Sensitized bacteria, 70, 143 advantages of, 70 Sensitized tuberculin, 70-71 Sepsis, 113, 221, 237 of guinea pigs, 22 of rabbits, 21, 23 Serum, color, 15 bacteriolytic, 231 obtaining of, 12, 13, 79 preservation of, 15, 16, 79 value in therapy, 231-232 Serum diagnosis; see under individual infec- tious diseases and under "Comple- ment fixation" of malignant tumors, 216 antitrypsin test, 122-124, 217 Freund-Kaminer reaction, 217-218 Meiostagmine reaction, 218-220 of meningitis, 175-177 of syphilis, 123, 124. see under "com- plement fixation" "Serum fast," 233 Serum sickness, 83, 223, 226-227 Serum therapy, 231-239 in anthrax, 239 in cholera, 239 in diphtheria, 83-84 in erysipelas, 236, 237 in hay fever, 229-230 in meningitis, 234-236 in pest, 238 in pneumonia, 237 in puerperal fever, 237 in rheumatism, 236 in scarlet fever, 236, 237 in sepsis, 237 in strep tococcic infections, 236-237 in tetanus, 88 in tuberculosis, 162, 233, 238 in typhoid fever, 239 in ulcus serpens, 237 Sessile receptors, 160, 161 Side chain theory, 158-160 "Simultaneous method," 30, 31, 77, 238 definition, 31 Small-pox, 25 Smith's phenomenon, 222-223 Snake poison, 96-100 serum, 100 Specificity, i, 5, 54, 105, no, 125, 134 of Koch subcutaneous reaction, 56 of ophthalmo reaction, 57 of v. Pirquet reaction, 55 of tuberculin reaction, 54 Spermatoxin, 151 Spider poison, 96 Spirillosis of fowls, 246 Spirochete infections, 240 Standard loop, 8 Staphylococci, 31, 44, 92, 199, 200, 205 vaccine, 209-210 Staphylohemotoxin, 92-94 Staphylolysin, 92-94 Sterilisatio Magna, 240, 245 mode of action, 240-241 Strauss canula, 13 Street virus, 27 Streptococci, 32, 173, 206 sera, 236-237 vaccine, 211-238 Subcutaneous tuberculin injections, 45-49, 56 Substance sensibilisatrice, 133; see "ambo- ceptor," "bacteriolysin" Summation of antigen, 156, 157, 174 Swine erysipelas, 30 pest, 36, 40, 41, 42, 43 Syphilis, 123, 124; see "atoxyl, ' salvarsan, "complement fixation." System of hemolysis tests, 147 Tabes dorsalis, 164, 165 Tauruman, 30, 71 Tebesapin, 72 Tetanolysin, 86, 87 Tetanospasmin, 86, 87 Tetanus, 86, 87 antitoxin, 88 cerebral, 86 sine tetano, 86 Thermolabile substances in serum, 15, 16, 133 Thermoresistant substances in serum, 15, 109, i33 Thyroid, 102 Titration of antigen, 178, 179, 185 of bacteriolytic serum, 137 of complement, 150 of hemolysin, 146-149 of luetic sera, 179-182 of virulence of cultures, 34 Toad poison, 96 Toxins, 74-104, 147 action of, 74, 75 definition, 85 dilution of, 18 obtaining of, 74-75 titration of, 76 of higher plants and animals, 95-101 Toxoids, 81, 147 Toxolipoids, 100 Toxon, 80 Toxophore group, 147, 159 INDEX OF SUBJECTS 263 Toxopeptids in relation to anaphylaxis, 228 to tuberculosis, 161 Transfusion of blood, 116-118 Transplantation of tumors, 216 Transudate, 173 Traumatic tuberculin reaction, 50 Trichophytin, 58 Trypan blue, 240 red, 240 Trypanosomiasis, 22, 166, 240-241 trypanocidal agents, 240 Trypsin, 102-104 Tubercle bacilli, 60, 61, 62, 172, 196 emulsion in opsonic determinations, 206 Tuberculin, 45-73, 155-163, 200, 201-203 action, 63-64 as antigen, 162 Beraneck's tuberculin, 63 bovine tuberculin, 71-72 "depot" reaction, 47 diagnosis, 45-49 cutaneous, v. Pirquet, 49-50 explanation of, 156, 157, 160, 161 Hamburger's method, 47 intracutaneous, 51 ointment reaction, 50-51 ophthalmo reaction, 52-54, 57, 58 subcutaneous reaction, 45-49 preparations, 62-63, 70 new tuberculin, 63, 69-71, 211 obtaining of, 45-46 old tuberculin, 45-59, 62, 65 original old tuberculin, 62 vacuum, 62 watery, 62 reaction, 46-54 theories, of Citron, 160-161 of Wassermann, 156, 157 therapy, 60-71, 155-162 antituberculin production during ther- apy, 155-162 contraindications, 68-69 general principles, 64-65 with bovine tuberculin, 71-72 with new tuberculin, 69-71 with old tuberculin, 65-69 with sensitized tuberculin, 70-71 Tuberculosis, 99, 114-115, 233 see under "complement fixation" prognosis in, from local reaction, 58-59 sera for, 99, 114, 115, 238 treatment of, by Friedmann vaccine, 30 in successive steps, 66 with nastin, 72 with tebesapin, 72 with tuberculin; see "tuberculin therapy" vaccination in, 30, 212-213 Tuberculosis of kidney, 67 Tumors, malignant; see under "carcinoma" Typhoid bacteria, 32, 106-113, 123, 134, 135, 171-173,192-193 immunization of rabbit with, 32 titration of virulence of culture, 24 extract, 44 Typhoid fever, 141, 185, 239, 255 protective inoculations, 33-35 serum, 106, 113, 140, 141, 239 vaccine, 33-35, 44, 211 U •Ulcus corneae, 238 Unit of antitoxin, 80 Urine, complement fixation test for tuber- culosis, 163, 174 Urticaria, 227, 229 Vaccination against rabies, 26-30 against small-pox, 25-26 against tuberculosis, 30, 212-213 against typhoid, 33-35 Vaccines (Wright), 209-213 dosage, 211-212 preparation, 209-211 Vacuum desiccator, 15, 16 tuberculin, 62 Venopuncture, 12, 13 Viper's poison, 100 Virulence of bacteria, 24 method for determining, 134 virulence of typhoid, 134 diminishing virulence, 232 increasing virulence, 135,233 Virus fixe, 27 Vital staining, 198 W Wall of leucocytes, 233 Wassermann's reaction; see "complement fixation" Watery tuberculin, 62 Weigert's law, 159 Wet cupping for obtaining blood, 13 Wet nurse, examination for syphilis, 169 Whooping cough bacilli, 173 Widal's test, 106-108 Wright's pipettes and tubes, 203 method for obtaining blood, 204 Zootoxins, 96-100 INDEX OF AUTHORS Allessandrini, 99 Amiradzibi, 218 Anderson, 87 Apolant, 216 Arloing, 114 Arndt, 93 Aronson, 236, 239 Arrhenius, 86 Arthus, 222, 223, 227 Ascoli, 218, 219 Audeoud, 57 Auer, 227 Aufrecht, 68 Bail, 23, 31, 36, 38, 40, 42, 44, 46, 139 Bamberg, 103 Bandelier, 48, 69 Bandi, 238 Bashford, 216 Bassenge, 44 Bauer, 99, 189, 190 Benario, 247 v. Behring, 2, 30, 71, 77, 79, 80, 88, 160, 189, 190 Beraneck, 63 Bergell, 239 Berger, 50 Berghaus, 82, 83 v. Bergmann, 102, 103 Bernstein, 220 Bertheim, 242 Bertrand, 99 Besredka, 223, 224, 239 Biedl, 227, 228 Bier, 14 Bitter, 245 Blaschko, 165 Blumenthal, 166, 242 Boas, 168 Bockenheimer, 88 Boer, 79 Bordet, 3, 31, 86, 120, 132, 143, 144, 152, 153, 154, 171 Borrel, 86 Borelli, 165 Brieger, 44, 102, 139, 187, 217 Bruck, 64, 92, 93, 155, 156, 164, 165 Buchner, 133 Calmette, 52, 88, 99, 100, 238 Castellani, in, 113 Chamberland, 31 Chantemesse, 239 Christian, 160 Citron, 38, 52, 57, 62, 154, 160, 165, 166, 178, 184, 225, 233 Coca, 98 Cohen, 173 Conradi, 90, 109 Courmont, 114 Craig, 1 66 Czaplewski, 8 Dagonet, 216 Denys, 62, 198 Deutsch, 239 Deycke, 72 Diatroptoff, 30 Doenitz, 83, 86 Doerr, 90, 91, 223 Doganoff, 50 Donath, 100, 101 Dopter, 91 Douglas, 198 Doyen, 212 Dreyer, 245 Dunbar, 229 v. Dungen, 98, 100, 116, 144 Durham, 105 Ehrlich, 2, 79, 80, 81, 82, 85, 87, 95, 100, 132 144, 147, 152, 158, 198, 240, 242, 246, 254 v. Eisler, 122 Ellerman, 55 Eppenstein, 52, 57 Epstein, 116, 118 Erlandsen, 55 van Ermenghem, 88 Escherich, 237 Ferran, 29 Ficker, 107 Fleischmann, 165 Fleming, 190 Flexner, 90, 97, 234, 235, 236 Flu, 245 Foix, 173, 236 Fornet, 22, 123 Fournier, 168 Frankel, C., 77 Franz, 56 Freeman, 203 Freund, 217, 218 Friedberger, n, 12, 16, 136, 161, 223, 225 Friedmann, 31 Friedman, 116 Friedemann, 223, 225 Fuld, 103 Garbat, 178, 187, 192, 239 Gay, 227 Gengou, 3, 143, 152, 153, 154 Ghedini, 170 Goetsch, 65 Graff, 218 Gross, 103 Gruber, 105, 233 265 266 INDEX OF AUTHORS Hamburger, 47 Hartung, 227 Hata, 243, 244, 246 Hecht, 189 Helmann, 58 Hendersen-Smith, 85 Herbst, 128 Heubner, 83 Hogyes, 29 Hoke, 44 Holzmann, 99 Hiippe, 44 Issaeff, 136 Iversen, 245, 246 Izar, 218 Jaffe, 142, 143 Jenner, 3, 25 Jensen, 216 Jobling, 234 Jochmann, 102, 234 Johannides, 246 Joseph, 51 Kaminer, 217, 218 Kelning, 58 Kempner, 89 Keysser, 161, 228, 255 Kikuchi, 44 Kitasato, 74, 77, 87 Kitashima, 77 Klausner, 124 Kleine, 115 Klinkert, 162, 163 Koch, R., 30, 45, 47, 48, 56, 60, 64, 71, 114, 242 Kolle, 32, 35, 44, 108, 138, 139 Koplik, 227 Korte, 141, 143 Kossel, 82 Kramer, 68 Kraus, R., 82, 85, 90, 91, 120, 178, 223, 227, 228, 234, 239 Kreibich, 124 Kruse, 90 Kuhn, 35 Kutscher, 114 Kyes, 97 Landsteiner, 100, 101, 144, 186, 188 Laubry, 170 Laveran, 241 Leclef, 198 Ledermann, 165 Leishman, 33, 198 Lenhartz, 67 Lesourd, 154, 192 Leuchs, 173, 192 Levaditi, 165, 197 Lewin, C., 216 Liefmann, 155 Liepmann, 216 Lignieres, 50 Loewenstein, 49, 162 Lohlein, 199 Lorenz, 31 Liidke, 192 Lustig, 238 Mac Fadyen, 239 Madsen, 78, 79, 83, 86, 89 Mallein, 173, 236 Mantoux, 51 Manwaring, 228 Maragliano, 62, 238 Marcus, 102 Marie, 30, 165 Markl, 238 Marmorek, 163, 174, 236, 238 Martin, 77 Marx, 82 Matthews, 213 Mayer, 44 McNeil, 191 Meier, G., 123, 165, 182, 183, 184, 186 Meakins, 191 Mendel, 51 Menzer, 236 Mesnil, 241 Metallnikoff, 72 Metschnikoff, 2, 138, 139, 151, 152, 196, 232 Meyer, 87 Meyer, F., 70, 83, 162, 165, 178, 236, 239 Meyer, K., 102, 103 Michaelis, G., 92, 93 Micheli, 165 Miessner, 128 Mita, 226, 228 Moller, 68 Moreschi, 154, 155 Morgenroth, 10, 13, 16, 83, 144, 147, 152, 158, 165, 255 Moro, 50 Moser, 236 Much, 99 Muller, 102, 186, 188, 191 Munk, 166, 187 Nakayama, 156, 158, 174 Negri, 26 Neisser, M., 92, 141-143, 155, 164, 165, 168, 193 Netter, 83 Neufield, 166, 199, 213, 214 Nichols, 1 66 Nicolle, 153, 246 Nocard, 26 Noguchi, 26, 72, 97, 100, 187, 189 Obermeyer, 31, 129 Oppenheim, 191 Ostertag, 31 Ottenberg, 116, 118 Otto, 89, 223 Pane, 237 Parvu, 170 Pasteur, 3, 26, 27, 41 Petit, 57 Petruschky, 66 Pfeiffer, 35, 44, 108, 131, 134-140, 198, 223, 226, 228 Phisalix, 99 Pick, 31, 129 Pickert, 162 v. Pirquet, 48, 52, 55, 223, 224, 226 Plato, 58 Plaut, 123, 164, 165 Porges, 122, 123, 186 INDEX OF AUTHORS 267 Portiers, 222 Potzl, 1 86, 1 88 Prausnitz, 230 Rabinowitsch, 158 Radziewsky, 136 Ransom, 87 Remlinger, 26, 30 Reschad, 72 Ribbert, 233 Richet, 222 Rietschel, 169 Romer, 51, 82, 95, 237 Ropke, 48, 70 Rosculet, 92 Rosenau, 87, 223 Rosenblatt, 160 Rosenthal, 90, 91 Roux, 26, 31, 74, 77, 82, 86 Ruppel, 70, 162, 234 Russel, 34 Sachs, H., 89, 97, 155, 186, 193 Sahli, 63 Salimbeni, 238 Salmon, 31 Salomonsen, 78, 79 Salus, 44 Schaudinn, 242 Schenck, 57 Schick, 223, 224, 227 Schittenhelm, 223, 226 Schleissner, 173, 236 Schone, 192 Schucht, 164, 165 Schulze, 92, 93 Schiitze, 125, 164, 165 Schwartz, 191 Schwarz, 82 Sclavo, 239 Seiffert, 57 Seligmann, 165 Shiga, 44, 98 Simons, 220 Smith, 31 Smith, Theob., 222, 227 Sobernheim, 239 Southardt, 227 Spengler, 62 Steinberg, 143 Steinhardt, 224 Stenitzer, 239 Stern, Marg., 165, 190 Stern, 141, 148 Stertz, 165 Strauss, 13 Szaboky, 99 Takaki, 86 Tallquist, 100 Tavel, 238, 239 Teague, 191 Teichmann, 59 Terni, 238 Todd, 90, 91 Topfer, 142, 143 Torrey, 191 Toussaint, 31 Traube, 218 Trebing, 102 Tschernogubow, 189 Tschistowitsch, 120 Uhlenhuth, 9, n, 125-130, 242 Vaillard, 84 Vannod, 191 Vigano, 218 Wassermann, A., 2, 3, n, 38, 64, 86, 89, 114, 125, 155, 156, 161, 164, 165, 178, 232, 255, 256 Wassermann, M., 228, 255 Wechsberg, 92, 141-143 Weichardt, 218, 223, 225, 229 Weidanz, 189 Weigert, 159 Weil, 36, 41, 156, 158, 174 Weinberg, 170, 190 Werner, 246 Widal, 1 06, 154, 192 Wolff-Eisner, 52, 54, 59, 223, 224, 225, 229, 231 Wollstein, 191 Wright, 3, 32, 33, 44, 108, 198, 199, 200, 209, etc. Yersin, 74, 238 Zeuner, 72 Zupnik, 87 YC 88631 360085 t ft UNIVERSITY OF CALIFORNIA LIBRARY $§i£^li: Illtll Illf