o ^ THE OPHTHALMIC AND CUTANEOUS DIAGNOSIS OF TUBERCULOSIS ,THE CUTANEOUS AND CONJUNCTIVAL TUBERCULIN REACTIONS ACCORDING TO v. PIRQUET AND WOLFF-EISNER) TOGETHER WITH A Discussion of the Clinical Methods for the Early- Diagnosis of Pulmonary Tuberculosis BY Dr. ALFRED W0LFF=EI5NER A PREFACE BY PROFESSOR H. SENATOR AND AN INTRODUCTORY NOTE TO THE ENGLISH READER BY C. THEODORE WILLIAMS, M.V.O., M.D., F.R.C.P. Consulting Physician to King Ediuard VII. Sanatorium and the Brompton Hospital ; Vice-President of the International Central Bureau for the Prevention of Consump- tion ; Vice-Cliairmnn of the National Association for the Prevention of Ccmsumption TRANSLATED FROM THE GERMAN BY BERNARD I. ROBERT With 2 Coloured Litho Tables, 11 Curve Tables, 15 Reproductions, AND Numerous Curve Figures in the Text NEW YORK WILLIAM WOOD AND COMPANY MCMVIII RD'TAl'THC 3'JS»t43 3 D A "^ r INTRODUCTORY NOTE. All means which tend to promote the early diagnosis of tubercle in its various forms are very acceptable to the clinician, as there is no doubt that the chances of arrest of tuberculosis depend largely on its early detection. Though great things have been accomplished by the aid of physical signs, of sputum testing, of radiography, and other modes of examination, there is room for improvement in the methods at our disposal, and in the fight against tuberculosis he would be a foolish warrior who knowingly would neglect the opportunity of adding a fresh and efficient weapon to his armoury. The present work by Dr. Wolff-Eisner on Ophthal- mic and Cutaneous Diagnosis is a valuable addition to our knowledge, as it describes, firstly, the cutaneous reaction of Pirquet, with all its technique, and gives the author's experience of its utility ; and secondly, it famishes full details of the con- junctival reaction, which has been generally associated with the name of Calmette, but which, on very strong evidence, is claimed by the author, as a co-discoverer with Calmette. The book is full of original work and is a most conscientious study of the two reactions in hundreds of cases, and what is most valuable, it contains a comparison with the results of other methods. It cannot be proved as yet that the ophthalmic method of diagnosing tubercle is infallible, but the evidence of this and other works testify powerfully to its future value, as a handy and useful instrument for the many labourers in this interestmg field, C. Theodore Williams. PREFACE. Within the last few decades — after its treatment had ceased to be regarded as the step-child of therapeutics, thanks to the work of Brehmer and his successors — medical interest has turned to the treatment of tuberculosis to an altogether unexpected extent. It has become more and more recognized that this disease, so much feared in days gone by, in reality represents a very fruitful field for medical activity ; that partial and even full recovery in a clinical sense is quite possible, the more so if the disease is recognized early and treated at once. Consequently the need of a method for the early recog- nition of tuberculosis has made itself more and more felt, and thus not only have the existing methods of diagnosis of recent years become more elaborate and refined, but new ones have been added, such as the demonstration of the presence of bacilli and Koch's tuberculin test, the Eontgen examina- tion, the cyto-diagnosis, the serum diagnosis, and finally the cutaneous reaction according to v. Pirquet, and the ophthalmic reaction according to Wolff-Eisner (and, later, Calmette). The very extensive literature already existing on the subject shows with what zeal these new methods for dia- gnosing tuberculosis have been taken up. Even the very latest reaction, during the nine months it has been generally known, has provoked such a number of publications in the most varied journals of all countries, that it is well-nigh impossible to acquaint one's self with all of them, to fully test the value of each separate view concerning the theoretical bases under- lying the reaction methods, and to find out which is the best way of applying them practically. Now the book of Wolff-Eisner — who is pre-eminently qualified to expound and elucidate this .subject — comes at the right moment. Having had a most excellent schooling in bacteriology, and being highly versed in practical medicine through years of work at the Berlin Medical University Clinic, the author masters the earlier as well as the more vi, PREFACE recent methods of investigation, is just as well acquainted with the theoretical bases underlying them as with their practical application, and takes care not to overrate any particular method, including his own. By demonstrating that hyper-sensibility is the "' basis underlying immunity," and also by his investigations into the nature of hay fever, the significance of albumin poisons, of endotoxins, &c., he has shown to what extent bacteriology can help us in deciding important questions of pathology, prophylaxis, and therapy. In this way his book, containing a rich store of investiga- tions concerning the newly disclosed diagnostic regions, be- comes a guide pointing with clear and critical vision to that which is most important and of the greatest essential value in the great mass of material already at our disposal ; it is to be welcomed both by the technical investigator and the practical physician, and it is therefore likely to have a wide circulation. H. Senatok. TABLE OF CONTENTS. Introductory Note Preface Introduction PAGE iv. V. 1 A. — General Part. The Development of the Diagnosis of Tuberculosis. Agglutination. The Tuberculin Diagnosis of Koch. The Need for other Methods 5 The Agglutination Method of Arloing ... ... ... ... ... 9 The Tuberculin Method of Koch 10 The Application of Tuberculin for Diagnostic Purposes ... ... 11 Technique of the Cutaneous Reaction, according to Pirquet ... ... 13 Technique of the Conjunctival Reaction, according to Wolff-Eisner 16 Tubei'culin Test Preparations ... ... ... ... ... ... 21 The Course of the Cutaneous and the Conjunctival Reaction ... 22 Unusual Forms of Reaction ... ... ... ... ... ... 24 The Conjunctival Reaction ... ... ... ... ... ... 26 Contra-indications for the Application of the Cutaneous and Con- junctival Reactions ... ... ... ... ... ... ... 30 The Result of the Reaction in Tuberculosis of the First, Second and Third Stages in Healthy and in Doubtful Cases ... ... ... 34 History- of Some Cases ... ... ... ... ... ... ... 41 Miliary Tuberculosis ... ... ... ... ... ... ... 43 The Behaviour of the Reaction in Healthy Individuals ... ... 46 Post-mortem Results ... ... ... ... ... ... ... 51 The Behaviour of the Reaction in Doubtful Cases ... ... ... 54 The Repetition of the Reaction ... ... ... ... ... ... 62 Nomenclature ... ... ... ... ... ... ... ... 66 Opinions of Authors on the Cutaneous and Conjunctival Reactions ... 67 Different Cutaneous Tuberculin Reactions ... ... ... ... 74 General Reaction in the Local Applications of Tuberculin ... ... 76 B. — Special Part. The Application of the Cutaneous and Conjunctival Reactions in the Diseases of Children ... ... ... ... ... ... ... 77 The Application of the Reactions in Dermatology ... 82 The Application of the Reactions in Ophthalmology ... ... ... 87 The Application of the Reactions in Gyniecologj' ... ... ... 90 The Application of the Reactions in Psychiatry ... ... ... 91 The Application of the Reactions in Veterinary Science and the RcRults obtained in Animals with the Reactions ... ... ... 92 Vlll. TABLE OF CONTENTS C. — Theoretical Part. PAGE The Comparative Value and Significance of the Cutaneous and Con- junctival Reactions ... ... ... ... ... ... ... 97 The Reaction in other Infectious Diseases 101 The Pathological Anatomy of the Ciitaneous and Conjunctival Reactions ... ... ... ... ... ... ... ... 105 The Prognostic Significance of the Reaction 110 The Prognostic Significance of the Cutaneous and Conjunctival Reactions ... ... ... ... ... ... ... ... 112 Tlieoretical Part of the Prognostic Significance of the Reaction ... 118 Further Particulars concerning the Mechanism of Tuberculin Im- munity, and of Immunity to Tubercle Bacilli ... ... ... 126 Observations concerning Tuberculin Therapy and the Treatment of Tuberculosis ... ... ... ... ... ... ... 136 The Historic Development of the Doctrine of Hypei-sensibility, of Hyper-sensibility Diseases, of Allergy, and the Local Reactions 141 The Outlook 148 The Clinical Methods of Early Diagnosis. The Sputum Investigation ... ... ... ... ... ... 155 Thermometry ... ... ... ... ... ... ... ... 163 The Rontgen Method and its Application for the Diagnosis of Tuberculosis ... ... ... ... ... ... ... 165 The Diagnosis of Tuberculosis of the Bronchial Glands 171 The Streak Diagnosis .. ... ... ... ... ... ... 177 Percussion of the Apices, accoi'ding to Kronig ... ... ... 177 Auscultation ... ... ... ... ... ... ... ... 190 The Morphological Examination of Exudations : the so-called Cyto- diagnosis ... ... ... ... ... ... ... ... 190 The Proteolytic Fermentation of Exudations, &c. ... ... ... 195 Inoscop3' ... ... ... ... ... ... ... 197 INTRODUCTION. The fact that the new methods of diagnosis, the cutaneous reaction of Pirquet and my conjunctival reaction, have roused so great an interest, suggested to me that it might be desirable to more precisely define my standpoint on the basis of personal experiences, considering at the same time the numerous works that have appeared on the subject ; and also, incidentally, to supply the practitioner with an introduction to the use of these methods in his practice. But there were also other reasons which led me to do so. In another part of this work the reader will find a number of enthusiastic references, especially to the conjunctival reaction, which medical men, delighted at having found a new and important method of diagnosis, have left almost entirely uncriticised, although opinions have been expressed by writers of authority, and several writings have been published under the auspices of important institutions. No one has ever disputed Pirquet's discovery ; on the contrary its predecessor, the cutaneous puncture reaction of Klingmiiller, has never been referred to in the whole of the literature. We will repeatedly return to the practical and theoretical importance of the cutaneous reaction. Pirquet, before making his method public property, communicated the same to me, together with experimental proofs, and thus on the day of his lecture I was already in a position to confirm his statements, and at the same time to make mention of the conjunctival reaction. In my short communication concerning this method, given as an observation during the discussion at a meeting of the "Berliner medizinische Gesellschaft " on May 1.5 (1907), I hesitated to draw far-reaching and even very evident con- clusions which would have increased the effect the commu- nication had on the audience, warned as I was by the fate of many a predecessor, and especially by remembering the conflict that is still going on to-day concerning the diagnostic value of tuberculin. In order to draw conclusions of a reliable order, I thought, it would be necessary to base them only on long-continued clmical observations of the reactions. Those 1 2 THE OPHTHALMIC AND CUTANEOUS who investigate the matter after me were less reserved, and in the course of days and weeks I read definite opinions concerning the "reaction of Calmette," as my reaction had, in the meantime, been called by a good many writers. All this could not induce me to abandon my cautious reserve, and publish immature results. Medical men and investi- gators who have a regard for thoroughness will not allow my conscientiousness and reservedness to be interpreted to my prejudice, nor will they in any way lessen the share due to me in the matter of this reaction. I am the last person to doubt that a distinguished and eminent investigator like Calmette, who can look back upon works of such extraordinary merit, could find, and has found, quite independently, the same reaction ; but I maintain that I have carried out the experiments and published them at an earlier date than he, and I am therefore entitled to at least equal recognition, which was withheld from me by Widal and all his co-workers in the case of the cyto-diagnosis, although at that time my position was equal, or rather more favourable, than is now that of Calmette with regard to the conjunctival reaction. I therefore find myself obliged to protest against the reaction being frequently designated in Germany and France as " Calmette's test," a protest which before now has been put forward by impartial parties, von Pirquet, Fritz Levy, and Stadelmann.^ First of all I must contest the grounds of this designation. Calmette and his followers^ say, that there is no doubt that I was the first to have suggested the experiment, but that I had not made any experiments myself, and thus to Calmette belongs the honour of having applied it to clinical work. What may be the thoughts of those who write and copy this ? When, on May 15, I made the communication, four weeks before Calmette, I had carried out experiments upon 150 persons. I declared that for the time being I would refrain from drawing even obvious conclusions about the reaction, in order to maintain the necessary reserve, and especially by reason of the fact that there are differences between the cutaneous and conjunctival reactions which it was first necessary to clear up. ' Pirquet, Vers. d. Naturf. u. Aerzte, Dresden, 1907 ; Ges. f. Kinder- heilk., Ber., p. 34 ; Fritz Levy, Ver f. innere Med., Dec. 16, 1907. Stadelmann, ibidem, Jan. 6, 1908. - Calmette, Ac. des Sciences, June 17, 1907 ; Paris and Petit, I.e. DIAGNOSIS OF TUBERCULOSIS 3 But iny short communication shows in a most definite manner that I have suggested this method for chnical purposes, and that I have made it known for chnical use. I clearly and distinctly stated : The method is of diagnostic as loell as of prognostic significance. All my inclinations point in a clinical direction, and I do not make experiments for the mere sake of experimentation, but ever in the hope that they may prove of clinical value. This remark applies to many of my works, and, as in the case of hay fever and other diseases, this reaction is a proof that such an endeavour may sometimes lead to clinical results. Thus, in this w'ork I propose to warn against the over- valuation of my own method, and to submit a critical description of what has been attained already and of what it is possible to attain. I will especially guard against the over-valuation of the conjunctival method at the cost of the excellent cutaneous method of Pirquet. von Pirquet, for obvious reasons, has not experimented very frequently with the conjunctival method, and most of the other investi- gators reject the cutaneous reaction and only avail themselves of the conjunctival method, and therefore, thanks to the friendly co-operation of Professor Stadelmann (for whose help, to which a great deal of the success of this method is due, I here express my deep-felt gratitude), I am practically the only one who has had the opportunity of thoroughly investigating by means of both methods a considerable amount of clinical material. I might have contented myself with merely collecting here the results of this new method. But both these methods, though decidedly excellent, are nothing without having been clinically tested. To indicate this I have also described the most important methods of early diagnosis. From this we may see how many good methods we have at our disposal ; of these many are still far too little in general use. During about eight years my multifarious investigations have impinged upon the region of earl}' diagnosis of tuberculosis from various sides, and now a curtain seems to have lifted which long had limited my out- look. It now appears that between the clinical-bacteriological method of tuberculin diagnosis and the cytological condition, there exists a closer relationship than was and could have been formerly supposed. If a medical man — of the old school — were told that 4 THE OPHTHALMIC AND CUTANEOUS nowadaj's, armed with a small bottle containing a dilution of tuberculin, one would venture to make the diagnosis of tuberculosis before it was justified by clinical investigation, a picture might easily arise in his mind of a time when medical art had become practically useless, and when the medical man need but provide himself with a vaccinating lancet and a number of diagnostic fluids, the number of the latter according to the extent of the practice to be expected. There is some truth in this jest ; there certainly does exist a danger that many in possession of such useful methods would neglect clinical investigation. This little work is intended to obviate this danger, by showing how very necessary seem to be other methods of diagnosis even to the discoverer of the reaction, who would be the first to have a right to over-esthuate the value of his discovery. Some of these other methods have not hitherto been generally applied to clinical work. The criticism which I am directing against my own method will, I hope, induce many investigators to thoroughly test the other processes recommended. Unfortunately, neither in Germany nor any- where else in the world need we be in want of material for testing these methods. The possession of new and undoubtedly good methods- — firmly based on aetiological diagnosis — should stimulate the development to its greatest extent of topographical diagnosis by means of clinical methods, and to allow these methods to become the common property of all medical men, just as is the case to-day with simple percussion and auscultation, which at one time w^ere the property of but a small school. DIAGNOSIS OF TUBERCULOSIS A.— GETs'ERAL PART. THE DEVELOPMENT OF THE DIAGNOSIS OF TUBER- CULOSIS. AGGLUTINATION. THE TUBERCULIN DIAGNOSIS OF KOCH. THE NEED FOR OTHER METHODS. The discovery of the local reactions and their uses for diagnostic purposes would seem to indicate the beginning of a new era in medicine. It opens out for the practitioner new possibilities with regard to his important work : that is to say, to prevent and combat tuberculosis, and also other infectious diseases. Hardly nine months have elapsed since the first communication concerning this process was made public, and already scientists all over the world are endeavouring to perfect the methods and to apply them more and more to new fields. The fact that in a short space of time more than 500 writings have appeared on the subject, together with the significance attached to these methods, constitutes a proof of the immense value and extent of the work accomplished nowadays by medical investigators. It is obvious that after so short a time problems which have opened themselves up through the newly-revealed reactions have not yet been fully solved, and in the place of those that have been solved new problems have sprung up. Yet it would seem that the time has arrived when it is necessar}^ to publish what has already been cleared up, and institute further inquiries in order to enable the increasing number of those whose co-operation in this work is desirable and necessary to afford us their help. The question whether new methods are needed has to be decided in the first instance by results, and these, without any doubt, appear to point in the direction of such a necessity. The diagnosis of tuberculosis in the early stages (the recog- nition of which is of the highest importance, it is true, is more advanced now than it was in the old Viennese school headed by Skoda and Traube. But it has not yet attained to such certainty as would be desirable in the interest of the case. It was often thought that in the course of this time progress had been made, and it is just now the twenty-fifth anniversary of the famous day of that memorable meeting of 6 THE OPHTHALMIC AND CUTANEOUS the Berlin Physiological Society, to which one often hears allusion made, at which Koch revealed the discovery of the tubercle bacillus. The grand solidity of his discoveries and the flawlessness of his arguments will for ever remain a classical example of scientific publication. Many of those present claim to have recognised the significance of the discovery at the time of its communication, and to have felt the approach of a new era of exact natural science in clinical diagnosis. The tubercle bacillus attracted the attention of everybody to such an extent that the majority of the investi- gators were, so to speak, hypnotised by it, so that they forgot all else. The diagnosis of tuberculosis was based only on the discovery of the presence of tubercle bacilli ; where bacilli were not found no tuberculosis was supposed to exist. In this purely bacteriological era, clinical investigation — which does not offer any specific symptoms by means of which tuberculosis could be discovered — may sometimes have been somewhat neglected, and when we bear this period in mind we can readily understand how it was possible that of late clinical scientists, such as Kronig, for instance, have said with some satisfaction that the clinical methods of investigation in the long run have proved to be superior to the bacteriological and again have become recognised. The results obtained by one method may for a time cause the other methods to be neglected, but the difficulties attached to diagnostical synthesis will always result in the reinstatement of the old methods, since it is only when we apply all diagnostic methods that are at our disposal that we can obtain a clinical picture which, like a mosaic, can only be composed of stones that supplement one another, thus approaching the truth. In this simile we find at once the full answer to the question put above as to the necessity of a new method. The recognition of the true value of the presence of the bacillus in cases of tuberculosis is closely connected with the sanatorium movement now making progress in Germany, which for the first time makes it possible to collect a large amount of material by which a judgiuent could be formed as to the success following" attempts to cure tuberculosis. Before this only practitioners, especially those holding positions as " family doctors," had at their disposal material that would be of use in settling this question. But no general use had been made of this material, probably by reason of its being scattered about. DIAGNOSIS OF TUBERCULOSIS / At present there is a current of thought which certainly undervalues the significance of sanatoria, especially as schools for hygienic living for certain sections of the public. But this is not the place to enlarge on this subject. This dis- approval of sanatoria is a result of the undoubted fact that the hopes put into them have not been fulfilled. These hopes were to some extent exaggerated, but one is bound to confess that the general results obtained were not even of such a nature as would be compatible with justifiable expectations. Yet these sanatoria have not been built m vain, if it were only that they taught us that the curableness of tuberculosis was absolutely dependent on an early diagnosis. The earlier the diagnosis can be inade out the better the chances of a cure. Thus, in order to obtain better results, the methods of diagnosis had to be improved. In the beginning it was thought to be undesirable to put individuals — suspected but not yet proved to be suffering from tuberculosis— into sana- toria where they would be in danger of infection, and their admission depended practically on the presence of tubercle bacilli in the sputum. But on the basis of the experiences gathered in those very sanatoria this view has generally been abandoned. It was this basis of admission which was the cause of the unsatisfactory results. It gradually became recognised that tuberculosis bacilli in the sputum did not represent the initial stage of the disease. This is not difficult to understand if we remember the anatomical conditions. Wherever the tubercle bacillus settles, whether in small vessels, or in the submucosa, or in the intra-alveolar tissues, it cannot possibly at first reach the outer world. It gradually forms a small tubercle which, when situated in a small bronchus or bronchiole, may cause an atelectatic collapse of a small alveolar region. Tubercle bacilli can only appear in the sputum when the bacilli or their products of metabolism have caused a caseation, which, when it breaks through the mucosa, admits the tubercle bacilli into the bronchus, and thus into the sputum. Everyone who is acquainted with these conditions can realise the fact that a fairly large number of bacilli must be present before they can be detected by the comparatively clumsy microscopical investigation, especially since in fairly early stages no suspicious lumps are as yet to be found which allow the tubercle bacilli to be detected so easily. The distinction made lately between open and closed 8 THE OPHTHALMIC AND CUTANEOUS tuberculosis simply serves to indicate whether any particular case of tuberculosis is one of mixed infection or not. In a strictly anatomical, clinical, and bacteriological sense tuber- culosis is to be regarded as an open disease from the time onwards that tubercle bacilli appear in the sputum, and from that moment the door is open to mixed infection. Everyone knows the difticulty experienced in diagnosing catarrh of the apex on a tuberculous basis, without being assisted by the presence of the tubercle bacilli- It is therefore to be deplored that as regards this impor- tant early diagnosis the tubercle bacilli, which represented the greatest advance in the region of tubercular diagnosis, are of little use. All the other methods dealt with in this work are destined in the first instance to supply a substitute for this diagnostic loss. For cases of advanced pulmonary tuberculosis there is no need for more refined methods, since those that are known and in general use are perfectly satisfactory ; the results of a physical investigation, as a rule, are so unambiguous that doubt can only be entertained in very rare cases. Also in such cases tubercle bacilli are to be found in the sputum with such absolute regularity that there is no necessity to look for a more refined method. We emphasise this fact here because we shall see later on that the most refined new methods fail to operate in advanced cases. On the other hand, the less frequent forms of tuberculosis not localised in the lungs — tubercular meningitis, pleuritis tuberculosa, peritoneal tuberculosis, and especially general tubercular infection (miliary tuberculosis) — sometimes present the very greatest diagnostical difficulties. It is only when in possession of a great deal of material that one can realise how often, in well-conducted hospitals, undiagnosed miliary tuberculosis is found to exist post mortem, how frequently miliary tuberculosis is wrongly suspected, and how many a time the diagnosis — and this is hardly conceivable by a be- ginning practitioner — wavers between typhoid, sepsis, and miliary tuberculosis. Here we are already a long time in possession of another and very real form of ophthalmic diagnosis, since one only ventures to make a definite diagnosis when tubercles are found in the fundus oculi. Also the local cutaneous tuberculosis, the lightest form of tubercular infection, sometimes presents diagnostical difficulties by reason of its similarity to a number of other DIAGNOSIS OF TUBERCULOSIS y cutaneous affections. The [etiological connections between glandular tuberculosis, scrofula, and pseudo-leucocythsemia are even in the present day still subjects of scientific discussion. For all these cases clinical science has long since felt in need of a diagnostical method which would assist the medical man in his clinical diagnosis. Since the discovery of the tubercle bacillus by Koch, only two methods have become known which were of assistance in the diagnosis of tuber- culosis — the method of the agglutination of the homogeneous tubercle bacillus culture of Arloing, and the subcutaneous injection of tuberculin for diagnostical purposes by Robert Koch, Although Koch's method is the earlier one, we will first deal with the agglutination diagnosis of tuberculosis, since the tuberculin method of Koch possesses points of close contact with the new methods, and therefore had best be dealt with together with the latter. THE AGGLUTINATION METHOD OF ARLOING. The agglutination method of Arloing is based chiefly oa the Gruber-Widal reaction of the agglutination of typhoid bacilli by means of immune serum, especially by means of serum of typhoid patients. The extraordinarily favourable diagnostical results obtained in cases of typhoid infection by means of the agglutination experiment, practically in- vited a transmission of this method, especially since it was found that very good diagnostical results were obtained with it in the cases of paratyphoid, coli, and similar bacteria. But already in typhoid certain difficulties are encountered in the diagnostic use of tbis method. No former attack of typhoid may have occurred — which it is not always possible to ascertain — in order to allow of diagnostical conclusions. Also in diseases that go together with icteric symptoms we often find important increases of the index, group agglutin- ation occurs, and also the serum of a normal person agglutin- ates, although not to a very important extent, but in such a manner that the difference is only a quantitative one. No connection can be found between the degree of agglutination and the kind and course of the disease ; some people with a very low serum index may overcome the infection, whilst others with a very high serum index may not be able to do so, and may die. To this there is added the difficulty that agglutination is 10 THE OPHTHALMIC AND CUTANEOUS a process which, without forcing the matter, can only be observed in movable bacteria, because agglutination consists in the conglomeration of the flagella, and in this way little heaps are formed. But this difficulty was done away with by the discovery that agglutmation can also be observed in the dead, and therefore flagellaless, bacteria — the agglutination of dead typhoid bacilli, according to Bordet and Ficker. This fact, which at first sight seems rather surprising, becomes intelligible by understanding that besides the agglutinin proper we always find precipitin, and that the clarification of the fluid containing the dead bacteria in reality constitutes a process of precipitation in the course of which the dead bacteria are drawn to the bottom in clots. (It must, however, be mentioned that lately the method of Ficker has been violently contradicted. Schrumpf, Miinch. med. Wochenschr., 1907, No. 51.) It was specially difficult to make sure of the occurrence of agglutination in the case of tubercle bacilli. The tubercle bacilli, by their slow growth, have the peculiarity that they order themselves spontaneously into groups, and cultures of this kind cannot, of course, be used in deciding whether an agglutination process has taken place or not. This only became possible after success in obtaining, by means of a fairly elaborate technique, cultures of agglutinable tubercle bacilli, spoken of as the homogeneous culture of tubercle bacilli. By means of continuously shaking the preparation during its growth it becomes possible to obtain, in fluidic culture media, a growth in which the bacteria lie singly and thus are capable of being agglutinated — that is to say, precipitated. This method has often been applied. The results, though not bad, were not sufficiently reliable for practical purposes. Also the course of procedure was too complicated, and there was no organic connection between the agglutination and the state of the disease. At present this method is compara- tively little used. (See Arloing and Courmont, Ber. des Kongresses z. Bek. der Tuberkulose, Berlin, 1899, p. 229. Beck and Kabinowitsch, Deutsch. med. Wochenschr., 1900, No. 25, p. 400 ; 1901, No. 10, p. 145.) THE TUBERCULIN METHOD OF KOCH. Koch's tuberculin reaction, on the other hand, has found great popularity, and to-day, despite all opposition, it is made DIAGNOSIS OF TUBERCULOSIS 11 use of to quite an enormous extent. The discovery by Koch of the specific tubercle bacillus toxin inaugurated a new era. The excitement in the medical as well as the lay world was even greater than that created at the period following the discovery of the tubercle bacillus. The cause of this feverish excitement lay in the fact that at the time it was hoped (not b} Koch, though) that there was a possibility of completely curing tuberculosis. The question concerning the significance of tuberculin as a cure has not yet been settled, but we do know that the hopes entertained at the time have not been fulfilled. Nor could the theoretical basis be maintained on which originally the tuberculin treatment was founded. But it will be seen from the transactions of the fourth conference of investigators of tuberculosis in May, 1907, how widespread, even to-day, is the therapeutic appreciation of tuberculin. It was even intended to form a society of adherents of the tuber- culin treatment, with the purpose of exchanging experiences gained by the application of this so frequently opposed remedy. This project — which, fortunately, has not been carried out, since a fruitful discussion could hardly be expected — at all events serves to show how difficult it must be to judge the therapeutic results of tuberculin, since after so many years of arduous investigation its adherents still find it necessary to discuss and exchange views with regard to the remedy. Contrary to what has been said concerning its therapeutic value, its diagnostic importance is not to be doubted, and it is hardly an exaggeration to maintain that this discovery stands at the beginning of the diagnostic science of tuberculosis, and our clinical knowledge concerning the distribution of this disease. It seems to have been the mission of tuberculin to completel}^ fill up the gap caused by the loss of the value of the tubercle bacillus for purposes of early diagnosis, and to constitute an extremely fine biological method by which to discover even the very first stages of tuberculosis. THE APPLICATION OF TUBERCULIN FOR DIAGNOSTIC PURPOSES. Whilst at first far too large doses of tuberculin had been applied for diagnostic purposes, fairly unanimous points of view have now been arrived at. Beginning with ^ to 1 milli- gramme, one may increase the quantity to 3, 5, or even 8 milli- grammes, when no reaction takes place, repeating the same dose when a doubtful reaction occurs (Koch, Petraschky, Beck, 12 THE OPHTHALMIC AND CUTANEOUS Max Wolff, MoUer). Against this method certain objections may be raised, to which we shall refer later. The French clinicians do not generally use larger doses than ^ milligramme (Hutinel, Grasset and Vedel, Claisse, Souques, Cawadias, Milian and Sicard, Labbe). "We should also mention the method of Lowenstein (and Ostrowsky), who repeats the same small dose even when the result of the reaction is a negative one. In this case diagnostic advantage is taken of the hyper-sensibility which exists in the patient, whilst a healthy person does not give a reaction after the second or third, but only after the seventh or eighth repetition of a small dose. But despite the indisputable value of this method, the subcutaneous injection of tuberculin has not come up to all expectations, and therefore further methods for the early diagnosis of tuberculosis were sought. The reason for this disappointment lies in a series of important shortcomings attached to this method, which prevent many clinicians and practitioners from making use of this reaction in human beings. In the first instance, certain dangers are attached to the subcutaneous tuberculin reaction. It is unnecessary to quote the extensive literature on the subject ; their existence is undoubted. But the subcutaneous injection of tuberculin also produces symptoms of disease which, it is true, have no lasting effect besides causing a rise in the temperature, which may frequently be very considerable ; lack of vitality ; pains in the limbs; various symptoms at the seat of the disease; and severe headaches, which may keep the patient in bed several days. All this renders the clinical tuberculin diagnosis impossible in practice and prevents its general use ; quite the opposite is the case with the new local reactions. The reason that the subcutaneous tuberculin injection has found so little favour in medical practice, lies in the fact that the diagnosis depends on a rise in the temperature. That is to say, that in the case of people whom one wishes to inject subcutaneously with tuberculin for the purpose of diagnosis, one has to observe the temperature very carefully for several days. For this purpose it would be necessary to take the temperature of the patient every two hours after the injection. The taking of the tempera- ture is too inconvenient, and this, together with the dangers referred to, is the principal reason why the subcutaneous injection of tuberculin has found so httle favour in practice. DIAGNOSIS OF TUBERCULOSIS I'S All these disadvantages are done away with bij the applica- tion of the local reactions. They do not expose the patient to the danger of being injured or having his sufferings increased ; they can be used in clinical practice ; they do not require a tedious observation of the course of the reaction ; and, most important of all, they produce reactions just as specific as the subcutaneous tuberculin reaction. The results put forward in this book justify the hope that the subcutaneous tuber- culin injection can, for diagnostic purposes, be replaced by the absolutely harmless local reactions. In the following chapter, w^e shall at once proceed to deal with the local reactions ; first with the technique, then with the practical clinical observations, and finally with the theo- retical results. TECHNIQUE OF THE CUTANEOUS REACTION, ACCORDING TO PIRQUET. Instrunients. — The cutaneous inoculation, according to Pirquet, may be performed by means of a vaccinating quill, vaccinating lancet, or Pirquet's raspatory. For the investiga- tion of large numbers the use of a platinum-iridium instru- ment is recommended. This inoculation is performed in exactly the same way as ordinary vaccination ; before inocu- lation the skin is superficially disinfected with alcohol or alcoholic ether. The danger of infection during and after inoculation is very slight, since it is only very rare that pustules occur, as is the case in vaccination ; therefore a secondary infection is excluded. In order to successfully perform the inoculation it is not necessary to open a minute blood-vessel for the purpose of producing a drop of blood. The opening of the most super- ficial lymph passages suffices ; this is clearly proved by the fact that Moro and Doganoff (by the cutaneous tuberculin test with 50 7o tuberculin salve) and shortly after Lignieres and Berger obtained the reaction by merely rubbing dead tubercle bacilli or undiluted (or w4iat acts better still, concentrated) tuberculin into the uninjured skin {Academic des Sciences, Oct. 28, 1907, Paris, and Wiener klin. Wochenschr., 1907). One can easily understand that the most superficial scarifications are sufficient, and even preferable, because a blood crust, especially during the first twenty-four hours after vaccination, may ob- scure the specific reaction. This slight operation, which is lim- ited to the most superficial epithelial layers, is most easily 14 THE OPHTHALMIC AND CUTANEOUS performed by means of the perforating raspatory of Pirquet. It is necessary to slightly stretch with the other hand the sm-face of the skin where the operation is to be performed ; this prevents the otherwise relaxed skin from being torn during the boring, or rather punching, process, which would cause unnecessary pain. Furthermore, the rasping by means of Pirquet's instrument has the advantage that the reaction following on the inoculation process takes a circular shape, Fig. 1. Vaccinating perforator, according to v. Pirquet. whereas the reaction after an ordinary inoculating incision would naturally take a long oval form. Taking the perforator, which is made heavier at the handle than at the blade, one presses the point on the skin and performs the perforation by rolling the handle between the fingers. It is not to be recommended to merely moisten the inocu- lating apparatus with tuberculin, as in ordinary vaccination, where the quill is merely moistened with the vaccinating DIAGNOSIS OF TUBERCULOSIS 15 lymph ; it is better in this case to place a drop on the skin and to make the scratch or move the perforator in this drop. This course of procedure is all the more to be recommended since the inoculation with tuberculin differs from ordinary vaccination by the fact that in the former there is no danger of implantuig virus that is capable of multiplying. Most frequently a 25 per cent, solution of Koch's old tuberculin is employed ; but I refer to another section of this book, where it is pointed out that we must also consider the possibility Fig. 2. Pipette bottles, according to v. Pirquet. of experimentmg with different concentrations and other tuber- culin preparations. Since there is danger, although remote, of transmitting syphilis from one human being to another, it is necessary to make the platinum-iridium vaccinating lancet glow in a flame before every new inoculation. Control Inoculations. — As Pirquet has mentioned already, it is to be recommended to employ a control inoculation. The sensitiveness of the skin with regard to traumatic irrita- tions is so great in some individuals that the slightest irrita- tion causes a reaction. By means of a control perforation with 16 THE OPHTHALMIC AND CUTANEOUS a drop of a sterile physiological solution of common salt, or with a 5 per cent, solution of glycerine and a 0"1 per cent, one of carbolic acid (corresponding to the concentrated solu- tion employed), it is possible to obtain an absolutely correct picture by comparing the wound reactions — which, if they appear at all, are very minute — with the specific reaction. In the case of inoculation in large numbers, I should advise the use of two inoculating perforators, one for control inoculations and the other for tuberculin inoculations, be- cause on account of its great heat-resisting property some tuberculin, still active, might be brought into the wound by means of the perforator, and thus produce a reaction, especially if one considers the minute quantity of tuberculin which is necessary to produce a reaction. This is not a mere theoretical point. We have seen in thirty inoculations, performed in succession, that in all cases in which a reaction occurred at the spot of inoculation also a slight reaction was to be observed in the control spots. The explanation of this fact was that only one lancet was used for both the tuberculin and the control inoculation, the instrument having been heated in a spirit flame between the two operations. The spirit apparatus was filled with ordinary spirit which was not of particularly high percentage, the flame was sooty and did not cause the lancet to glow in a quarter of a minute. This low temperature of the flame was the reason why the heat had not destroyed the tuberculin that had remained on the lancet. TECHNIQUE OF THE CONJUNCTIVAL REACTION, ACCORDING TO WOLFF-EISNER. The technique of the conjunctival tuberculin diagnosis is, if possible, even more simple than that of the cutaneous inoculation. One proceeds according to the method used by all medical men for treating conjunctivitis : the introduction of a drop of fluid into the conjunctiva, or the instillation of atropine for diagnostic and therapeutic purposes. The lower eyelid is drawn down and the drop is instilled. But since it is necessary that the conjunctiva, if only for a short time, be bathed to some extent in the fluid instilled, some further measures are needed, which altogether take less time than their enumeration here. In some individuals the cavity obtained when the lower eyelid is held down is extremely DIAGNOSIS OF TUBERCULOSIS 17 flat, so that the drop easily flows out again. It is advisable that the head should be held somewhat back, in order to counteract the tendency of the drop to run out. Then the lower lid is held down for about half a minute in order to ascertain the bathing of the mucous membranes in the solu- tion, and then the patient should be enjoined to remain, his head bent back, for another half minute without closing his eye, in order to prevent the drop from being pressed out of the eye by the contraction of the lids. Protective Dressing. — In cases in which special value is attached to the result of an experiment, it may be advisable to prevent the eye being rubbed by the fingers, and causing a traumatic inflamed irritation or secondary infection. This may be done by means of a protective dressing of the eye, the watch-glass dressing of the ophthalmologists or a trans- parent celluloid protective cap — such as is often recommended in vaccination — not a bandage (monoculus) which itself may cause irritation. In our numerous investigations such pre- cautions have not been necessary ; the inclination to rub and scratch the eye was but extremely slight. Hoiv to undertake the Beaction in Existing Conjunctivitis. — The self-evident question how to apply the reaction in a case of existing conjunctivitis has to be answered. In our records (see tables) we have carefully noted these cases, in order to exercise special care in making use of the results. If both eyes were equally inflamed no difficulties were present, since the eye not operated upon afforded a comparison, and thus allowed us to give the reaction its proper value. Fears expressed by Calmette, for instance, concerning the use of the reaction in the case of conjunctivitis do not seem to have been justified. Since conjunctivitis is of extremely frequent occurrence, the use of the conjunctival reaction would have been very much impaired if these theoretical objections of Calmette had been justified, and if, in these cases, any injury had been done or too strong a reaction had been observed. As a matter of fact, the instillation of tuberculin had not the least effect on the conjunctivitis, and in all cases where a reaction occurred it remained absolutely within the ordinary limits. As a rule, there was no difficulty in recognising the reaction by a comparison with the other eye. But greater difficulties are encountered where the con- junctivitis is only one-sided. In our investigations we have fairly frequently met with one-sided conjunctivitis, and 9 18 THE OPHTHALMIC AND CUTANEOUS we have found that it is more frequent than one would generally be inchned to suppose. Since, clinically speaking, the specific reaction only manifests itself as a conjunctivitis in those cases in which there is one-sided conjunctivitis, we have no eye to compare with. With regard to such cases, we have abandoned our principle to only use (in order to avoid mistakes) the left eye of the patient, and have instilled the tuberculin into the healthy eye. Now, if in this healthy eye a reaction occurs which is equal to, or exceeds, that of the inflammation, we may without hesitation regard the reaction as being positive ; if the reaction remains weaker the decision is more difticult. Yet it is possible, especially if one pays a great deal of attention to such cases, to decide by the symptoms of the inoculated eye alone, without the control eye, as to the positive or negative result of the reaction. The above-described measures to avoid the drop of tuber- culin solution from being pressed out are of very great practical significance for ascertaining the result of the reaction. This has also been pointed out by Desplats {Sociefe des Sciences Meclicales de Lille, July 10, 1907), Petit (Ophthal- moreaktion, 1907, Paris), Masson (p. 21), and Stadelmann (in his lecture to the Verein fiir innere Medizin, on January (3, 1908). The practitioner who has only occasion to apply this reaction in a few cases will soon have his attention directed to this point. The measures of precaution have to be very carefully observed in cases where a great many individuals have to be examined, and the few cases in which signs were found of disagreement between the tuberculin, cutaneous or subcuta- neous reactions on the one side, and the conjunctival reaction on the other, were in all cases easily cleared up when, under the observation of these precautionary measures, a direct and distinct reaction was obtained by the instillation of a drop of tuberculin solution into the other eye. We shall see further on that a repetition of the experiment in the same eye has to be absolutely avoided ; therefore in the same individual the experiment can only be repeated once, and that in the other eye. From this it is evident that from the first the measures of precaution have to be regarded ; these measures, fully applied, yet allow the reaction to be performed in the shoit space of time of about a minute. (Compare, however, with the chapter dealing with our experiences in cases in which the instillation was repeated.) DIAGNOSIS OF TUBERCULOSIS 19 Concentration of the Solution. — We make use of a solution of Koch's old tuberculin in a 0"8 -gev cent, sterile physio- logical salt solution, mixed together simply by means of sterile pipettes and brought into a sterile vessel. It is to be recom- mended to bring the amount to be used in one day into a watch glass and from there on to the conjunctiva by means of a pipette. As the original solution will have to be used for several days, in applying the conjunctival reaction, there is a possi- bility of conveying septic infection from the solution to the conjunctiva by means of the pipette, the contact of which with the conjunctiva cannot always be avoided. According to our experience the development of bacteria in a weak tuberculin solution renders it inactive and gives rise to an important source of errors, to which w'e must call attention (Stadelmann). We therefore recommend that the solution for the conjunctival reaction should be prepared afresh at least once a week. To make it last longer, Eppenstein, instead of using a physiological solution of common salt, sub- stitutes for it a 3 per cent, solution of boric acid. Opinions still differ as to the correct strength of the solu- tion. At first I made use of a 10 per cent, solution, whilst later on, at the request of Stadelmann, I carried out experi- ments with a 1 per cent, solution. Fritz Levy works with 2 per cent, and 4 per cent., Eppenstein with ^ to 4 per cent, solutions. The difficulty in deciding lies in the fact that repetitions of the reaction with increasing strengths are not feasible for reasons to be dealt with later on, and it is therefore impera- tive to decide on a degree of concentration which, without causing harm, may allow of an exact diagnosis at tlie very first application. Since I did not see any evil effects,^ even with a 10 per cent, solution, it seemed to me that there were no objections to the use of strong solutions. One must decide on a degree of concentration (by means of prolonged efforts) ' A case of which I was informed hy Kronig, of suppuration of the Meibomian gland after the appHcation of a 1 per cent, tuberculin solution, I attribute to secondary infection or bacterial impurity of the tuberculin made use of ; the possibility that injury might be done with a 10 per cent, tuberculin solution is not, of course, excluded. A strong solution, there- fore, is only to be applied after certain precautionary measures have been taken. (See the chapter deahng with contra-indications and with its application in ophthalmology). 20 THE OPHTHALMIC AND CUTANEOUS which will give a distinct reaction in affected cases without causing a rash in healthy people (donhtful cases) on account of the strength of the tuberculin solution used. As mentioned above, we have carried out our experiments with a 1 per cent, solution, and were completely satisfied with the results obtained : they were neither too weak nor too strong. Such a solution prepared from the ordinary old tuberculin of Koch, with 20 per cent, glycerine added, con- tains at most 0"2 per cent, glycerine and 0"005 per cent, carbolic acid, an amount which certainly causes no symptoms of irritation, according to our observations, just as little, in fact, as the " fausses reactions" of Calmette. According to Eppenstein (Med. KL, 1907, No. 36) and Levy, even a 2 per cent, glycerine solution does not cause any symptoms of irritation. Fron:i an ophthalmological point of view the occurrence of a reaction after the application of a 2 per cent, glycerine solution has occasionally been observed, but these cases are so rare, and then only limited to ophthalmic patients, that the ordinary medical man does not meet with them. We find that one of the chief advantages of the con- junctival method in the form we recommend lies in its cheap- ness and simplicity. All medical men, even in the most distant parts, are in a position to apply this method without any assistance. Its cost varies from x^o to ^V P^^'*' of a penny per reaction, according to whether one prepares the solution oneself or has it prepared. This point brings us to the technique of Calmette. In order to avoid the irritating action of glycerine on the mucous membrane, he recommends the use of tuberculin precipitated by alcohol. He takes a six weeks old culture of tubercle bacilli obtained from the cow, heats it or destroys it in the autoclave at a temperature of 110" during twenty minutes, then evaporates down to one-tenth at a temperature of 80° to 90° and then filters ; finally he precipitates by means of 95 per cent, alcohol. By the addition of a small piece of sodium chloride the precipitation is hastened considerably. He then filters through a filter paper, takes the precipitate and dries it in a vacuum. These processes are repeated twice for the purpose of further purification ; finally a whitish powder is obtained which is free from glycerine, and is supposed not to contain any resinous or waxy substances, but which yet contains the active bacteria (see the theoretical part). DIAGNOSIS OF TUBERCULOSIS 21 TUBERCULIN TEST PREPARATIONS. The " Institut Pasteur " of Lille produces the preparation described above : (1) in powder form in doses of 5 milli- grammes each, to be used for the preparation of a 1 per cent, solution by adding ten drops of sterile water ; (2) a sterilized 10 per cent, solution in sealed glass tubes. The " Hochster Farbwerke " have lately produced a similar preparation which, like Calmette, they have designated " test tubercuhn." We strongly object to the use of the latter preparation. A number of strong reactions that have lately become known, and which might discredit the whole method, are to be regarded as due to its action. According to Levy the "Hochster" test tuberculin is too strong, and according to Wassermann (oral communication) this is due to the tuber- culin being saturated with alcohol and the solution being calculated not on the basis of the original volume, but on that of the weight of the dry substance. 100 c.c. old tuberculin yield 10 grammes dry substance ; if this is employed for the preparation of a 1'/^ solution, this solution in reality rep- resents a 10% solution according to the standard provided by the original fluid tuberculin. It is therefore not surprising if strong reactions are obtained by the use of this preparation. The so-called 1 per cent, solution has to be diluted about ten times. Nothmg can be said against these tuberculin preparations, except that they have no advantage over Koch's old tuber- culin, since no harm has ever been observed by us to have been caused by glycerine, if one uses solutions of not more than 0*2 per cent, concentration. It is surprising that Cal- mette, without stating his reasons, should have made use of bovine tubercle bacilli. Although the dissimilarity between the tubercle bacilli of the cow and those of human origin has not yet been absolutely proved, according to Koch's experiments the bacilli may be quite separate, and this hypothesis must be taken into account. It is therefore advisable to make use of solutions prepared from tubercle bacilli of human origin, unless one intends to institute investigations concerning the frequency of diseases which bovine tubercle bacilli may produce in man. The only objection w^e have to these other preparations is that they increase the cost of the reaction to a very consider- able extent ; whilst it is just the cheapness of the conjunctival 22 THE OPHTHALMIC AND CUTANEOUS reaction and the simplicity of its application which constitute one of its greatest advantages. The cost of a reaction by means of this " test tuberculin " is from 85 d. to Is., against tIo to £y part of a penny when old tuberculin is used. Nor can Calmette's idea concerning the purification of tuberculin make any claim to originality. Koch himself had already prepared a tuberculinum depuratum by adding to the tuberculin one and a half times its volume of absolute alcohol, and by repeatedly washing the precipitate that had formed after twenty-four hours with 60 per cent, alcohol. In a similar way Klebs prepares tuberculinum depuratum, begin- ning also by producing a precipitate by means of alcohol. He precipitates the fresh tuberculin by adding five to ten times its quantity of absolute alcohol, and then proceeds with further methods of purification (quoted after Pick's represen- tation of antigene in Krans-Levaditi's " Hdbch. der Technik und Methodik der Immun-Forsch.," Jena, 1907, p. 364). Maragliano has also prepared an aqueous tuberculin free from glycerine. The only difference which exists between these prepara- tions and that of Calmette consists of the fact that in the former the glycerine is removed by being precipitated by alcohol, whilst Calmette does not allow any glycerine to act on the tubercle bacilli. However, glycerine is the extraction substance Kar e^o'x^rjv, which Koch, after very elaborate experi- ments, found to be the best adapted for the extraction of tubercle bacilli, and most of the other investigators have con- formed to this view. In attaching but little importance to the purification of tuberculin we are of the same opinion as Koch, who thinks that the purification of raw tuberculin is not of very great moment. THE COURSE OF THE CUTANEOUS AND THE CONJUNCTIVAL REACTION. (1) The Cutaneous Beaction. — In consequence of the inocu- lation after about an hour and a half a very slight redness appears at the spot of inoculation and also at the spot of the control inoculation. This is to be regarded as due to the slight traumatic irritation caused by the actual performance of the inoculation. As a rule this redness disappears after an hour or two, and in an inoculation leading to positive results a new redness appears after three hours, which does DIAGNOSIS OF TUBERCULOSIS 23 not appear on the control spot. The redness then increases till it reaches a maximum, which in different cases may occur after a variable number of hours, very frequently after between twelve and twenty-four hours. The time from which the reaction may be said to take its course varies. In judginc( the degree of the reaction a great deal is left to the judgment of the operator. In order to distinguish this degree we should advise the consideration of distinct reactions only for the purpose of diagnosis. These distinct reactions we divide into three degrees : (1) Distinct reaction ; (2) Strong reaction ; (3) Very .strong reaction. Those who apply the cutaneous reaction often will not be in doubt as to what represents a very strong reaction ; little difdculty is encountered in distinguishing between the distinct and strong reactions. The reaction can occur in two different ways. Twenty-four hours after having reached its maximum, that is to say, altogether after forty-eight hours, it may, under certain conditions, decrease to such an extent that hardly a sign of it is left ; on the other hand, it may increase in the course of the next few days. We attach great importance to these conditions, and we have therefore in our tables recorded in each case the course of the reaction during the first four days. The type of reaction, which appears quickly and soon disappears, is that designated by the French as reaction precoce et precise, and corresponds to the reaction which is obtained in experiments of hyper-sensibility. Here, too, it is really a condition of hyper-sensibility, which is dis- tinguished either by a reaction which quickly appears and quickly disappears, or by one of excessive intensity. At about the spot of inoculation the following effects are observed : The first, as a rule, is an injection conditioned by local hypersemia ; then follows exudation, which causes the spot on the surface to appear somewhat in relief. This can be detected better by touch than by sight. From the very beginning the exudation often is so tense and so well defined that one may be in doubt as to whether this is not an infiltra- tion, and as a matter of fact very frequently the original pomphus changes into a papule. Since we make so important a distinction between exudation and infiltration in the employ- ment of tuberculin it is to be regretted that we cannot decide from the beginning w^iether we have to deal with the one or the other. 24 THE OPHTHALMIC AND CUTANEOUS It) is remarkable that in many cases, and also in any particular case, the proportion of injection, infiltration, and exudation varies, and, what is more remarkable still, even during the course of the reaction any of these may show fluctuations, which may easily be overlooked in clinical observation, but which one readily appreciates when one endeavours to represent them by curves. For instance, the redness may surround the region of infiltration in the form of a wide halo, and then it may suddenly disappear for a few hours, appearing again in the form of a halo round the papule in the centre. UNUSUAL FORMS OF REACTION. We must particularly mention some unusual forms of reaction which we have noticed in a very few cases. In one case in which a cutaneous reaction was extraordinarily strong, the tense papule in the centre gradually softened and a vesicle was formed filled with a dull grey substance, which could hardly be distinguished from a vaccine pustule. In another case in which no ver}' strong reaction took place, the centre^ — which did not exhibit any infiltration, — vesiculated, showing a moderately injected border at the periphery, which shaped itself into concentric rings, and these gradually paled. At the same time the continually increasing centre exhibited desqua- mation, so that the process vividly reminded one of a case of herpes tonsurans (a herpetiform kind of reaction). In one case herpes-like vesicles appeared round the pustules ; these dried up without bursting, il am convinced that a great many forms of reaction will still be discovered and described, especially when dermatologists, who have had a great deal of practice in the cognition of such processes, will take a greater interest in these forms of reaction. As a matter of fact Bandler and Kreibich have described different forms of cutaneous tuberculosis as being the results of the reaction (see the chapter on the " Application of the Reaction in Dermatology" and in "Diseases of Children"). It was not mere speculation which caused us to determine these forms of reaction. Apart from the fact that probably a distinct clinical importance has to be assigned to each form, the different reactions that follow an identical inoculation of tuber- culin also explain polymorphism of the tuberculous affections of the skin. DIAGNOSIS OF TUBERCULOSIS 25 Forms of Late Reactions. — Special interest attaches to a form of reaction which does not exhibit a subsidence of the symptoms after four daj's, but on the contrary shows a dis- tinct increase; so that, for example, during the first two days there is a distinct, during the third day a strong, and during the fourth day a very strong reaction. Still more pronounced, and, at any rate, easier to diagnose, is another form of late reaction in which only on the third or fourth day does a reaction take place, von Pirquet believes such a reaction to be a torpid one, and does not exactly know what sig- nificance to attach to it. We (Stadelmann and myself) believe it to be of great importance and designate it as late reaction. Very often a papule of long duration is formed in the case of late reaction. This may also occur in other forms of reaction, although it would be rarer. Such a late reaction may be perfectly unaltered even after three or four weeks. The firmly infiltered papule of a bluish-red appearance reminds one of the early stages of an anatomical tubercle. Lymphangitis during Cutaneous Beaction. — There are but very few other clinical symptoms which have to be men- tioned. In only one case have we observed the occurrence of lymphangitis with swelling of the cubital glands, but this disappeared within three days without further symptoms. A similar observation has been made by Kronig in his experi- ments ; in investigating the case I found a slight, but very distinct, retrogade lymphangitis in which the redness extended to the periphery. In these cases it has not been definitely ascertained whether during or after inoculation a slight infec- tion had occurred, such as may sometimes take place in vaccination, or whether we have to do with the results follow- ing a very strong reaction. This latter probability, I think, is the more likely one, since the affection appearing after the introduction into the body of heterogeneous albumin without bacterial infection also manifests itself as fever and by the swelling of glands. The subjective symptoms caused by the cutaneous reac- tion are extremely slight. A few times itching was observed in inoculated patients, which became more troublesome in long-persisting forms of reaction. But this cannot be of great objective importance, since it has never been observed that patients scratched the reaction spots ; nor was itching com- plained of, but its occurrence was only found out after having directly questioned the patients as to its presence. 26 THE OPHTHALMIC AND CUTANEOUS We must, however, mention the appearnce of phlyctaenae which were observed in children, following on inoculation. (See chapter on this subject.) After the injection has paled down and the infiltration disappeared, the skin assumes a brownish-red hue, which may for a long time remain, finally to desquamate. THE CONJUNCTIVAL REACTION. There are very few remarks to be made with regard to the conjunctival reaction ; after a period of from six to twenty- four hours the conjunctiva reddens slightly, which may be recognized by comparison with the other eye. Calmette, and other French investigators, as also Eppenstein, maintain that the caruncle reddens first. We cannot exactly confirm this observation, but the fact that the patients complain of strong irritation in the caruncle after the solution has been instilled on to the conjunctiva, speaks in its favour. Again, since the injected fluid is evacuated through the nasal ducts, and is somewhat obstructed at the caruncle, the latter remains longest in contact with the tuberculin solution, and thus it could be understood that the caruncle would exhibit the earliest and most marked symptoms. First Degree of tlie Beaction. In the first degree of the reaction no more than a redness of the conjunctiva is produced ; in the second degree the redness is more intense, and goes together with a fibrinous exudation of mucus ; in the third degree all these symptoms are increased in intensity, whilst at the same time there occurs a softening of the conjunctiva and a serious infiltra- tion (chemosis).^ When the 1 per cent, solution of old tuberculin is employed, symptoms indicating the third degree are of extremely rare occurrence. In second and third degrees of the reaction one often finds that the follicles of the conjunctiva are seen quite distinctly. The strong reactions produce the symptoms of ordinary con- junctivitis : irritation, photophobia, and a feeling as if foreign bodies were present. In all but a few cases the clinical symptoms are very slight, and are confined to a feeling of ' Very rarely slight haeiuorrhages are observed to occur on the con- junctiva (ecchymoses Letulle). DIAGNOSIS OF TUBERCULOSIS 27 heaviness of the eye-Hds, and to that of foreign bodies above referred to ; but they soon disappear. In the case of sensitive individuals the apphcation of boric acid fomentations, and in all cases the use of a drop of 3 per cent, cocaine solution, have at once put an end to all discomfort. Under certain circum- stances a combination of cocaine and adrenalin (8 per cent, and 1 per mille.) might be tried. The same has been found useful in analogous cases of hyper-sensibility of the conjunctiva from irritation by pollen. Secondary Symptoms. Even in cases in which highly concentrated solutions were applied there was no sign of serious consequences. The symptoms pass away in a few days, and, as a rule, entirely disappear about the fourth day. A late conjunctival reaction is very rare : so far we have only been able to make sure of two cases. This may be because, when there is only a slight opportunity for reaction to occur, the tuberculin does not remain in the same place for a long time, and also because it is easily absorbed and thus enters the general circulation. We must also consider that the tuberculin, when instilled into the eye, does not retain its original strength for long, as it is quickly diluted by the lachrymal secretion which follows its application, and it is then partly carried away through the nasal duct. Nor have we been able to observe long-continued reactions on the conjunctiva; but it is necessary to pay further attention to this point, since Cohn has observed reactions lasting up to a fortnight {Berliner klin. Wochenschr., 1907, No. 47). As mentioned above, unfavourable secondary symptoms ^ have never been observed. One may safely assume that up- wards of 10,000 cases have been reported on up to the present. The number of those treated according to the conjunctival method is probably much larger. At first my colleagues ex- pressed their doubt as to whether one could venture to attempt the reaction on the eye, an organ of such vital importance. From the very beginning I expressed my firm conviction that no danger was attached to the reaction, although theoretically the objection could be raised that a bacterial infection might be superimposed on an already existing conjunctivitis. But my experiences with hay fever enable me to state that in about ' See. however, the chapter on Contra-Indications. p. 30. 28 THE OPHTHALMIC AND CUTANEOUS 200 cases which exhibited the severest conjunctival symptoms imaginable during eight to ten weeks continuously, I have never observed the slightest injury to the interior of the eye, not even when, through the introduction of pollen extract, the process going on in the conjunctiva was brought to a most intense degree of irritation. The symptoms, therefore, are limited to the conjunctiva ; between the conjunctiva and the uveal tract there exist no direct communications, or if they do, they are of very slight influence. Gommunicating Tracts in the Eye hetween the Conjunctiva and the Uvea. I am indebted to my friend Dr. Emil Levi, of Stuttgart, the oculist, for the description of the following anatomical conditions. The vascular systems of the conjunctiva and the uvea are distinct, but they are connected by means of the rami per- forantes of the anterior ciliary arteries. Together with the conjunctival vessels they form the corona vascularis sur- rounding the cornea, the so-called marginal network. This connection, however, is, comparatively speaking, insignificant (see Leber's table of the vascular course in Fuchs' " Lehrbuch der Augenheilkunde "). The most detailed description is found in Leber, that of the circulation and the nutritive conditions of the eye in Graefe-Samisch's " Handbuch der Augenheilkunde," Second Edition, vol. ii., 2. This vascular communication would explain the ciliary injection we once observed in a patient after the application of the conjunctival reaction, and in another case a pericorneal injection ; it not only explains the occurrence as such, but also the rarity of the occurrence. A transmission of the reaction to the interior of the eye, apart from any existing vascular connection, might also be brought about by means of lymph spaces and lymph passages (through the anterior chamber and the cornea), since drugs, such as atropin, &c., so much in use in ophthalmology, can only reach the interior of the eye in this way. However, in the case of albuminoid substances, less absorption seems to take place (compare also Leber on the absorption of colloid sub- stances through the cornea, he. cit., p. 387). But compared to this absorption, the one that takes place by means of the circulation, plays a role of even less importance (see p. 375), DIAGNOSIS OF TUBERCULOSIS 29 From a practical point of view, the absence of symptoms in the interior of the eye in the appHcation of the conjunctival tubercuHn reaction, is to be explained by the fact that the amount of tuberculin reaching the uvea is too small to pro- duce any irritation. But we must take into consideration that, after the first inoculation, the vessels, especially those of the interior of the eye (and more particularly m cases of changes in the cornea), may show different appearances with regard to the absorption of tuberculin. Communicating Tracts between both Eyes. Whilst it is certain that communication exists between the uveal tracts of both eyes — which on a purely clinical ground can easily be concluded from the occurrence of panophthalmia — no communication exists between the conjunctivse of the two sides. As I have been informed by ophthalmologists, Eversbusch has endeavoured to demonstrate the existence of such communications, but he has failed in his attempt. Our own observations are in harmony with this view ; it was only in very rare cases that we could detect a trace of redden- ing in the other eye. In order to explain this it is not neces- sary to assume the existence of a communication ; it is more probable that the patient with his finger introduced a trace of tuberculin into the eye that had not been tested. Cohn has observed that, when the instillation is repeated, there occurs not only a strong reaction in the eye in which the tuberculin has been instilled before, but that the eye that had not been tested exhibits hyper-sensibility when tuberculin is instilled into it, an occurrence which Cohn does not en- deavour to explain. It is difficult to interpret this fact unless one accepts the existence of a common tract. Many would not feel inclined to accept as an explanation the occurrence of a sympathetic transmission along nervous paths, the less so as Cohn's observations have not to do with the transmission of an inflammation but with that of a condition of over-sensi- bility, so that no other explanation remains but to assume that the tuberculin reaches the other along the ciliary nerves. But before framing great hypotheses it is necessary to ab- solutely exclude the simplest and most likely explanation — that of transmission of tuberculin by the finger — -by means of experimental arrangements, and to see whether in that case the hyper-sensitiveness of the other eye, described by Cohn, will still show itself. 30 THE OPHTHALMIC AND CUTANEOUS We may assuredly exclude the transmission of tuberculin along a tract, since no direct communication exists between the conjunctivae. On the other hand, according to ophthal- mologists, the possibility of transmission by means of sym- pathetic irritation along the ciliary nerves — especially in an already existing condition of irritation — vt^ould not be excluded (see Schirmer's " Die sympathetische Augenerkrankung," Grsefe-Samisch handbooks, vol. vi.). The most probable explanation of the accompanying in- flammation of the second eye was the transmission of tuber- culin by the hand of the patient. This possibility, as accounting for the existence of hyper-sensibility of the other eye without preceding inflammation at the first application of the reaction, may only be excluded when from the beginning the second eye has been made inaccessible by means of a watch-glass bandage, or by the celluloid protective dressing for inoculation, which I have recommended before. Other dressings, such as the much used monoculus, cannot be recommended, since they themselves may cause irritation. For this special purpose I even regard the celluloid dressing, which admits air, to be more appropriate than the ophthalmological watch-glass band- age, which does not admit air. The occurrence of irritation in the other eye is too rare to require systematic bandaging of the healthy one. On the other hand it would conduce to exact scientific results if the eye treated with tuberculin were in this manner pro- tected from being touched. CONTRA-INDICATIONS FOR THE APPLICATION OF THE CUTANEOUS AND CONJUNCTIVAL REACTIONS. Contra- Lndications for the Conjunctival Reaction {Conjunctivitis) . The discussion of the contra-indications follows naturally on the short anatomical and pathological observations of the preceding chapter. For this reason we will depart from our usual custom and deal first with the contra-indications of the conjunctival reaction. Calmette regards conjunctivitis as a contra-indication (see our remarks on the Technique of the Reaction). Although also Emii Levi {he. cit.) maintained, from a theoretical point of view, that the possibility was not absolutely excluded that altered conditions of absorption might be present in already existing inflammation, and that DIAGNOSIS OF TUBERCULOSIS 31 in such cases the reaction might influence the uvea, yet this idea proved to be merely a theoretical one, since in all cases of conjunctivitis no symptoms could be observed in the uvea, nor even more intense symptoms in the conjunc- tiva. I will not neglect to mention a case, communicated to me by Kronig, in which, eight days after a conjunctival reaction, an inflammation of the Meibomian gland developed which had to be incised several times. This isolated case can hardly be regarded as a contra-indication in any direction whatever, since such an affection may originate spontaneously as well as follow on an already existing conjunctivitis. In the manner I have indicated in the chapter on technique, the solution when used in a great number of cases, might easily contain bacteria which are brought from one eye into the other by means of the pipette. It is, therefore, quite probable that a case like this is not to be attributed to the tuberculin, and may possibly be prevented by taking aseptic measures. Wiens and Giinther (Miinch. med. Wochenschr., 1907, No. 52) attribute a strong reaction to an already existing conjunctivitis. This assumption is to be regarded as an erroneous one, since strong reactions are the result of the use of strong solutions (see " Ausfiihrungen iiber Tuberkulin- Test," by Hochst) and the repetition of the instillation (see Wolff-Eisner, Miinch. med. Wochenschr., 1908, No. 2). In the same number similar communications appeared on Contra-Indications in Conditions of Irritation in the Interior of the Eye. On the other hand, I enjoin great care in dealing with conditions of irritation in the uvea (iris), &c., and I believe the best thing to do is to ask the patient, before applying the reaction, whether he has ever suffered with his eyes or has ever been under the treatment of an oculist. Cases of iritis are not so frequent as to in any way mipair the diagnostic results, and I should not like to bring general discredit on the conjunctival reaction by reason of such isolated cases. However, the presence of tubercular changes in the interior of the eye I regard as an absolute contra-indication for the reaction, that is to say in the ordinary manner. A minute amount of tuberculin, which arrives in the eye affected with tubercle, would produce a strong reaction, the course of which, as far as the eye is concerned, can with difliculty 32 THE OPHTHALMIC AN]"* CUTANEOUS be predicted. If one wishes to apply the reaction in any case one must proceed until further experience has been gained in a way similar to what is usual in tuberculin therapeutics ; that is to say, one must apply so small a dose that one is niaster of the reaction. Instead of with 1 per cent, solutions one might perhaps begin with solutions of 1 : 100,000. The conditions for the tuberculin diagnosis in the eye are especially favourable. Whereas in other cases one can only once apply the reaction to the conjunctiva (see chapter on the Eepeti- tion of the Reaction), this does not hold good in the case of tuberculosis of the interior of the eye, since the reaction need not be judged from the appearance of the conjunctiva but can be observed in the tubercular focus itself. It is a very peculiar fact that, in their attempts to produce focus reactions with minute doses by means of the local apph- cation of tuberculin, ophthalmologists have not discovered the conjunctival reaction. In internal medicine attempts have been made to advance the science of diagnosis and thera- peutics by the introduction of tuberculin into the seat of tuberculosis. I need only mention here the diagnosis by tuberculin inhalations and the unsuccessful experiments of pulmonary injections. In the case of ophthalmology this therapeutic application, which would necessarily have led to the discovery of the conjunctival reaction, would have been very near, since, especially in ophthalmology, reports as to the very favourable results of the subcutaneous application of tuberculin in cases of tuberculosis of the eyes have fre- quently been issued, and ophthalmologists have this advan- tage over other medical men : that they are able at any time to see the reaction of the focus, and need not draw indirect conclusions by means of secondary methods. Observations in Cases of Tuberculosis of the Eye. (See also the Special Part, on the Application of the Reactions in Ophthalmology.) I almost regret not to have published before this these practically self-evident contra-indications, w^iich clearly follow from the theoretical bases underlying the use of tuberculin, and the anatomical structure of the eye. The}^ will, I hope, be still in time to prevent further harm. On October 7, 1907, Kalt communicated a case to the Ophthalmological Society DIAGNOSIS OV TUHKRCULOSIS 88 of Paris, iu which he had apphed the reaction to a patient suffering with corneal opacity, synechia, and opaque refractin^^ media ; there were no pericorneal injections, and from this it could be concluded that the changes were of long standing. On July '24 he instilled into the eye one drop of tuberculin 1 : 100, and the next day he recognised conjunctival and pericorneal injection. After six days the sclerotic focus was enlarged and had reddened ; after fourteen days the iris was in a very irritable condition, the injection of the vascular zone being very marked. On October 8 the symptoms had come to an end with sclerotation of the cornea. Kalt took the oppor- tunity to repeat his experiment in a ten-year-old boy, who had small tubercles on the iris and in the corner of the eye, which up to then had followed their course with but few symptoms of inflammation. But within eight days the reaction set up a severe inflammation which was not expected after the simple aspect it had worn before. Morax attributes the severe symptoms mentioned by Kalt to a combination of circumstances, which, however, I do not think to be very probable, though I do agree with him in attributing Terrien's discovery of the occurrence of small confluent nodules in the conjunctiva two months after the application of the con- junctival reaction to the possibility of a drop of tuberculin exciting a latent tuberculosis of the conjunctiva, although it cannot have been its cause. I am inclined to accept this contra-indication, although in one case Citron has applied this reaction in a case of tuber- culosis of the iris without the iris having been influenced, and Stephenson (British Medical Journal, No. 2,442, 1907), calling it " Calmette's serum reaction," as also Petit, warmly recom- mend this method for diagnosing the tubercular nature of diseases of the eye (iridocyclitis, sclerotitis, choroiditis). Other Secoiidarij Symptoms in the Application of the Conjunctival Beaction. Eppenstein recommends a solution of onl}' i per cent, in the case of children suffering or suspected to be suffering from tuberculosis, in order to prevent the occurrence of too strong a reaction. He has observed two cases in which tuberculous children, after the application of a 1 per cent, tuberculin solution, developed a slight keratitis, and once, in a boy who liad been suffering from lichen scrofulosum, a recurrent kera- 3 34 THE OPHTHALMIC AND CUTANEOUS titis occurred following on a conjunctival reaction. In these children there is no direct connection between the conjunctival reaction and the keratitis because the cornea is sufficiently near to the reaction that is taking place. Such children frequently suffer from keratitis ; a direct connection, however, could only be assumed if children not suffering with it would develop keratitis after cutaneous inoculations. All this con- cerns the hypersensibility with regard to the toxin of tubercle bacilli. (See Pfaundler, Milnch. Ges. f. Kinderheilk., ref. Mon. /. Kinderheilk., vol. vi., No. 3; Moro, Wiener Klin. Wochenschr., 1907. See also the chapter dealing with the application of the reaction in the diseases of children.) For more details I refer to the special part on the Application of the Reaction in Ophthalmology. THE RESULT OF THE REACTION IN TUBERCULOSIS OF THE FIRST, SECOND, AND THIRD STAGES IN HEALTHY AND IN DOUBTFUL CASES. Introductory Remarks. It is advisable to deal simultaneously with both forms of reaction, the cutaneous and the conjunctival. It is true that this way of dealing with the subject makes it more difficult to make use of the literature on the subject, since it is onl}^ Mainini (Milnch. vied. Wochenschr., 1907), who so far (up to January 1, 1908) has employed both reactions simultaneously. For this reason we are only able to adduce in general outlines the results obtained b}^ other writers on the subjects with either of these methods. Nor do we intend to publish every one of our 500 cases with their diagnoses, reactions, and further symptoms, as has been done by Petit in his work on ophthalmologic reactions, not only with regard to his own observations, but with all those obtainable in the literature on the subject. This would be altogether unnecessary for our purpose, for which it is of greater advantage to deal with the complete symptoms of patients at the time. The results have been taken from the material collected by Stadelmann, as well as my own, with the kind help of Drs. Blume, Mass, Miinzer, Steinberg, and Teichmann. Actually, the number of our observations has been far greater than that dealt with in these pages. First of all I have omitted my own 150 experiments, which were more DIAGNOSIS OF TUBERCULOSIS 85 or less instructive, as also all patients inoculated after Novem- ber 25, because the publisher was rather in a hurry concerning the appearance of these communications. These later in- oculations have only been taken into account in so far as they exhibited peculiarities not in accordance with the general rules ; for this reason the conclusions reached would not in any way have been altered. Following the customary rule we divide the tuberculosis patients into three groups : — First Stage : Symptoms at the apex ; unilateral or bilateral infiltration of the apex ; fever absent or slight. Second Stage : Infiltration of the upper lobes ; fever. Third Stage : Considerable infiltrations ; formation of cavities ; hectic fever ; decrease in weigfht. Our initial experiments clearly show^ed that there were great differences in the three stages as regards the manner of reaction, and this induced us to retain the distinction between these stages from the beginning. Tuberculosis of tlie First Stage. From the table it may be seen that out of 20 tuberculosis patients in the first stage, on whom the cutaneous reaction was applied, 16 gave positive, 4 negative results. In 5 the reaction was regarded as merely indicative of the disease, in 10 as showing the existence of the first stage, and in 1 of the second stage. We found that the majority (about 80 per cent.) responded, but the reaction was not a verj- strong one, five times showing merely a trace of the disease, which we do not regard as a pronounced reaction. The second degree of reaction occurred only once, the third not at all. The conjunctival reaction was positive in 14, negative in 6 cases (that is, positive in about 70 per cent.) ; in 4 cases it showed traces ; in 4 it gave the first degree ; in 4 the second ; and in 2 the third. The cutaneous reaction failed in 4 patients ; the conjunctival in 6. The difference is not a very great one ; the absence of response in these two cases may be attributed to errors as indicated above. But it is undeniable that in cases of tuberculosis of the first stage the conjunctival reaction is inclined to be the stronger of the two, so that this reaction frequently is the more distinct one. These figures, however, do not allow us to draw conclusions in favour of either reaction (in two cases the cutaneous reac- 36 THE OPHTHALMIC AND CUTANEOUS tion is more marked than the conjunctival, whilst in four the conjunctival reaction is more marked than the cutaneous). Tuberculosis of the Second Stage. Of 26 cases of tuberculosis in the second stage, 18 re- sponded to the cutaneous reaction, and 15 to the conjunctival reaction. The cutaneous reaction gave 8 traces, 8 reactions of the first degree and 2 of the second. The conjunctival reaction gave one trace, 9 of the first, 4 of the second, and 1 of the third degree. In 2 cases the cutaneous reaction was stronger, in 3 cases weaker than the conjunctival reaction. Again, these figures do not permit one to draw definite con- clusions as regards either reaction. Tuberculosis of tlie Third Stage. Of 18 cases of tuberculosis of the third stage only 3 responded to the cutaneous reaction, of which 2 only gave traces, and 1 the third degree of the reaction. The con- junctival method gave .5 positive and 13 negative results. Of these 1 gave traces, 2 were of the first, and 2 of the second degree of reaction. One of these latter was the same patient who, in the cutaneous reaction, responded in the third degree. These figures, like the former, do not show any very important differences between the two reactions. Conclusio)is. These results allow us to draw the following important clinical conclusions : the reactions call forth a response in a great number of cases ; but the reactions are not in proportion to the extent and the degree of the tuberculous changes that have taken place, and the symptoms caused by the disease. The strongest reactions observed did not occur in clinicall}^ manifest cases of tuberculosis, but in cases in which the patient had not even been regarded as a suspected sufferer before this test had been applied. It is evident that in the cutaneous, as well as the conjunctival, reaction, the number of responses decreases in percentage as the tuberculosis pro- gresses, and they fall to a minimum in the most advanced cases, the very severe cases of the third stage. The tables would wear a somewhat different aspect if we would retouch them in accordance with the very latest data, i.e., the results of post-mortem examinations. But we have held on to our DIAGNOSIS OF TUl^EllCULGSIS 37 rule and only registered the cases as they have been reported upon, according to the inoculation, by the investigator in question, w^ho was absolutely neutral as to the results of the inoculations. The reactions in clinically manifest cases of tuberculosis have this in common, that they respond to the reaction within a few hours, that they reach the greatest intensity shortly afterwards, and that they regress just as quickly. Only one case of late reaction has been observed in ()4 cutaneous applications. No further remarks con- cerning the course of the conjunctival reaction are neces- sary, since, as we have stated above, even in non-clinical cases of tuberculosis late reactions have but very rarely been observed on the conjunctiva. Judging by the slight differ- ences between the figures in either reaction, it can hardly be decided whether either method is to be preferred to the other (in cases of tuberculosis). We can only form a proper judgment when we can take the other categories into con- sideration, especially the reaction in healthy individuals. In tuberculosis of the third stage, the occurrence of the reaction is the exception ; our statistics give '28 "pev cent, (conj.) and 15 per cent, (cutan.) Most of these only give traces, and those cases in which the reaction was a strong one are exceptions and will be dealt with specially later on. However, our results, which I believe to be absolutely reliable and in- disputable, are altogether contradicted by the statistics of A. Frankel, who, in tuberculosis of the third stage, obtained positive reactions in about 45 per cent, of his cases. ^ (See Cohn, Berl. klin. Wochenschr., 1907.) The statistics of Frankel are based upon comparatively few cases, and the difference as far as the conjunctival reaction is concerned (about 80 per cent.) is not so great as would have been the case if only the results of the cutaneous reaction had been considered. For this reason we can here only repeat our advice to apply, if possible, both reactions, and generally ' Eppenstein, who did not distinguish between the three stages, obtained positive reactions in practically 100 per cent., but only after having repeated the instillation. After single instillations, out of 40, only 24 responded. Since he did not distinguish between the stages, we may not, from our point of view, regard his material as reliable. So also Schenck and Seifert (MiiucJi. Med. WoclienscJir.) obtained response in 100 per cent, of their patients by twice repeating the conjunc- tival reaction. At the first instillation the responses amounted to about 80 per cent. (78-57 per cent.). 38 THE OPHTHALMIC AND CUTANEOUS speaking — after having taken into consideration all possible secondary influences — to ascribe a higher diagnostic value to the conjunctival reaction, and a greater prognostic significance to the cutaneous method. Again, in considering Frankel's figures, one must consider the fact that he proceeds from the theoretical basis that, if there is no response in cases of evident tuberculosis, this is to be attributed to a faulty reaction. And since the reaction has so great a diagnostic significance, we can easily understand how an extremely weak reaction may be regarded as positive, if this is suggested by the knowledge that the disease is present. The figures quoted by Klieneberger in the Mi'inch. Med. Wochenchr., 1907, No. 52, and used by him to criticize the method, agree very nearly with our own. (Negative reactions about 50 per cent. ; out of 7 cases of tuberculosis 2 were in the second stage, and 1 in the very early stage ; out of 4() cases of healthy individuals, 8 [ = about 16 per cent.] responded in a positive way to the conjunctival reaction ; when repeated 78 per cent, gave a positive reaction. This agrees with Fritz Levy's statistics.) These figures agree to a surprising extent. The fact that he draws conclusions which I regard as unjustified does not in any way impair the similarity of the results. From the result of our observations — to the effect that the number of the responses decreases as the disease pro- gresses — it is evident that, when it is ascertained that tuber- culosis does exist, the result of the reaction enables us to conclude a posteriori which stage the disease has reached. However, if no response is obtained, this does not so much indicate the stage of the disease ' as the absence of the Lenhartz (Arztl. Verein Hamburg; Milnch. Med. Wochenschr., 1907) obtained positive cutaneous and conjunctival reactions in 22 cases of tuberculosis ; in 15 cases they were varied inversely according to the strength of the tuberculin solution employed. Letidle (C. B. de la Societc de Biol., 1907) obtained in 66 cases (conj. i 63 positive reactions, of which 50 + + +, 2 +-f -}-, 11 +, 3 - (2 moribund, 1 diagnosis dubia). Schubert (Ges. f. Xatur- u. Heilkunde, Dresden, MiincJi. Med. Wocheji- ^chr., 1907, p. 2504) obtained out of 19 cases, 17 positive and 2 negative reactions. ' Eppenstein and other writers on the subject were not able to find a clear connection between the strength of the reaction and the stage of the disease. Smce it is true that such a connection does not exist from their particular point of view, they must have lost sight of the prog- nostic significance of the reaction. DIAGNOSIS OF TUBERCULOSIS 89 capacity to react, and this enables us to draw important conclusions ; for a number of non-reacting cases exist in the first and second stages, whilst in the third stage we find only a few cases in which reaction is positive.' Now it is not likely that a first stage of tuberculosis could be regarded as a third stage simply because there is no reaction, but it is quite possible that patients in whom the capacity to react is absent very quickly pass from one stage of the disease to another, and it is this point which has been absolutely justified by the results of our observations, to which, from the beginning, I have directed the greatest attention. It is extremely important to diagnose in a re- liable manner tuberculosis in its very early stages, and where it is only suspected ; but clinicians have no difficulty what- ever in diagnosing an evident case of tuberculosis from the first to the third stage. It is of much greater importance to be able to make a jjrognosis in a simple manner, or, putting it in other terms, to be able to recognize whether the body is in a condition to resist the progress of the disease, by means of treatment, living in the country, suitable climate, sanatoria, seaside residence, 6i.c. The stage of the disease offers but few indications by which one can prognosticate the course of the disease. An initial catarrh of the apex may lead to death in the course of a year, whilst a case of advanced phtliisis of the second or third stage may allow the patient several years of life under comparatively' favourable conditions. A clinician who pointed out this fact called the attention of his pupils to the possibility of such a lengthening of life in the case of tuberculous patients when the external conditions were of a favourable nature (see, for instance, Striimpell, " Lehrbuch der inneren Medizin "). At last we begin to see what these favourable conditions consist of: viz., a reactive capacity of the body, which may be increased by favourable external conditions, or decreased to a minimum by unfavourable conditions, such as insufficient nourishment, &c. ' It must be mentioned that in Kronig's cases, which he has been good enough to shoAV me. the number of strong reactions in the third stage was greater than witli us. Almost without exception the\- were cases of advanced phthisis, which, however, had taken a very slow coiu'se. Klieneberger is of opinion that no response in the early stages of the disease indicates some defect in the method, whereas we regard it as being one of its most important assets. 40 THE OPHTHALMIC AND CUTANEOUS Now it cannot be denied that medical men, who have had a great deal of clinical experience with regard to tuber- culosis, have often been enabled to predict the course of the disease. But even a medical man with the greatest ex- perience could but base his prognosis on a very long, clinical observation, and it will readily be granted that before such experience had been attained he must frequently have failed in his prognosis. There are very few diseases in which so many living witnesses for many years have belied the doctor's hopeless verdict as is the case with tuberculosis of the lungs, and, more frequently still, a favourable course of the disease is predicted, whilst instead the victim falls an early prey to the scourge. It is not to be doubted therefore that a reaction which guides our clinical judgment with regard to a prognosis of the disease along the right channels would be a very great boon, and, as an aid, is equal in value to a diagnostic method. In order to prevent misunderstanding we should explain that a positive reaction indicates the existence of the capacity to react, whereas the unresponsiveness to the reaction indicates a lack of this capacity. We will refer later on to the theoreti- cal significance of the reactive capacity, or, what is the same, the sensibility or insensibility respectively. I will only say here that, according to our observations, the absence of the reactive capacity is in all cases to he regarded as an ominous sign, whilst the positive reaction, on the other hand, merely indicates the reactive capacity, i.e., that the body is capable of fighting a battle. The result of the battle does not depend on the one factor, the reactive capacity alone, but on a great many other factors, especially on the virulence of the bacteria in question ; but the reactive capacity in this struggle is a factor of importance which is not to be underrated. And when the reaction indicates that the body is capable of combating the infection, we are justified in assisting the body in its struggle with all the means at our disposal. From the figures quoted above it would follow that, as the disease progresses, the reactive capacity of the body decreases ; from 80 per cent, in the first stage it drops to about 60 per cent, in the second and '20 per cent, in the third. Now I should like to add that we may not identify the quick progress of this process with what is usually understood by an " ad- vanced " or "inhibited" disease; as already stated, a process .showing slight changes may quickly become intensified, whilst, DIAGNOSIS OF TUBERCULOSIS 41 on the other hand, a slowl}' progressing disease may in the course of years become inhibited. Therefore the decrease of the reactive capacity need not necessarily correspond to the stage which the disease has reached. If, notwithstanding, on the first glance there seems to exist a remarkable corre- spondence between the stage of the disease and the reactive capacit}', this has to be attributed to the fact that as a rule a patient with catarrh of the apex resists infection, and there- fore possesses reactive capacity, whereas in the patient with advanced pulmonary symptoms the reactive capacity has been exhausted in the struggle, and the disease quickly intensifies and leads to death. How, in the course of the infection, the reactive capacity and the sensibility gradually cease, we will leave to be dealt with in our theoretical observations. It seems to me that it would be of interest to quote one or two especially significant histories of cases, so as to give some example of the prognostic significance of the reaction, and to point more particularly to such cases which apparently to some extent contradict our explanations by showing a nega- tive reaction in the first stage, and a positive reaction in the third. For this purpose I make use of our own material, that is to say, of the excellent compilation by Professor Stadel- mann, to whom I will again express my thanks for kindly allowing me to use it in this publication. HISTORY OF SOME CASES. We will first of all deal wdth a case which, clinically, was regarded as being in the first stage ; negative reaction was obtained. Ovinsk (Pavilion 17), admitted on October 1-5 : very few pulmonary symptoms ; slight ulcers on the larynx. •21.x. 1907 1st 2ud 3rd 4th day 1st 2ud 3rd trace - - Cutaneous Conjunctival i.e., a swiftly passing trace of redness, lasting only one day. The reaction did not agree with the apparently unimpor- tant lung symptoms. Percussion sound everywhere normal ; breath sounds vesicular ; over the right apex a fairly distinctly prolonged expiration, accompanied by a few very dry, crackling sounds. This, therefore, according to the clinical examina- tion, was a simple case of catarrh of the apex. To this was added hectic fever, for the appearance of which no explaua- 42 THE OPHTHALMIC AND CUTANEOUS tion could be found. In the next few weeks infiltration and catarrhal symptoms appeared over the two superior lobes ; yet the clinical result could still be regarded as insic;nificant. Notwithstanding these unimportant clinical synjptoms, and despite the absence of diarrhoea and severe bodily pains, a rapid loss of strength took place which, after about six weeks, led to a high degree of cachexia and to the formation of a great hasmorrliagic exudate; the patient died on November 11. The autopsy gave few progressive processes in the two superior lobes, but brought to light an extensive tuberculosis of the pericardium and of all the abdominal organs, as also enormous intestinal ulcers. From these facts it appears that the negative result of the reaction enabled us to form a better prognosis than the clinical investigation ; also that the contradiction between the clinical lung symptoms and the reactive capacity enabled us to draw the conclusion that the disease in the lungs would either rapidly intensify, or that clinically the extension of the disease in the lungs had been diagnosed inaccurately, or that also in other parts of the body extensive tuberculosis existed. This, therefore, was one of those cases of the first stage of tubercu- losis in which the altogether negative results obtained by the cutaneous reaction, and the slight traces indicated by the con- junctival reaction, were explained on post-mortem examination. The following will be one out of numerous cases in which a positive reaction was obtained in the first stage of lung tuberculosis. A patient who clinically was regarded as being in the first stage (Gade, Pavilion 17), showed symptoms of an infiltration of the left apex, slight fever, and a good many catarrhal symptoms over the lungs. In the sputum bacilli were found ; — 4th day 1st 2nd 3rd + + + Cutaneous + + + + decreasing Conjunctival This great difference suggested a probable mistake in the technique, and a few weeks later the reaction was repeated on the other eye. A distinct positive result was obtained of between the first and second degrees. The case, therefore, was to be regarded as of fairly favourable prognostic value. This case was under observation for three months ; the patient was slightly feverish ; sometimes the fever disappeared DIAGNOSIS OF TUBERCULOSIS 43 altogether ; the process in the lungs sometimes showed signs of regression ; progress could not be definitely ascertained ; the patient increased in weight (6 lb.), walked about, and was fairly well subjectively ; there was but little coughing and expectoration of sputum, and catarrhal symptoms were few. Another patient in the first stage (Kleindienst, Pavilion 17) gave strong conjunctival and cutaneous reactions with a distinct indication of infiltration and a decided and direct early reaction (see charts). The reaction was one of the strongest observed. The circumference of the papule amounted to "28 mm. by the use of a 2-5 per cent, solution of old tuberculin. The disease took a distinctly favourable course ; the general condition was good ; the patient for the most part was slightly feverish (37-3H'C.) ; the lung symptoms partly regressed, and he gained 5 lb. in weight. From the result of the reaction we may conclude that the patient had reactive capacity. We could observe that it was this reactive capacity which enabled him to arrest the process during his stay at the hospital, and to improve his health. MILIARY TUBERCULOSIS. AVe have not had a very great deal of experience with regard to the result of the reaction in cases of miliary tuber- culosis. But, according to the literature, the reaction is a negative one in most cases. We have observed a few cases of partial distribution of miliary tuberculosis, especially at the base of the brain. In one case (Stramm, Pavilion 17) in which, according to the clinical diagnosis, signs of extensive infiltration were found in the left upper lobe, together with a meningitis tuberculosa, the reactions were negative. As a rule a negative result is mentioned in the literature, also in meningitis tuberculosa ; but this result seems not to be a constant one, since in another case under our observation the reaction in meningitis tuberculosa gave a positive, in a third case again a negative, result. In children, the reactive capacity seems to last longer (nearly to close on death !) in miliary tuberculosis than is the case in adults, where it disappears some time before death takes place. But the negativeness of the reaction is only relative. Experience with the subcutaneous injection of tuberculin shows that it can be overcome by large doses. von Pirquet obtained the following results in children 44 THE OPHTHALMIC AND CUTANEOUS suffering with miliary tuberculosis {Wiener klin. WochenscJir., 1907, No. 88) :— 20 -10 days 10 — days (oite mortem 9 + 13 + 13 - = Positive reaction. — = Negative react "lu three cases the decrease in the reactive capacity could be definitely ascertained ; this began about three weeks ante mortem, and in some cases it decreased to zero, in others it did not decrease to quite that extent, whilst in one case a child showed strong reaction even three days before death."' The course of those cases representing progressive pul- monary changes of stages II. — III. and III. which have produced a distinct, positive reaction, must be specially mentioned. The greater number of such cases represent exceptional conditions. Case 1. — Post mortem showed but slight alteration of the lungs, and death was due, not to phthisis, but to amyloid degeneration. Case 2. — The disease, after rapid progress, unexpectedly came to a standstill, and took a favourable course. Case 3. — Took a very favourable course, and suddenly died of haemoptysis. Cases 4 and 5. — The disease took a progressive course despite positive reactions (see specifications below). The history of these cases is very interesting and instruc- tive (see Stadelmann, Deutsche Med. Wochenschr., 1908). Case 1, a patient in the third stage (Pavilion 1, Giihlsdorf), gave very strong cutaneous and strong conjunctival reactions. Post-mortem examination showed very severe amyloid degeneration of intestines and kidneys. Compared with this the tuberculous affection of the lungs was but slight, since it only represented stage I. — II. The strong reaction in this case is thus easily explained, although it cannot be exactly shown what the amyloid degeneration was caused by. Case 2. — A young girl (Hein, Pavilion 1 ; second stage of tuberculosis ; under observation from March 12 to July 2.5) with extensive catarrhal symptoms ; sputum contained tubercle bacilli ; high fever ; continuous decrease in weight, so that clinically no doubt was entertained that the disease would take a rapid and unfavourable course. The result of the reaction on May 9, 1907, was strongly positive (strong con- junctivitis with purulent secretion, reaching a maximum DIAGNOSIS OF TUBEKCULOSIS 45 after thirtj^-six hours, and then regressiiif^;). According to my explanations concerning the prognostic significance of the positive reaction, there was a possibihty in this case that the process might still be arrested and perhaps take a more benign course. And indeed the fever suddenly ceased, and the subjective condition became so favourable that the patient was able to remain in the open. Soon the expectoration ceased altogether, the patient's weight increased 10 lb. in the course of four weeks, and she was dismissed with diminished resonance over the lower lobes, and weakened breath-sounds with slight rattling. Case 8. — In one female patient (Lindeinann, 1) we found tuberculosis of the second to the third stage. Three weeks before death there was a strong positive response to the con- junctival, a negative response to the cutaneous, reaction. This led us to expect that the disease would not progress much, and that death was not yet near ; the further course of the disease was in accordance with this assumption. The fever decreased, remained insignificant, and sometimes was entirely absent. The weight of the patient did not decrease ; she sat up for hours when she felt well. Suddenly she got a strong attack of haemoptysis, and died in the course of a few days. We do not believe that this case is in contradiction with our theories. The reactive capacit}^ indicates the possibility of a favourable course, based on the capacity of the body to prevent the extension of the symptoms of the disease. Haemoptysis is accidental, and, although it cannot be com- pared with a traumatic affection, it yet represents a coin- cidence not directly connected with the extension of the symptoms of tuberculosis. Following on these cases we will mention the exceptions, all of which turn on the point that, despite a positive reaction, the course of the disease was not so favourable as might have been expected after the favourable experiences in the other cases. Case 4. — A female patient (Madame Schulz, Pavilion 1), who gave strong reactions, but who nevertheless showed severe clinical symptoms with high remittent fever. For six to eight w'eeks the disease has not realh^ become worse, yet the affection in the lungs increased. At the end of the observation the same hyper-sensibility existed. The reaction, when repeated, was again a strong one. 46 THE OPHTHALMIC AND CUTANEOUS Case 5. — Another patient (Schroder, Pavihon 17) showed, post mortem, extensive King symptoms and fresh dissemina- tions over the inferior lobes despite strong reaction. Such exceptions are very rare. Stadehnann, in writing on the subject, says : " There certainly are exceptions, but they seem to be extremely rare." They occur in cases in which the reactive capacity of the body could not win the battle against the other factors in operation. On the other hand, cases showing a defective reactive capacity hardly produce any exceptions as to their unfavourable course. Of those clinically regarded as the first stage one has died (Kohler, Pavilion 17) and yost mortem it was seen that the changes were actually those of the third stage, and that the reaction had given a better idea of the disease than the painstaking clinical investigations. Of those in the second stage who did not produce a reaction one died, whilst four passed on into the third stage and are expected to die within a short time. Of 18 patients in the third stage 15 did not give a cutaneous reaction and 13 did not give a conjunctival, and out of these 14 have already died, after six weeks. Also, according to the observations of Pirquet {Wiener Mm. Wochenschr., 1907, No. 38), the reactive capacity in children decreases shortly before death. Out of five tests only one reacted positively six to nine days before death. Out of 24 cases investigated during the last ten days of life 13 reacted negatively and only 11 positively. The decrease can also be measured quantitatively. Days ante mortevi Dilutions Cutaneous reaction 14 11 6 1 : 1000 1 : 100 1 : 61 + THE BEHAVIOUR OF THE REACTION IN HEALTHY INDIVIDUALS. Now that we have dealt with the course of the specific reaction in individuals in whom distinct tuberculous changes have taken place, that is to say, in those individuals in whom it would be easiest to obtain a picture of the course of the specific reaction, I think it desirable to follow the course of the reaction in healthy individuals. By the term " healthy " DIAGNOSIS OF TUBERCULOSIS 47 individuals we mean people who are being treated in hos- pitals for other diseases, and who, according to the most painstakmg clinical investigations, do not give us any reason to suspect the presence of tuberculous changes. It could, of course, not be doubted that amongst these so-called " healthy " individuals there were a few in whom tuberculous changes had taken place, and a great number with latent and healed up or inactive tuberculous centres. We think it desirable that such healthy individuals should also be examined in order that their behaviour with regard to the reaction should become known ; suspected individuals always present difficult conditions, since amongst them we find partly tubercular patients in the beginning of the first stage of the disease, and partly healthy individuals. It will therefore become much easier to judge the results when we shall be able to point to distinct differences between non-tuberculous and tuberculous people in their behaviour with regard to the reactions. Our material amounts to 192 cases, exactly half of which, by cutaneous reaction, gave positive results, whilst the other half gave negative reactions. Of the positive cases 28 exhibited traces, 59 the first degree of reaction, 8 the second, and 1 the third degree. In 26 persons there was late reaction. As we shall see further on, special significance attaches to the late reaction, and one might even be justified in not regarding the latter as a positive reaction. But even after this deduction 70 positive cutaneous reactions remain against only 35 positive conjunctival reactions, of which latter 14 gave traces, 17 the first degree and 4 the second degree of reaction. From the number of positive conjunctival reactions one may deduct two late reactions. The Difference in the Besiilts of both Beaciions in CUnicalhj Unsuspected Individuals. — In tuberculous patients hardly any essential differences could be found between the results of the cutaneous and the conjunctival reactions, the one reaction being not even stronger than the other. We here find for the first time a much greater responsiveness in the cutaneous reaction, that is to say, a responsiveness which is to its disadvantage, because we have to do with clinically healthy persons. In thirty-two cases the cutaneous reaction was stronger than the conjunctival reaction, whilst in only three cases the opposite was the case. This frequent response in cutaneous reactions 48 THE OPHTHALMIC AND CUTANEOUS in apparently healthy individuals has caused even its dis- coverer to regard the reaction as of diagnostic use in the case of children only/ and for this reason Citron (Berl. nied. Ges., July '24, 1907) simply regards the cutaneous reaction as useless (" all persons show this [cutaneous] reaction, except some cachectic individuals"), whilst the majority of French writers, who have worked with the ophthalmic reactions, have hardly taken the trouble to carry out any investigations with the cutaneous reaction.- AVe must here state that the possibility of coincidence in the material or of mistakes is excluded, the more so since the results arrived at by other authors point to the same thing, and even show a tendency to more clearly indicate the differences (Frankel, for instance, only obtains 5 per cent, reactions ; Eppenstein at most 5 per cent, with the conjunctival method). Bandler and Kreibich {Deutsche med. Wochenschr., 1907, No. 40) obtained 88 cutaneous reactions out of 02 healthy individuals, &c. Probable Explanation. — It is not a very easy matter to explain these differences in the results, but from a clinical point of view this divergence of both methods is of very happy auoury. The difference between the reactions is as follows : The cutaneous reaction indicates the presence in the body of an inactive and encapsuled tubercular centre, which does not manifest clinical symptoms, and which may not even be capable of again becoming active ; in other words, it mdicates that at one time or another the body has in some way come in contact with tubercle bacilli. The ophthalmic reaction, on the other hand, merely indicates that active or semi-active ' Burnet {Soc. de Biol.., 1907) has observed the course of the reaction in himself and his colleagues, that is to say, in clinically healthy in- dividuals ; it seems to have been a reaction in the sense of our late reaction. - It is interesting to note that Schenck and Seiffert had tlie same result witli regard to the conjunctival reaction as we have had with the cutaneous, by repeating the conjunctival reaction three times in clinically healthy persons (positive reaction in 50 per cent.). This supports our conclusions to an extraordinary extent (see the theoretical part). Lenhartz in forty healthy people found both reactions to be negative (four had not been treated with the conjunctival method) ; in all other cases fovir gave positive reactions. Schubert always obtained negative reactions in healthy individuals. Mainini (Miinch. med. Wochenschr., 1907, No. 52) is the first to follow our idea, and to apply both reactions in each case. DIAGNOSIS OF TUBEKCULOSIS 49 tubercular centres are present. These conclusions, apart from clinical observations, can be arrived at by studying the figures. Now it is much more probable that out of 192 apparently healthy people, 35 (= about one-sixth) would possess imper- fectly cured tubercular centres than 96.' Further confirma- tion of the actual presence of tubercular centres in the body is obtained when the tuberculin reaction (see this chapter) coincides with the other. Greater Clinical Significance of the Conjunctival Reaction. — It is not to be doubted, therefore, that the conjunctival reaction certainly can claim and does possess the greater clinical significance. Clinicians do not interest themselves in activities of the past ; they wish to know whether in any particular body they are investigating there are still tuber- cular centres claiming their attention and care. They feel comparatively little for inactive, encapsuled, healed up, expired, harmless processes. I do not mean to say that the cutaneous reaction should be neglected ; one may not maintain that it is without any clinical interest whether in any par- ticular case a tubercular process has healed. In fact, in the investigation of a great number, and for purely scientific purposes, such observations are of the utmost importance. The similarity in results of both these reactions make the cutaneous reaction a very miportant means of control, since technical mistakes do sometimes occur ; the more so since the non-appearance of the reaction is indicative of important prognostic conclusions, which might be based on a mistake if the conjunctival reaction alone were relied upon. It is especially important to ascertain that the cutaneous reaction only very rarely fails, but that if it does fail in tuberculosis, it fails together with the conjunctival reaction ; and it is only when we find that both reactions fail to respond in one and the same case that we may safely draw the so very important prognostic conclusion. The cause of the difference in these two methods probably lies — as has been pointed out before — in the different absorp- ' We cannot possibly enter into a discussion with those authors who never, or hardly ever, obtain conjunctival reactions in clinically health}' individuals. Even if we assume that the authors in question have greater capability than ourselves in instituting clinical investigations, yet autopsy and subcutaneous tuberculin injections clearly show that tuberculous centres may exist in clinically healthy individuals ; it is at least very probable that our observations are the more correct ones. 4 50 THE OPHTHALMIC AND CUTANEOUS tive capacity of the parts of the body in which the reaction is performed. The skin, having but shght power of absorption compared to the conjunctiva, retains the concentrated tuber- cuHn until the ceUs — which for a long time had ceased to be influenced by tuberculin by reason of the tubercular centre being latent — regain their reactive capacity through tlie con- tinuous stimulation.^ The quickly absorbing conjunctiva, however, remains for too short a time in contact with the tuberculin, the anatomical conditions aiding the expulsion of the instilled fluid. It is evident from our explanations concerning the reasons inducing earlier authors not to attach any clinical value to the cutaneous reaction that the latter must have found an even larger percentage of positive reactions than ourselves. And, in fact, von Pirquet found that only new-born infants and young children do not react, and that with increasing age the percentage of positive reactions increases until, with very few exceptions, all adults respond positively, except in very advanced cases. It is difficult to explam why we should have obtained reactions in fewer cases. It may be because of our technique ; also we do not regard as positive — contrary to Pirquet — the very slight traces of a reaction, but only clear and distinct reactions, the only course to pursue for practical purposes and in experimenting on a large scale. The autopsies prove most decidedly that we were right in our conclusion (which in the first instance had been based on figures only) with regard to the clinical superiority of the conjunctival reaction, and that the cutaneous reaction is so very sensitive that it brings to light the presence of inactive tubercular centres in the body. I append a table compiled by Stadel- mann, which shows this very clearly. The Autopsies sitpport the Value of the Cutaneous Reaction for Discovering Latent Tubercular Centres. — In five cases in which no response was obtained, either in the cutaneous or the conjunctival reaction, no trace of tuberculous changes could be found on. post-mortem examination. The same held ofood for a sixth case which had shown traces in cutaneous reaction, whilst out of eight cases in which a response was obtained in the cutaneous, but not in the conjunctival re- action, six possessed old encapsuled and healed up tubercular ' In order to avoid being misunderstood, we emphatically refer to the Theoretical Part. DIAGNOSIS OF TUBERCULOSIS 51 centres in different parts of the body, whilst in only one case there was visible tuberculosis ; in one case only — that of a cachectic individual — recent tuberculosis was found in the apex. (In this case, according to Stadelmann, an experi- mental mistake has been made. I am inclined to leave the question open.) Considering the enormous number oi post-mortem examina- tions carried out at the Friedrichshain Hospital, the number of cases above mentioned is not very great. This, however, is due to the fact that we have not taken into considera- tion any post mortems on advanced tubercular patients, because little interest is attached to them ; a great number of people come into the hospital in a dying condition, the reactions not being applied, since no definite conclusions could be drawn from their behaviour. POST-MORTEM RESULTS. We will give the post-mortem results in the form of a table. The reason for the small number of investigations has been discussed above. Future investigation wall have to be carried out, especiall}^ along the lines of post-mortem examinations, because the time elapsed since the discovery of the reactions has not been sufficiently long for the collec- tion of a great deal of post-mortem material. The following points have to be taken into account for the scientific appreciation of the post-mortem results. That there are no apparent tuberculous changes is no proof of the absence of tuberculosis. This has been made clear by Orth many years ago. But even if microscopical alterations cannot be found, experiments on animals may often prove the presence of tuberculosis (see the elaborate treatment of the subject by Kabinowitsch, Berl. klin. Wochenschr., 1907). For all highly scientific purposes this course must be followed. Oi\\ev post-mortem results confirm our statements : — VON PiRQUET [Wien. klin. Wochenschr., 1907, No. .38) in Children. Author Cases Cutaneous reaction Post-mortem examination von Pirquet Comby ( Fresse vied. , Aug. 10, 1907.) 31 ) 52 4 ) 6 1 + 1 + Tuberculous changes, at auy rate caseated glands. No tuberculosis. Tuberculous changes. No tuberculosis. There was only one case with reaction without microscopically visible tuberculosis. D -5 E o c d i o No tuberculous processes of any kind. Remarks z; o 5 1- - 1 1 1 1 1 1 1 CO 1 1 1 1 1 1 1 ^ 1 1 1 i 1 1 1 - 1 1 1 1 1 1 1 c « 5 1 (^ 1 1 1 1 1 1 1 1 « 1 1 1 1 1 1 1 Trace 1 \- 1 1 1 1 1 1 1 DinK"0»is Ulc. endoc. . . Myocarditis Leucocyth;emia Pericarditis.. Hepatic cirrh. Sclerosis of arteries, cholelith. . . Syphilis III., aortic insufficiency 1 is (1) Matthes . . (2) Lciw (3) Burgh. .. (4) Lieb (5) Fuchs .. (6) Wiese .. (7) Hiichstn. .2 .= -^^ O f-i '— ■^ a, &, a p< ce c3 eg ^ i^.S a o o a > 1^ 'co o 03 03 CO : : s OJ Pi h a 03 "Vi &4 s o :0 cfi WO •^ ^ ^^ ^ , .^-^ ^^ , — ^ , — . ::i 'N CO ;2^ L- DIAGNOSIS OF TUBERCULOSIS 58 Von Pirquet (Wien. klin. Wochenschr., 1907, No. 38): In patients found tuberculous on post-mortem examination, the reaction had been partly positive, partly positive when re- peated, partly negative ; out of 13 : 6, 3, 4, respectively ; in totally calcified tubercular centres the reactive capacity in one case disappeared without any explanation being afforded by the autopsy. Late and Lasting Beaction the Reaction of Healthy Indi- viduals. — As we have already noticed, a late reaction only occurs extremely rarely in tuberculous individuals. We have already dealt with the type of the reaction in the case of actual tuberculosis, and the curve figures further on clearly show the rapid increase and decrease of the reaction. Judg- ing by the great number of cases with positive reaction in healthy individuals, it would seem as if the late reaction is the reaction of non-tubercuious individuals, or perhaps of people with healed up tuberculosis. A^ery painstaking clinical obser- vations and post-mortem results support this view. Great significance is attached to the late reactions, since they will enable us to understand the hypersensibility reactions. We will return to this point in the Theoretical Part. The Conjunctival Reaction in Typhoid Patients. — After having dealt with the reaction in healthy individuals we will now refer to those observations which show that in the case of typhoid patients and convalescent persons a special reactive capacity exists with regard to the conjunctival instillation of tuberculin. Cohn, going by his own observations as well as by the literature, calculates about 80 per cent. Although, of course, the calculation of percentages on so small a number of observations is not very reliable, yet this number is very remarkable, especially since Kraus, Lusenberger and Kuss {Wien. klin. Wochenschr., 1907, Xo. 45) found out of twelve cases of typhoid eleven in which a conjunctival reaction by tuberculin w^as obtained. A. Frankel suggested that individuals saturated with bac- terial toxins would exhibit some sensibility with regard to every form of bacterial albumin, which at the same time would go to prove the often doubted specific nature of the albuminous substances constituting the separate bacteria. Kraus arrives at similar conclusions. Testing the problem on our own material, apart from any theoretical reasoning, we come to entirely different conclu- sions. In eight typhoid patients the reaction proved to be 54 THE OPHTHALMIC AND CUTANEOUS absolutely negative, and I regard the fact that in most cases the cutaneous reaction, which had a tendency to turn out positive, has remained negative as a control to the correctness of these observations. Wassermann and also Fritz Levy inform me that in typhoid patients they did not obtain a reaction, whereas in convalescent persons it occurred frequently. We have not been able to obtain positive reactions either in typhoid con- valescence or in articular rheumatism. THE BEHAVIOUR OF THE REACTION IN DOUBTFUL CASES. We now come to the most important, and at the same time the most difficult, part of our explanation, the behaviour of the reaction in doubtful cases, of which only twenty have been under protracted observation. All these cases were suspected on clinical grounds of being in the very early stages of pulmonary tuberculosis, but no certainty was obtained. Sixteen showed positive cutaneous reactions, twelve positive conjunctival reactions. Just as is the case in healthy indi- viduals, positive results are more frequently obtained with the cutaneous than with the conjunctival reaction, but the dif- ferences here are not so great. To begin with, we must assume that amongst these doubtful cases there were actually some who were in the early stages of tuberculosis and others who were non-tuberculous. Therefore, in diagnosing the reaction, we will base our judgment on the types of reaction in tuber- culous individuals of the first stages and in healthy persons. After what has been explained before we will disregard the late reactions for the moment and regard the latter as the reactions of healthy individuals ; this reduces the differences between the two methods to two. Those cases in which both reactions have been obtained we will regard, duly considermg the possibility of errors, as being in the beginning of the first stage of pulmonary tuberculosis. The classification of jthe reactions does not present any further difficulties. Of the cutaneous reactions six showed traces, seven the first degree, three the second degree ; in the conjunctival method seven showed traces, three the first degree, two the second degree. In two cases the cutaneous reaction was stronger than the conjunctival. The result of the reaction in doubtful cases is the touch- stone for its diagnostic value. Above we have laid stress DIAGNOSIS OF TUBERCULOSIS 55 on the diagnostic significance, but this is not a matter of importance m cases which present hut little difficulty to the experienced chnician. As far as cases of manifest tuberculosis are concerned, the greatest advantage of the reaction lies in its prognostic value. This attribute of the reaction lias been curiously overlooked by all French writers who have introduced the reaction into clinics according to their lights, although from my very first communication to the Bed. med. Gesellsch., May 15, 1907, I have specially emphasized this point. Difficulties encountered i)i the Use of the Reaction in Doubtful Cases. — In doubtful cases the appreciation of the diagnostic value of the reaction is much more difficult than in the manifestly tuberculous cases. I cannot possibly under- stand how most writers on the subject can have come to a conclusion in the course of a few weeks. Despite the much longer time I have spent in investigating this subject, I do not regard our observation as sufficient, although during the six months of our actual investigation we found a few cases which to some extent corroborated our views. Also in these cases we had to apply the usual diagnostic methods, and, when necessary, to test the identity of the result of the sub- cutaneous tuberculin reaction with that of the cutaneous and conjunctival reactions. The fact cannot be denied that the discovery of a substi- tute for the tuberculin method — the danger of which has only recently been confirmed by a new example — would in itself constitute a great attainment, even if the new method had no other advantages ; and indeed, our observations, chronicled in the following table of Stadelmann, and which coincide with the observations of Levy and others, show that a far reaching correspondence certainly exists. Explanation of the Tables. — As is evident froui the tables, the subcutaneous tuberculin injection was commenced with O'OOl gramme of tuberculin, which was increased to 007 gramme and even 0"008 gramme if the reaction did not appear. The last-mentioned dose is too great, so that few, including ourselves, attach any diagnostic importance to it. The positive cutaneous-conjunctival reactions have in all cases corresponded with the control tuberculin injections, a fact with which most writers have agreed. The value of this cor- respondence lies in the fact that if further investigations con- firm this behaviour, the cutaneous and conjunctival reactions 56 DIAGNOSIS OF TUBERCULOSIS will, for all practical purposes, replace the subcutaneous in- jection of tuberculin. However little I may be in favour of the subcutaneous application of tuberculin, I must yet enjoin clmicians to test the concordance of the two reactions most thoroughly in order that we may be able, as soon as possible, to replace the subcutaneous tuherculin reaction by tJie local reaction. We must mention that the table shows a number of cases, clinically not suspected, in which a positive local reaction was obtained, and in which the subsequent tuberculin reaction also gave a positive result. This, therefore, according to our present knowledge, clearly indicated the existence of tuberculosis. It is of interest to notice that out of the four clinically not suspected cases in which a cutaneous and a conjunctival reaction had been obtained, two reacted on a very minute dose (0"001 gramme). Examples. — We will adduce a few histories of diseases on account of the great interest attached to them. Leipold (Pav. 6a) had sanguineous pleuritic exudation, expectoration, coughing, and sputum resembling raspberry jelly containing uninuclear cells. Clinically it was thought that tumour of the lung existed (on account of the appearance of the sputum, and because of its distinctly hsemorrhagic character). The conjunctival reaction was positive, and later on tubercle bacilli appeared in the sputum. The tubercuHn reaction was also applied, producing 38*2^ C. by a reaction on 0-003 c.cm. of tuberculin. Another case (Kirste, Pav. 17) is worth mentioning in which tuberculosis was assumed to exist on account of the general condition constituting an etiological basis for a pleuritic exudation (his father having died of pulmonary tuberculosis;. The result of both local reactions was nega- tive, which is remarkable in view of the fact that Levy in his material, which contained a good many cases of pleuritic exudation, has always observed strongly marked reactions in such cases. Nor was the suspicion as to the presence of tuberculosis supported by the clinical course of the disease ; the exudation was soon absorbed, the fever left soon after, and permanently, the sputum was examined several times and was never found to contain any tubercle bacilli, and finally it disappeared altogether. Tubercuhn injection of O'OOl gave a temperature of 37'4^ C, and it was only at the injection of 0"007 c.cm. (!) that it rose to 38'4'' C. GeneraL symptoms were observed. Eh en ^^ w ' ., , o , >2 ^a" TJ (1) 3 o c o Li a *^ o % 1% 1 '3 B 05 5 o 3 o .2 a ^ o a-2 a o *^ 0) -a a QJ o 4J 3 Q 01 a o Ph O oo o 5 - - n - .-. X .• c« ? o CM 1 1 1 1 '*^ «• 1 1 (N 1-1 1 '"' - ^ (M ■* '^ 'Jq « ■^ CO "* O <^ :j ^ t- "* cs T-H ^ ^ » 15 tb t- O) cb 1-1 CO .i c -^ o o O T-l ,—1 55 ^"l CI ■•=. •^ rt 3 ~ 91 »o cq tH »c o o CO tH .r o §1^ A CO 1 -* CO tM a ^ O 05 C -g o m A CO (M cq (M 1 o aj • : ^ u s o d 1 <-( , (A rt ►J 2 o C^l 1 1 i-H 7-1 1 1 1 ^'T^ 00 CO - c 00 ■N 9> CM E-i vi * o: cp O % ^ 00 >o op y. o '"' ■2s 2 •« o CM ;o CO to 03 CO u o rH iH T— ) 05 2 2 =3 3 3 'S a ^ ^ > a 'O 75 CO CO .2 X 05 — •^ 0) O s (M _o n 2. "" 1;^ "So &■ — ■ 3^ o (D o Si o s o c^ u: cS 3 X 3 2 ^ a -*^ -*^ ^ "*^ -^ _ 'U hH h-H -H I— 1 ^- -- O "^ -3 > O) a, > ^ = .>< o -a 3 g i-^ Ml 3'^ CO I! » o - eS I u II 85 a! ja g o -3 O '> ~ 3 c3 •-< ao o_2 o "" a 3X.2 — ° 9 "^ . 8 H II II II 5 ^O A ^ 'O H K hH H o H K 1-1 < H & 2 o s lac s gfc a 3 3 " cS op CO > 3 "o o o '5 3 ■r. O o a a a g 'O O C •^ 3 Ji . .< ^ &^ a •^ 1^ a CO o a O O ce o o '3 o a. •^ CC S H CC CO a a '3 c3 o o ^1 o 5 >! 5: 3 "3 ^^ c ■- « k— i ^ 5c ^ o o o &-I ^ ^d d C 5" a ^_q^ "S 9 a s-Too 5 CO ^ ^ > 0) o o o a> .r > HI > o3 a. > _ t*-( o &> si a 1 o o 6 9 8 o p 6 a '^ -H 3 o > O O 9 > 5) 00 CO o o _S 6 ^ C5 >*- CO < ^ < <: < < 1 5 O o o + + + + a 03 + + + + + + a X c3 11 + + + + 7. < o &H + + + ns -t- _ ^ + S + + + — ■f. ("^ + 5 S^s + + a 1 + o 'O o >- + + + + 1 c3 5^ cS •^ + + + + 5 v- ^ &H a o %4 3 .2 + + 'C + .:, "V cr " u _'b na si s f^ -g 13 o O "2 'a o & -j: 3 a a o a o o o P- CC CC a a a ® a -u '35 "3 3 ■-; o >1 a 'ci -o "p^o 3 "o U * o ._|'p. s^ o fM VH t— I E H ■"I '<" t^ > . — s ^^^ c3 r-i ^— ^ >■ < Pj, >• ^"^ > ^ ci >■ C- > C^ i: o CC CO a < CC '3 g < s ST ' — ' ■ — • a^ o a S 3 .5 — -a ■;7 1^ .a -3 a s ^1 ■^ =3 S iC ■7. =^ X &. a "^ a c (J f ) ei — >; n rr-! S» ^►-) "n a — ^ CG After the tuberculin injection dis- tinct hoarse breathing and crepi- tation over the left apex ; no sputum. The patient remains clinically suspected. No sputum. General state of health good ; but declined in sanatorium (slight catarrh, apex of right side). After repeated seai-ch tubercle ba- cilli found in sputum. Disturbed general health ; fever remained from 24—48 hours, then decreased. An old tuber- culous cavity found. Hereditary connections ; no spu- tum. Fever nearly gone ; no tubercle bacilli in the sputum. No bacilli found in sputum. After the tuberculin injection slight increase of fever during four days; no sputum ; uo symptoms of pul- monary infiltration. 5 i 3 O 1 3 EH Only reacts on 0*007,* show ing fever up to 39*2° C. On 0-001 no reaction, on 003 fever up to 38 -2" C. On 0*001 no reaction, on 0-003 fever up to 38-2° C. On 0001 after six hours fever up to 38*4° C. On 0-001 no reaction, on 0-003 fever up to 39-5° C. On O'OOl no reaction, on 0-003 fever up to 38-6° C. Only on 0-007 tuberculin fever up to 38-0° C. On 001 fever up to 390° C. z; o o Ed 03 "3 > o 3 '5° o C3 - Trace Trace + repeated trace + , trace, trace trace s § c 5 =) Q CI Trace Late reaction on fourth day + -f i./n., -f- trace Trace -f, +, +, repeated -f Trace, trace, trace, trace o 5 Catarrh, apicis. Sus- pected of tuberculosis Catarrh, apicis. Sus- pected of tuberculosis Pleurit. exud. Slightly suspected Pleurit. exud. Sus- pected of tuberculosis Suspected of pulmon- ary tuberculosis Hscmoptysis ; infiltra- tration of the left apex Tuberc. apic. sinistr. . . Pleurit. peritonit. tu- berc. 2 5 (8) Gost. (Pav. 17) . . (9) Post. (Pav. 17) . . (10) Leip. (Pav. Ga) . . (2) Brenu. (Pav. 1) . . (G) Griitzm. (Pav. Ga) (5) Chask. (Pav. Ga).. (13) Schoss. (Pav. Ga) (14) Kriig. (Pav. Ga) . . 60 THE OPHTHALMIC AND CUTANEOUS Only three days elapsed between the apphcation of each injection. Nother (Pav. 17) had a strong pleural effusion on the right side. After puncture the fever disappeared, the effusion was reabsorbed, and the patient felt quite v/ell. Clinically no suspicion of tuberculosis was entertained. The cutaneous and conjunctival reactions were very strong : — 1st i 2nd 3rd 4th , 5th days Cutaneous + + decreasing Conjunctival , + | + + + decreasing The diagnosis of pleuritis tuberculosa was supported by the result of a subcutaneous tuberculin injection 0"003 ; reaction up to 38'6' C. In the Theoretical Part I will deal with the diagnostic objections against the repetition of control tuberculin injec- tions. Since we must start from that which may be regarded as proven, I will not emphasize this point here, leaving it to everyone to draw his own conclusions from the results of the tuberculin reactions when a high dose is applied. It would, however, appear to me that the local reactions have already been shown to possess distinct advantages over the subcutaneous injection as far as their clinical diagnostic significance is concerned. I will leave it to the logical and unprejudiced clinician, Stadelmann, to pronounce a synthetic judgment on the value of the reaction. He expresses himself in the followmg words : " The justification of a suspicion of tuberculosis by the re- action can only, of course, in the first instance be found in the course of the disease and in clinical observation, which naturally has to extend over a fairly long period, so that a final opinion cannot be given easily and within a very short time. In a number of cases, however, we have actually succeeded, and I may express satisfaction at the fact that our critical and objective judgment in all cases corresponded with the result of the reaction." The Experiences of other Authors icith the Reactions. — Eppenstein has investigated seventy-six doubtful cases and has obtained positive reactions in about 40 per cent. An absolute figure cannot be given since he does not distinguish between those who reacted at once and those who reacted only after a repeated application. It is only worth while to mention six cases of chronic exudative pleuritis with positive reactions. DIAGNOSIS OF TUBERCULOSIS 61 Lenhartz, in sixty-three doubtful cases, obtained twenty- three positive cutaneous and conjunctival reactions (eighteen of which had hereditary connections) (about 80 per cent.). Schubert, in seven doubtful cases, obtained three positive conjunctival reactions (about 40 per cent.). TABLE V. CONJUKCTIVAL REACTIONS IN INTERNAL DISEASES. Author Combemale (p. 34) Charmeil (p. 36) Jouveuel (p. 37) Desbonnets (p. 38) Halipre (p. 38) Desplats (p. 39) Letulle (p. 41) Prouff (p. 42) . . Grillot (p. 43).. Grasset (p. 46) Dujon et H. (p. 46) Derscheid (p. 46) Montagnon (p. 47) Braillon (p. 48) Soulce (p. 50).. Metraux (p. 50) kl. W., 1907, 19, viii Citron (B 1907) Ferreina (p. 52) Uriarte (?) Denys (p. 53) . . Schubert (M. in. iv., ivui, j Klieneberger(3/. »(.. W., 1907 Wieus and Giinther (21. m. 1907, No. 52) IMainini (2f. in. W., 1907, No. 52) Xuniber of tuber- culous patients Reaction T7., 1907, p. 2054) "" No. 52) W., 16 1 7 1 5 12 75 5 6 8 1 7 16 6 39 15 31 2 116 8 19 17 11 12 16 1 7 1 5 9 72 4 6 14 4 34 14 25 1 109 8 17 8+2? 6 11 Suspected Healtliy 3 5 2 1 1 11 9 11 4 3 2 3 26 Total 435 391 45 76 45 8 2 19 8 1 1 72 earlier (p. 55) Le Fort (p. 50) Bary (p. 57) . . Surgical Cases. 6 6 _ 1 - i - 1 4 4 — 4 2 14 ! 12 2 3 — 1 16 8 6 28 31 8 5 14 7 10 12 IS 9 40 44 8 24 38 26 48 397 N.B. — Tkie pages indicated refer to the publication of von Pirquet. Schenck and Seiifert in twenty doubtful cases obtained six (30 per cent.) positive conjunctival reactions at the first instillation, six (30 per cent.) at the second instillation, three (15 per cent.) at the third instillation. Total 75 per cent. ( 62 THE OPHTHALMIC AND CUTANEOUS Citron obtained 80 per cent., Eppenstein 40 per cent., conjunctival reactions in doubtful cases. Further investigations are recorded in Table V., which surveys the results obtained by the other authors not men- tioned in the text. Our own investigations may be found in the tables printed a few pages previously. They contain the results of the cutaneous and conjunctival reactions, at the same time dis- tinguishing between the different stages of tuberculosis. Tbis way of tabulating has several advantages. THE REPETITION OF THE REACTION. As in all cases in which processes of hypersensibility play a part a reaction cannot be repeated without producing a decided change. Von Pirquet has found that in some cases in which the cutaneous inoculation was repeated the reaction increased in strength and became positive where first it had been negative {Wie)i. Jdui. Wochenschr., 1907, No. 38). His remarks only refer to the cutaneous reaction,^ but the hyper- sensibility is even much more pronounced in the conjunctival reactions. A great deal in regard to this matter has already been collected, and we attach to it great theoretical signifi- cance. We will only point out here that, for purposes of diagnosis, one may not repeat the reactions ad lib., since in doing so unforeseen conditions arise. I do not mean to say that repetitions should never be allowed ; the}^ are quite harmless as far as the patient is concerned and are at any rate of scientific and even (although to a very limited extent) of diagnostic value. The Significance of the Positive Result of the Conjunctival Reaction when Repeated. — As more extensively dealt with in the Theoretical Part, I am of opinion that the positive result of the reaction, when repeated, indicates the presence of latent tubercular centres. The circumstances, however, have to be taken into account in so far as the reaction occurring after repeated instillation may not be simply registered as positive and be used as such unreservedly for the diagnosis ' Compare the chapters on the application of the reaction in the diseases of children and in dermatology, in which we refer to special obser- vations of scrofuloderma-like changes in cases where the cutaneous reac- tion has been repeated. DIAGNOSIS OF TUBERCULOSIS 63 of tuberculosis. The neglect of this rule explains, for instance, why Eppenstein obtained 100 per cent, reactions in cases of tuberculosis ; he fails to tells us, however, in how many healthy individuals a similar result was obtained after re- peated instillation.^ Fritz Levy is the first in Germany who has called attention to the fact that hypersensibility results when the conjunctival instillation is repeated (quoted after Cohn, Berl. kiln. Wochenschr., 1907, who confirmed his remarks by personal observations). These observations are of importance because the repetition of the subcutaneous injection of tuberculin is to be objected to on the same theoretical grounds. This will be more fully expounded in the Theoretical Part. As may be expected the literature on the subject provides us with a few analogous remarks. Auberet and Magne {Journal de Medecine de Bordeaux, August 15, 1907) have made similar observations. Petit is of opinion that these observations detract from the value of the ophthalmic reaction and is delighted to find that Mongour and Laude (see Bulletin Medical, September 4, 1907, p. G5), in ten cases have instilled tuberculin twice, and, after enumerating these cases, he concludes that only in one case (which he attributes to an experimental mistake) a change in the reaction was observed at the second instillation. Levy, on the other hand, finds that hypersensibility results in 70 per cent, of his cases ; it seems to take at least three days to appear. Cohn has confirmed Levy's observations for twenty cases (the percentage is not given), and he beheves that there is only a difference of degree in the production of hypersensi- bility between non-tuberculous and tuberculous individuals, in so far as in tuberculous individuals the reaction after the second instillation is a very strong one. Generally speaking, I agree with him, except that, according to my own conception, non-tuberculoics individuals do not produce the hypersensibility reactions, but that it would be better to distinguish between active and latent tuberculosis than, like Cohn, between tuber- culous and non-tuberculous cases. I have embodied my reasons for this view in the Theoretical Part, and here I will * As mentioned before, bis table does not show in bow many cases be has x'epeated the reaction in chnically health}' individuals. At all events the number of positive results obtained by him in healthy persons is a very small one. 64 THE OPHTHALMIC AND CUTANEOUS confine myself to quoting Cohn himself : " In the course of life hypersensibility is acquired in the same way as the capacity for cutaneous reaction by von Pirquet's method. According to a communication by Dessauer infants never react, not even after a second and a third instillation." The Reinflammation of the Reaction Sjiots. — It is a remark- able fact that the spots where the reaction had taken place become reinfiamed when afterwards tuberculin is injected subcutaneously. This is equally true for the cutaneous as for the conjunctival reaction. A great number of our own inves- tigations support this statement.^ Petit has also mentioned a few cases which have been communicated by Slatineano and Guerin.^ {Bulletin Inst. Pasteur, August 30, 1907, and Ree. de. Med. Veterinaire d'Alfort, August, 1907). According to these reports reinflammation after subsequent subcutaneous injection may sometim.es also occur in individuals in whom in the first instance neither a cutaneous nor a conjunctival reaction had been obtained. Since this reaction of non- tciberculous persons rests absolutely on the same principles as the primary reaction occurring in tuberculous individuals, viz., on the presence of tuberculin, producing therefore a specific though artificial reaction, it seems to us unjustifiable to designate it as fausse reaction,. Such a designation would not matter if it were not for the fact that its adoption at the same time would cause false theoretical views ; as a matter of fact, enough confusion has been caused already by the designation fausse reaction. ^ Stadelmann has made the remarkable observation that the reinflam- mation of the reaction spots occurs before the general reaction, but as a rule only in such cases in which a general reaction actually takes place. In addition to the above-mentioned French authors, the reinflammation of the reaction spots after subsequent tuberculin injection has also been observed by Lenhartz {Aerztl. Verein, Hamburg, October 29, 1907, ref. Miincli. med. Wochenschr., 1907, No. 48), Bandler and Kreibich {Deutsche med. Wochenschr., 1907, No. 40), Fritz Levy {Ver. f. inn. Med., December 16, 1907), and Baginski (Berl. med. Ges., December 18, 1906). - In France, Lemaire in August last pointed to this phenomenon, and at the same time he showed that the cutaneous reaction may become positive if at the same time tuberculin is injected. Experiments on animals by Calmette, Breton and Petit showed that after great doses of subcutane- ously injected tuberculin the conjunctival reaction remained negative, whilst after small doses it became positive {Soc. de Biol., 1907, No. 28). Moro has also observed positive cutaneous reactions in scrofulous children when tuberculin had been injected subcutaneously at the same time. DIAGNOSIS OF TUBERCULOSIS 65 Hypersensibilitij of the N on- in stilled Ei/e. — Cohn has observed that in tuberculous individuals, after tuberculin has been instilled into one eye, the other eye also may become hypersensitive. The theoretical significance of this fact need not trouble us here ; we will only say that this hypersensi- bility is to be explained by the absorptive capacity of the con- junctiva. Special attention must be paid to this observation since we recommend the repetition of the instillation in the other eye if a technical mistake is suspected. For this reason the repetition on the other eye would not be alto- gether objectionable from a technical point of view ; but one should pay great attention to the correct application of the first reaction, since this causes a change in the original reactive conditions of the body, lasting a considerable time. Fortunately the observations of Cohn show that such an increase of sensibility after a first reaction only occurs in cases of actual tuberculosis, so that after all some diagnostic value is attached to the observations in the other eye. As far as I am aware, no good description of this rein- flaming of the reaction is found in the literature. For this reason I refer to Table III. on p. 58, where we have indicated the results of the reaction, as also any symptoms that have appeared after a subsequent tuberculin injection. (1) Case Sturzbach, Pav. 1, who during twenty-four hours showed a strong reinflammation of the conjunctival and cutaneous reactions, which had almost disappeared. The symptoms were more marked than in the first reaction. Dis- tinct pomphi on the skin surrounded by a strong red halo. A late increase of fever was remarkable. (2) Lohrk., Pav. 1, in whom reactions were applied on October 14 and November 9. First reaction conjunctival minus ; cutaneous late reaction on fourth day. After repeti- tion, conjunctival plus plus, cutaneous late reaction on third day. Both reactions were just traceable on November '23, on which day O'OOl gramme of tuberculin was injected. The next morning a very strong reinflammation of the concluded conjunctival reaction was observed (the second eye remained entirely untouched, which is remarkable in view of Cohn's communications). Only the second cutaneous spot showed reinflaming of the reaction. In this case, too, fever only set in tw^enty-four hours after the local symptoms on the reaction spots. (8) Altmann. The particulars of the cutaneous and con- 5 66 THE OPHTHALMIC AND CUTANEOUS junctival reactions may be seen from the table. The conjunc- tival reaction reinflamed strongly about sixteen hours after the tuberculin injection ; not so the cutaneous reaction. Again, in this case, the fever increased after about thirty hours, in fact, after the appearance of th-e conjunctival reaction. (4) Koth. Producing a reaction after the second dose of tuberculin, the conjunctival and the cutaneous reactions re- inflamed. (5) Biirg, Pavilion Ga. The result of the first cutaneous reaction was : first day, plus ; second day, trace ; third and fourth days, minus ; conjunctival : first day, trace ; second day, plus ; third and fourth days, plus. The other eye showed slight inflammation. After thirteen days the reaction was repeated ; cutaneous first to fourth day, plus ; (conjunctival instillation in the other [right] eye), first day, plus ; second day, plus, plus ; third day, plus ; fourth day, plus. The eye instilled on tbe first occasion again showed slight inflamma- tion. Three times tuberculin was injected ; only after a dose of 0*005 did fever occur the next day up to 39"(3° C, and it was only after this dose that the cutaneous reaction spot became reinflamed. This took place, therefore, before the appearance of fever, but only when a dose was administered sufficient to cause a general reaction. On the other hand, Colin has also observed the reinflaming of the reaction after tuberculin injection without a general reaction. Some theoretical value may be attached to the verification of this fact, for which reason 1 am directing attention to it. Meanwhile our own observations have given similar results in a few cases. NOMENCLATURE. This chapter will not detain us long. Von Pirquet's reaction was appropriately designated by its discoverer the cutaneous reaction, to which name we will adhere. Dermo-reaction by Rubbing Tuberculin into the Ski>i. — The reaction obtained by Lignieres and Berger {Acad, des Sciences, Paris, October 28, 1907) by rubbing tuberculin into the un- broken skin was also designated cutaneous reaction. This designation would have given rise to a great deal of confusion if it had not been for the fact that this reaction found but little favour, being without advantages and involving more work. It is, however, unjustifiable on the part of the authors DIAGNOSIS OF TUBERCULOSIS 07 to appropriate the name "von Pirquet's reaction" for their own method, indicating von Pirquet's method by the name dermo-reaction. It would have been much more appro- priate to designate the new reaction " dermo-reaction." Two months before these French authors the same discover}^ was made by Moro and Doganoff (Wien. klin. Wochenschr., No. 3, 1907). Moro has continued and elaborated his experi- ments w'ith the salve reactions {Milnch. med. Wochenschr., 1908). We have most thoroughly tested his methods, but we cannot find any advantages of this " ointment reaction " over the cutaneous reaction. Conjunctival Beciction or Ophthalmic Reaction ? — I have given the name " conjunctival reaction " to the process of instilling tuberculin into the eye. I am very much in favour of maintaining this name, not only because of its right of priority to the French designation " ophthalmic reaction," but also because this name is the more appropriate, as has been distmctlj^ shown, amongst others, by Levy {Verein fiir innere Medizin, December 16, 1907) and Schenck and Seiffert {Milnch. med. Wochenschr., November 12, 1907). The re- action takes place on the conjunctiva and not in the eye. The advantage of the reaction consists in the fact that it takes place at a sensitive, protected, and yet easily accessible spot, w^hich is not of vital importance and which does not transmit the inflammation to the eye. If the name " ophthal- mic reaction " expressed its nature in so far as it were to include the uveal tract (iris, choroid), the reaction could hardly have been recommendable. The classical method of the diagnostic application of tuberculin, the subcutaneous injection, had in future best be designated as the subcutaneous method and the tuber- culin subcutaneous reaction ; the other methods occasionally applied are pulmonary infusion, pulmonary reaction, tuber- culin inhalation, rectal reaction and the rectal method. OPINIONS OF AUTHORS ON THE CUTANEOUS AND CONJUNCTIVAL REACTIONS. The inventor of the cutaneous reaction (v. Pirquet) himself did not regard his method as appropriate for adults ; other inves- tigators were mclined to follow his example, and thus we are not able to bring forward any favourable opinions concerning the cutaneous reaction by other writers, except in so far as SCHEMES FOR EEGISTER The application of the reaction is so simple that no further aid is necessary. How- ever, if one applies the reactions rather frequently, and if one desires to have a synop- sis of one's material, one is assisted to a considerable extent by a scheme for tabulating observations. Recent -observations have been somewhat confused ; one investigator speaks of a positive reaction when he has obtained a reaction after the first instillation, another also when it is obtained after the instillation has been repeated. The results of one Date of Application : Sanatorium : SCHEME FOR THE APPLICATION OF THE By Professor Stadelmann A.— For Tuber Xame and number C'lNJCNCTIVAL REACTION Suspected STAGE I Tubercle BACILLI -f++ Spaces for tabulating any further observa Subcutaneous tuberculin injection Secondary symptoms Results of fo.st- mortem examination, or| course of the disease i Date of Application : Sanatorium : SCHEME FOR THE APPLICATION OF THE By Professor Stadelmann B. — For Non-tuber Name AND Ndmber Hereditary Connections P = Pater M = Mater Pr=Propinqui Conjunctival reaction - + -+ + + -f Days Lasting reaction ? days Spaces for tabulating any further observa Subcutaneous tuberculin injection Secondart symptoms Results of post- mortem examination, or course of the disease 1 ING OBSERVATIONS. investigator, therefore, can hardly be compared with tliose of another, and this is the reason that, for the most part, we had to rely on our own observations. In order that everyone's observations shall become of general value, we very strongly recommend the following schemes for tabulation, wliicii provide for all practical neces- sities, enabling us to at once find a reply to all questions of a statistical nature. The schemes are published (in German) by Curt Kabitzsch (A. Stuber's Verlag, Wiirzburg), and can be ordered through every medical bookseller. (Ten tables, pro- viding for 200 patients' records, for Tid.). CUTANEOUS AND CONJUNCTIVAL REACTIONS. AND Dr. Wolff-Eisner. culoiis Persons. Table No. Lasting reaction '■■ days J R EPETITKiN ClTANEOUK REACTIO.V + ++ ++ + Lasting reaction ?days Control Eye In the same eye In the Other eye - Repetition Whether liealthy before instillation Days 1 1 2 1 3 4 Days 1 1 2 1 3 4 1 1 Days 2 1 3 1 4 Days 1 1 2 1 3 4 1 1 1 1 i 1 1 1 tious that may be of interest (other concentrations, otlier tuberculin pi'eparatioiis, &c.). CUTANEOUS AND CONJUNCTIVAL REACTIONS. AND Dr. Wolff-Eisner. ciilotts Persons. Table No. Repetition is the same eye Repetition in OTHER EYE THE CUTANEOl + s Reaction + + + + 4- Lasting reaction '? days Repehtion Days Days Control eye Days Days 1 2 1 3 1 4 1 1 3 4 ' •2 3 1 4 ■i-^ 3 4 1 tions that ma\ be of interest (other concentrations, other tuberculin preparations. &c.). 70 THE OPHTHALMIC AND CUTANEOUS it has to do with children (see the chapter on the diseases of children in the Special Part). On the other hand, a great many unfavourable opinions have been expressed. Abrami and Burnet i'oand that the cutaneous reaction did not act in tlie first and second stages of pulmonary tuberculosis, without being able to explain the reason ; thus they declare it to be inconstant and irregular, without it being possible to find out why. There are tuberculous patients who do not react and, on the other hand, non-tuberculous patients who do react. The same observations have been made by A. Slatineano.^ The following opinions have been expressed concerning the conjunctival reaction : — LetuUe : The conjunctival reaction with tuberculin is a simple, certain, and harmless clinical method. The con- clusions to be drawn from positive results are of absolute value, unless the intensity of the reaction influence any conclusion. Prouff believes the scope of the new diagnostic method to be immense ; he thinks that the ophthalmic reaction will enable us to show that a whole Breton farm is attacked if once tuberculosis has gained an entrance. Grillot : " The results are evident, and prove the great value of the process." Grasset and Rimbaud : " Our first few investigations already make us declare that we uphold the cutaneous re- action, which we believe to be of the greatest service in the diagnosis of tuberculosis." Derscheid : " My experiments have been interesting, and they confirm those of Calmette." Montagnon : " The results are encouraging, the process is easy, and one can understand that the service to humanity will be immense if the reaction proves to be infallible." Braillon says he is perfectly convinced of the great practical value of the reaction. ^ Mainini {Milncli. med. Wochenschr., 1907, No. 51) is, besides Stadel- niann and myself, the only one who has applied the cutaneous reaction in adults along with the conjunctival. Feer (Miincli. med. WoclienscJir., 1908, No. 1) reconiniends a more frequent use of the cutaneous reaction, at all events in children. He starts from the mistaken assumption that the conjunctival reaction causes keratitis. This, however, is due to a special hypersensibility, the same phenomenon which has also been observed in the cutaneous reaction (see my reply in Miincli. uwd. Wochenschr., 1908, No. 2). DIAGNOSIS OF TUBERCULOSIS 71 Citron : " The oplithalinic reaction is bound to l)ecome an invaluable diagnostic method for the practitioner. Its appli- cation is easy ; dosing and taking the temperature are avoided ; it does not cause any inconvenience to the patient, not even in feverish cases, and is a certain indication of tuberculosis. The absence of the reaction, however, does not prove the absence of tuberculosis." Denijs : " The figures speak for themselves and encourage us to further apply this diagnostic method. It has the advan- tages of being simple and harmless." Bazy congratulates himself on having obtained so simple and rapid a means for diagnosing tuberculosis ; he believes that this method will play a very great part in aiding the diagnosis of tuberculosis of the urinary passages. Comby has obtained good results with the conjunctival method in 300 children (Sem. med., 1907, p. 371, in 69 cases ; and SoG. d'hop. sem. rned. inf., No. 50). He declares "the process, without reserve, to be simple, easy, certain, and harmless in healthy eyes." Sicard {Sem. med., 1907, p. 823) : " Seule I'ophthalmo- reaction est d'une application facile, et parait, dans ses resultats globaux, concorder assez bien avec les enseigne- ments de la clinique. La sous-cuti reaction et la cuti reac- tion sont, par contre, des procedes infideles et quelquefois nocifs." Francke, E. {Deutsche med. Wochenschr., 1907, No. 48) : " The reaction never failed in any case in which it was with certainty expected ; in several doubtful cases it proved to be of important diagnostic value." These opinions have been cited partly after Petit. I do not think it necessary to quote any more favourable opinions. The standpoint I represent with regard to these views is that the reaction can render extremely valuable services in the diagnosis of tuberculosis, and especially that it is of very high prognostic significance. Thus far I fully agree with all these opinions, which lately have been confirmed by Levy after considerable experience. But yet I still recommend a critical and continuous investigation of the method, because it is not hundreds but thousands of patients who must be investigated, and also because in suspected cases years must pass before the significance of the reaction can properly be judged. Our own post-mortem results speak distinctly in favour of the method in so far as the conclusions we drew 72 THE OPHTHALMIC AND CUTANEOUS from the reaction proved to be correct. So far definite opinions can only be given in the cases of healthy individuals and in tuberculosis of the first to the third stage. I here make my standpoint clear because in France and Germany a trend of thought seems to gradually come into existence which is as far beyond the mark on the one side as the original blind enthusiasm vi^as on the other. This trend of thought against the reaction justifies my original cautiousness in com- municating my first discoveries, w^hich caused Calmette and his followers to begrudge me the merit of having created this method for clinical purposes. Of these critics we might first mention Mantoux, who, at the French Congress for Internal Medicine, in October, 1907 (see Milnch. mecl. Wochenschr., 1907, No. 49), expresses the opinion that the ophthalmic reaction, whether positive or negative, should be interpreted with great reserve. He arrived at this conclusion because out of 200 children under 16, only sixteen ( = 8 per cent.) reacted positively, and of these only three could be suspected of suffering from tuberculosis, whilst ten «o;i-reacting children on clinical investigation were found to be in the first stage, according to Grancher. At the same Congress, however, Scherb (Algeria) arrived at opposite con- clusions. To him the ophthalmic reaction represents a very sensitive method of diagnosis, and is of special interest when it reacts positively in those cases in which chnically it had not been expected, or negatively where one might have expected the opposite. It seemed to him that great benefit might be derived from it in the case of infants. Barbier {Soc. des. Hop., Paris, December 6, 1907, ref. Sem. med., 1907, No. 50) has built up a criticism of the method on the observation of one patient. The latter, in whom a positive reaction was obtained, had pleuritis ; the 'post-mortem showed the cause of the pleuritis to have been a hydatid cyst of the liver. Barbier had drawn false conclu- sions as to the nature of the pleuritis from the positive reaction. Souques obtained twelve reactions in fifteen non-tuberculous individuals which partly corresponded with the reactions ob- tained after subcutaneous tuberculin injection. From this he concludes that the action of tuberculin cannot be specific. Sicard has found manifest tuberculous changes post viortem in two cases, although the reaction was negative, even after having been repeated. He recommends a certain reserve in drawing conclusions, especially in the case of adults. DIAGNOSIS OF TUBERCULOSIS 78 Massary, again, obtained very satisfactory results. He, however, obtained a positive reaction in a case of general carcinosis, and therefore concludes that the conjunctival re- action is not infallible as a method of diagnosis. He only attaches a certain value to it if, positive or negative, it corresponds with the clinical symptoms, thus supporting or detracting from a diagnosis. Mery conceives the method to be of importance, but not absolutely reliable. Salvolini represents the ophthalmic tuberculin reaction (Riform. med., 1907, No. 4'2) as a good though not absolutely reliable diagnostic method, because it may fail in distinct tuberculosis, whilst, on the other hand, it may appear in non- tuberculous cases. I have quoted all these unfavourable reports fairly ex- tensively,^ like the others of a more favourable nature ; I have omitted a great number of the latter so as to avoid any suspicion of wishing to retouch the results. The criticisms referred to find their explanation partly in the fact that they are not sutiiciently critical, partly in the somewhat naive conception that the conditions of the reaction are so simple that the result simply and without further thought indicates the presence or absence of tuberculosis. Surely tuberculosis bears a somewhat too polymorphous character ; also the reactive capacity differs too greatly in active and latent tuber- culosis. As far as tbe very complicated circumstances can, so far, be understood, we have analysed them elsewhere, and the reader will find that they are rather more than less complicated. Other objections at once indicate a lack of critical power from a clinical point of view. Thus, for instance, the objec- tion of Barbier, who did not recognize the etiological condi- tions of pleuritis, and who does not contemplate the possibility that a tuberculous centre may be present in the body, the seat of which could not, of course, be indicated by the reaction. So also Sicard, who is surprised to find tuberculosis on autopsy which had not been indicated by a positive reaction ; and Massary, who mentions a similar observation ' Some objections have been raised hy ophthalmologists {Ophth. Ges., Berlin, 5, -January 16, 1908). In the cases referred to the reactions had not been applied accordmg to my principles ; they were taken in most cases with the preparation of Calmette, or they had been applied to scrofulous patients or people suffermg with the eyes. 74 THE OPHTHALMIC AND CUTANEOUS without remembering how often it has been pointed out that dying persons do not react, and who does not stop to consider that in such cases there is no need for a diagnostic method which is only of value in the initial stages of the disease. DIFFERENT CUTANEOUS TUBERCULIN REACTIONS. In my first communication I have advocated the use of different reactions obtained with various kinds of tuberculin ; this view has found strong supporters in Detre {Biidapester kgl. vied. Ges., November 11, 1907) and Kentzler {Wie7i. klin. Wochenschr., 1908, No. 1). A difference does not exist in all cases ; in our own curve figure the different reactions take an almost equal course (see curve figure). Kentzler, in order to explain the difference produced by the use of human and bovine tubercle bacilli, has experimented with these as well as with Koch's old tuberculin. His results are tabulated as follows : — Number of cases Koch's tuberculin Human filtrate Bovine filtrate + 49 21 1 20 + - + - Cases of unsuspected tuber- 1 culosis j" Suspected tuberculosis Pulmonary ., 1st degree ■2nd ,, ,, 3rd ., 69 34 27 27 23 20 13 26 27 3 5 2 io 64 32 25 17 23 1 2 6 68 32 27 21 23 Total cases of tuberculosis . . 98 75 28 21 77 •• 98 It is worth mentioning that his experiments agree with ours, at any rate as regards the fact that he found hardly any positive reactions in the third stage of tuberculosis. He acknowledges that his investigations did not lead to any view as to a possible difference between human and bovine tuberculosis. It seems to me that this is due merely to technical mistakes, resulting from an incorrect method of application. Since the author in question himself mentions the existence of a close relationship between the different kinds of tubercle bacilli, one may assume a group reaction. For this reason full value must he attached to the observation of quantitative reaction conditions. DIAGiS'OSIS OF TUBERCULOSIS 75 This very important point, has been altogether neglected by Kentzler. If the way in which he conducted his experi- ments had been correct, the hmuan and bovine tuberculin tests together would have provided at least as many positive reactions as Koch's tuberculin alone, the latter being prepared from human tubercle bacilli. According to his tables, how- ever, this is not the case ; the reactions not only are of very much rarer occurrence, but are also much weaker. This would have been still more evident if the tuberculin and filtrate reactions had been represented by a curve chart. Thus it would seem as if Kentzler had used too weak a tuberculm, probably a kind containing too few bacteria, owing to the manner of filtration (through filter candles) (see our remarks on the prognostic significance, &c., Wolff-Eisner and Teichmann, Berl. kiln. Wochenschr., 1908, No. 2). The results of the experiments with filtered tuberculin proved the truth of this conception. When our tuberculin was filtered through Berkefeld candles fewer cutaneous reac- tions took place : that is to say, they did not occur in indi- viduals who formerly had reacted on the same tuberculin unfiltered ; if a reaction did occur it always was of far less quantitative value. We (myself and Stadehnann) have found the same in a very great number of cases of the subcutaneous tuberculin reaction. These discoveries show that the tuberculin reaction is not a fermentative one, but is absolutely quantitative, and there- fore it is imperative in the cutaneous and conjunctival reac- tions to empirically prepare a concentration which will allow the greatest diagnostic and prognostic value to he gained from the reaction. The most acceptable concentration seems to be the Euete-Enoch tuberculin obtained from the Medizinische Warenhaus A.G. at Berlin, Karlstrasse 31 ; for the cutaneous reaction one should use a 25 per cent., for the conjunctival a 1 per cent, concentration. It is, of course, possible, in a case of manifest tuberculosis in which a negative reaction has been obtained, to force the appearance of the reaction by employing a stronger solution ; but in view of our explana- tions we do not see any clinical advantage in this course of action. In assuming a difference between bovine and human tuber- culosis, these quantitative conditions have to be taken into account. If a patient reacts on active 76 THE OPHTHALMIC AND CUTANEOUS Tuberculin of bovine tubercle bacilli Tuberculin of human tubercle bacilli Solution 1 : 10 + + + 1 ; 100 + + 1 : 1000 + Solution 1 : 10 + + 1 : 100 + 1 : 1000 - one may safely conclude that the disease is due to bovine tubercle bacilli. This is the only explanation that suggests itself to me. We refer to the extensive publication on the subject by Detre, Wien. klin. Wochenschr., 1908. GENERAL REACTION IN THE LOCAL APPLICATIONS OF TUBERCULIN. The principal advantage of the local reaction lies in the fact that it does not cause a general reaction, as is the case with the subcutaneous injection. In my first communica- tion I mentioned the fact that general reactions may occur after the local application of 39 m 1 I j f"' I I I I tuberculin, but they occur so very rarel}^ and even then are so insignificant, that they can 38 I \ I ' / \! ! M I I ' I I only be observed when specially looked for. Von Pirquet, for instance, has overlooked them, 37 1 I I i ' — ^ \ \ ' /^ ^ l>-l since he neglected to take the temperature every two hours. I have added a curve figure, 3g| I I I I I I I I I I I I I I taken on May 6, 1907, which indicates a general reaction very clearly, the patient being otherwise without fever. A general reaction, in tuberculous individuals, may also appear when the local reaction pro- duces a negative result ; in fact it seems to occur preferably in this case. The theoretical insight we have now obtained on the subject makes this fact fully comprehensible. As a matter of fact, in cases of fever it is difficult, as a rule, to prove the occurrence of a general reaction, and it is in these cases that the local reaction very often remains negative. 35 1 ( i ^- it - ^ /\ 1 \ i / \ \J / 1 1 \7i \ » i 1 l\/\ ,A ! i ! y \ y \ ' , / \ / V 1 1 ! 1 , ' i 79///3S73ff/3S7Sl Reaction applied on May 6, 1907. Conjunctival reaction + Cutaneous reaction very slight. General symptoms -|- DIAGNOSIS OF TUBERCULOSIS i I This general reaction following the local application of tuberculin is so rare, and so insignificant, that no practical importance need be attached to it. After having proved its occurrence we do not attach any further importance to it. Lemaire mentions that in two cases of fever a distinct increase in the temperature was observed after the cutaneous inoculation ; also one patient, without fever, had a temperature of 39"5° C. after a cutaneous inoculation ; the same tempera- ture had prevailed a fortnight previously after a subcu- taneous injection with tubercuhn. Burnet has observed only very slight general symptoms. One or two other authors recently have mentioned the appearance of general symp- toms after the cutaneous and conjunctival reactions. This proves that my view with regard to its occurrence is correct. B.— SPECIAL PART. THE APPLICATION OF THE CUTANEOUS AND CONJUNCTIVAL REACTIONS IN THE DISEASES OF CHILDREN. Von Pirquet recognized from the very beginning that the reactions are of the utmost importance in the diseases of children, since they enable us, contrary to those in adults, to detect the moment of infection with an almost definite certainty. The basis on which this valuation of the reaction rests is the fact that newly-born infants never, without excep- tion, show a positive reaction. Prouff (see Petit, p. 75) did not even find a reaction m newly-born infants whose mothers did show a positive result (fifty-four cases, in twelve of which the mothers had relapsed). We are adding the results obtained in infants in the form of a table. Bang and Xocard, by means of autopsy, arrived at the conclusion that in the case of oxen, calves are never born in a tuberculous condition, but only become tuberculous after birth. This rule also holds good for man. Now the most important point is, that infants who do not give a reaction are so sensitive to tuberculous infection that at a very early age tuberculous infection simply means tuberculosis followed 78 THE OPHTHALMIC AND CUTANEOUS by an early death, that is to say, that at that age the mortality of tuberculosis equals 100 per cent. As the age increases the number of positive reactions increases ; and vice versa, the percentage again decreases when death through tuber- culosis is near (after Nageli, Virchoio's Archiv., vol. 160), and at the age of 18 months amounts to 29 per cent. Whilst tuberculous changes are never found j^ost mortem in the first infant stage, the number of such discoveries increases w^ith the age. Tuberculous changes discovered as matters of secondary importance point, however, to the fact, at all events in the majority of cases, that the infection did not show any tendency to spread, or even that to some extent it has healed up (see also Hamburger's observations, Wien. Min. Wochenschr., 1907, No. 36). Authors Number of infants Positive reaction Nega- tive reaction Observations con- cerning positive reactions Other observations Prouff and 54 54 The mothers of 12 Petit (p. 75) of these infants had reacted Czerny ? ? — — Baginskv 9 ? — — Sperk..' 109 1 108 — Died in 21st month Langstein (B. 100 1 99 Tuberculous post — kl. Tr.,1907, mortem No. 44) Aronade {Med. 46 1 45 Hereditarily con- Of those reacting &L, 1907, No. nected and dis- negatively, it was 51) tinctly tuber- culous found postmortem that 17 were from tuberculosis Engel aud 1 See text . . Out of 100 children Bauer (B. A- Z. 22 reacted Tr.,1907, No. 37) The frequency of tuberculosis in children increases with their age, especially between the first and third years. In the first six months of life tuberculosis is never found as a secondary matter at autopsy, whilst at the age of puberty the extent to which it is found is expressed by 53 per cent. The mortality curve of tuberculosis of course takes an inverted course, giving 100 per cent, for the first half year of life, descending to 50 per cent, at puberty, and showing tuberculous changes. Healed up cases of tuberculosis have not been known to occur before the age of 3 to 4 years, and become more and more frequent as the age increases. DIAGNOSIS OF TUBERCULOSIS 79 Von Pirquet expresses his view with regard to the dia- gnostic value of the cutaneous reaction as follows {Wien. klin. Wochenschr.. 1907, No. 38) : " The positive result of Months OS 3-G 6-12 1 2 ?-4 h-a CIO lO-U over 7i / / / so% r' ^ ^ V / / fO% 0% / Frequency of tuberculosis in 988 cases in infants and children of different agesi according to von Pirquet. 0-3 3-6 e-is 1-2 2-U k-6 G-10 10-lU over l!f 70% ^ V so% / 1 jj%/ 30% zo% I3^S~^ -^ 1 0'... S'i __^ Frequency of latent tuberculosis according to von Pirquet. the reaction indicates with certainty the presence of active or inactive tuberculous changes " (thirtj^-one confirmed by autopsy, one doubtful case). The work of Eabinowitsch (Berl. Idin. Wochenschr., 1907, No. '2), which is provided 80 THE OPHTHALMIC AND CUTANEOUS with a complete list of the literature on the subject and with the cases, proves that one may not conclude as to the non- existence of tuberculosis merely on the ground of the signs found at autopsy. Organs which seem to be healthy macro- scopically may be found to contain tubercle bacilli on micro- scopical investigation — the first work on this subject was published by Orth in 1876 — and even organs appearing microscopically normal may be found to contain the bacilli in the animal experiment. Ages in Years. 9 -a r,-70 10 It 1^ 13 I'j- Frequency of active and of latent tuberculosis in children. -' Tuberculosis on the basis -| \- Latent tuberculosis found of Hamburger's post- on autopsy. mortem results. - o On the basis of the cuta- o - o After cutaneous inocula- neous method. tion. Curves according to von Pirquet. In the case of active tuberculosis the reaction may dis- appear during the last days of life. Ebstein, Ganghofner, Bayer, Escherich, Schick, Schloss- mann and Binswanger^ have come to the same conclusions with regard to the subcutaneous tuberculin injection ; the cutaneous reaction, therefore, may be regarded as a substitute for the subcutaneous tuberculin reaction. This is of special • Arcli. f. Kinderhcilk., 1906, vol. 43, p. 121. DIAGNOSIS OF TUBERCULOSIS 81 importance in the case of children, because Baginsky and Neumann {Berl. klin. Woclienschr., 1907, No. 9) and others altogether disparage the subcutaneous injection of tuberculin. I add a table (p. 80) showing the results of the conjunctival reaction in children ; these results speak very much in favour of the reaction. Engel and Bauer {Berl. klin. Woclienschr., 1907, No. 37) fully agree with these statistics as far as children of a certain age are concerned ; they differ, however, in the case of infants, because in one case, after having obtained a positive reaction, they did not find any tuberculous changes on autopsy, and in four other cases with positive reactions they could not clinically find any symptoms indicating tuberculosis, nor did they obtain a reaction by means of the subcutaneous injection of tuberculin. In the case of infants, therefore, they prefer the subcutaneous tuberculin injection. Diseases of Children. Xumbpr of tuberculous children Reaction Suspecterl H-althy + + 3 6 5 2 + Delearde (p. 59).. Ausset (p. 52) . . Mongour (p. 64) Comby (p. 68) . . 9 2 1 2 9 2 1 2 - 1 5 1 14 6 7 4 47 20 13 8 Surgical Cases. Gaudier (p. 70) . . 4 4 _ 4 1 9 ^louton . . 25 22 3 3 1 4 8 Bosc 36 35 1 10 7 - 10 Moro-Doganoff and Schleissner have also obtained favourable results with the cutaneous reaction. They agree with Pirquet in their statements concerning the occasional absence of the reaction in cachectic children and in the final stage of miliary tuberculosis and of meningitis tuberculosa. Schleissner prefers this method to the subcutaneous tuberculin reaction, on account of its harmlessness. Occa- sionally a very high degree of hyper-sensibility to tuberculin is found to occur in children, especially in scrofulous children, and I will here mention the cases that have come to my know- ledge. In adults, such a high degree of hyper-sensibility is apparently found only in cases of local tuberculosis of the skin. Pfaundler {Milnch. Ges.f. Kinderheilk., June 14, 1907) and Moro-Doganoff {loc. cit.) mention the case of a four-year-old 6 82 THE OPHTHALMIC AND CUTANEOUS child in which the reaction was apphed on account of fever after measles, and which had an outburst of scrofulides lasting twelve days, a fortnight after the reaction. Another child showed a strong cutaneous reaction, with the formation of vesicles, after the application of undiluted tubercuh'n. After eleven days a conjunctivitis phlyctcienulosa appeared on both eyes, lasting seventeen days ; simultaneously a polymorphous eruption not unlike lichen scrofulosum made its appearance. The same thmg was found in the case of a nine-year-old child suspected of pulmonary tuberculosis, fifteen days after a cutaneous inoculation with 25 per cent, tuberculin.^ So also a two-year-old child with coxitis and morbilli showed conjunctivitis phlyctsenulosa a fortnight after the cutaneous reaction ; another two-year old child, ten days after a strong cutaneous reaction, had a re-inflammation of a past conjunctivitis phlyctcenulosa and of two scrofulo-dermata. Also Engel and Bauer {Berl. klin. Woclienschr., 1907, No. 37) have noticed strong reactions in scrofulous patients ; sometimes they even observed tubercle-like formations after cutaneous inoculations (see the following chapter). THE APPLICATION OF THE REACTIONS IN DERMATOLOGY. Dermatologists have mostly used the cutaneous inoculation, which obviously is more natural to them than the conjunctival reaction. As far as my knowledge goes, dermatologists have never published any results obtained with the latter. The results obtained, however, concur with our own observations. Bandler and Kreibich have obtained thirty-four positive re- actions out of thirty-eight cases of local cutaneous tubercu- losis, whilst in four cases which were complicated by severe ^ This appearance of phlyctsenae not merely after the conjunctival but also after the cutaneous tuberculin reaction induces me to refer in a few words to the views as to the origin of keratitis held by ophthalmologists. The pui-ely exogenous staphylococci theory (Burchard and Eeich) we may certainly regard as non-existent. It is generally agreed that endogenous stimuli play a part. Leber and Bruns obtained phlyctena-like formations by injecting dead tubercle bacilli. But the identity of these formations with the real phlyetenie is disputed. Staub assumes the existence of a toxin irritation, which should have to be supplemented by an exterior irritation. See Axenfeld, " Bakteriologie des Auges," p. 232. G. Fischer, Jena. DIAGNOSIS OF TUBERCULOSIS 83 pulmonary tuberculosis the reaction was negative, in two cases also after subcutaneous injection of tuberculin.' The comnumications of all authors seem to point to the fact that very strong reactions may always be expected in cases of lupus and skin tuberculosis ; this corresponds to the well-known hyper-sensibility of the skin of scrofulous people with regard to super-infection (Bandler and Kreibicb, Pfaundier, Moro, Doganoff, Oppenheim, Uffenheimer, and others). In a similar sense we have to interpret the fact communicated by Nagelschmidt and which we have mentioned before, namely, that local inoculation with old tuberculin causes an ulcer in a tuberculous centre in the skin.- The views concerning the pathogenesis of a number of tubercular affections of the skin, and especially of scrofulides and their significance, are in a continually vacillating con- dition. Oppenheim (" On Changes in the Skin in Adults, following the Reaction of Pirquet," Wien. klin. Wochenschr., 1907, No. 32) doubts whether tubercle bacilli are at all present in the skin in a number of tuberculous processes, and comes to the conclusion that tubercle bacilli at any rate do not occur often. Starting with this view, he carried out experiments in order to find out whether, by repeated appli- cation of tuberculin, changes analogous to those of scrofulide could be obtained. He actually succeeded, by repeated appli- cation of tuberculin to the same spot in the skin, in obtain- ing papulous efflorescence of a soft consistency, which finally became brownish-red and flabby, and surrounded with nodules the size of a millet seed, bearing the character of lichen scrofu- losum. In another case of lupus the pomphi changed into vesicles, these latter into sharp-edged ulcers which, notwith- standing treatment, w^ere still present after a fortnight. He regards these observations as somewhat analogous to those in super-infection in syphilis by Finger and Landsteiner. Pfaundier, Moro, and others have made similar observa- tions, which we have recorded in our chapter on the reaction in diseases of children. ' Out of sixty-two noa-tuberculous skin cases thirty-eight gave reactions, which, however, showed distinct variations when compared with those obtained in lupus patients. - Nobl {Wieti. Jdin. WochrnHchr.. 1908, No. 1. p. 27) describes a reaction lasting forty-eight hours in a scrofulide of an epithelioid tuber- cular character, showing an erythematous infiltration, 25 niui. in breadth, with a central vesicular elevation the size of a bean. Inoculation in a scrofulide caused a papule the size of a bean. 84 THE OPHTHALMIC AND CUTANEOUS Our observations enable us to offer a solution to a number of problems still unsolved in dermatology, and therefore we will give a short sketch of the present state of our know- ledge with regard to tuberculosis of the skin, and especially of scrofulides. Lately a great deal of progress has been made in diagnosis (subsequent to the Finsen therapy), and also in the construc- tion of clinical symptomatology. The progress made in a diagnostic direction has gone hand in hand with the use of subcutaneous tuberculin injections, which were especially used by Neisser, Jadassohn, Klingmiiller and Buschke. The subcutaneous tuberculin injection causes a central reaction in the case of tuberculosis of the skin, which is somewhat analogous to the cutaneous reaction of Pirquet. The local reaction after the subcutaneous employment of tuberculin is most distinct in the case of lupus ; it is most frequently absent in the case of tuberculosis verrucosa cutis of infants. Besides lupus proper, the nature of which may be regarded as understood, there are other forms of cutaneous tubercu- losis which must be mentioned. We may feel certain that their nature will soon become more clear to us by the frequent application of the local reactions ; the work done by derma- tologists with the cutaneous reaction certainly embodies a promise of this nature. First we find a description of lupus pernio by Jarisch, Kreibich, and Polland, which is signified by infiltrations of a bluish- red to faint red colour, and which is probably only lupus produced by a disturbance of the circulation. Then there is lupus follicularis disseminatus, a cutaneous manifesta- tion, partaking at the same time of the nature of miliary tuberculosis. This form has a subdivision, a kind which only occurs in the face, and which has a more favourable prognosis. The products of the disease are of a tubercle-like structure found by Schassberg, Bettmann, and Finger, together with tubercle bacilli. Jadassohn showed that the diseased parts are pathogenetic with regard to animals (guinea-pigs), and that after they have been removed by an operation there are recurrences of the disease ; Delbanco, on the other hand, believes in the existence of toxin of tubercle bacilli secreted by the sebaceous glands. All lupus-like skin diseases may be defined as diseases which have a tendency to destroy the skin, forming scars DIAGNOSIS OF TUBERCULOSIS 85 as they progress, showing remissions, hut which never heal spontaneously. Quite contrary to these we find clinically the scrofulides, consisting of hnematogenous dermatosis, which appear suddenly, and are frequently symmetrical ; they have a more benign chai'acter, showing but little inclination to ulcerate and to form scars. ^ They are found most frequently in children who have tuberculosis in other organs.- Accord- ing to Werther (Ges. f. Natur. und Heilk., Dresden, ref. Miuich. 7ned. Woehenschr., 1907), diagnostic significance is to be attached to these comparatively benign formations, which in a pathological sense are to be regarded as lympho- mata, indicating tuberculous formations in other parts of the body. To this group belong folliculitis, lichen scrofulosum, erythema induratum scrofulosum (Bazin, Hebra), in which hard nodules appear in the depth of the corium, the skin over this part having a bluish-red hue. iVlthough this forma- tion shows little inclination to regressive metamorphosis, yet Hutchinson has observed such a process.^ I will briefly refer here to three forms of disease which, from a pathological point of view, have a tuberculosis-like appearance, and which often have given a positive result with the subcutaneous tuberculin reaction, although it is still uncertain whether they have to be classed with the group of tuberculous skin diseases. I am referring to lupus erythema- todes, pityriasis rubra, and angio-keratoma Mibelli. I must mention that Bandler and Kreibich have obtained in one case of lupus erythematodes, by means of the cutaneous reaction, a distinct, and in another case a weak reaction, and that lichen scrofulosum and lichen syphiliticus could not be distinguished by means of the reaction, probably on account of the fact that both affections are of a scrofulous nature. ' It may be of interest to mention in this connection that L. Dor, jun. (XVIII. Congr. de la Soc. Frang. d'Ophth., 1906), regards the frequently mentioned phlyctaense, as also episcleritis and a few tubercles on the choroid, as scrofulides, at the same time distinguishing between tubercu- losis and paratuberculosis. Dr. Emil Levi, of Stuttgart, to whom I am indebted for this communication, does not know what the material demonstrating these distinctions consists of. - See also Stock, " Tuberculosis as Etiology of Chronic Inflammations," Grafe's Archiv, vol. 66, i. ' The scrofulide question at present occupies the centre of dermato- logical discussion. It monopolized nearly the whole time of the Convention of the Wiener dermatolog. Ges. on November 6, 1907 (see Wien. Min. Wochenschr., 1908, No. 1) 86 THE OPHTHALMIC AND CUTAKEOUS 111 order to decide the question concerning the nature of the lichen and the hipus erythematodes, widespread investi- gations have to be carried out in view of the frequency of tuberculosis. In most of the affections referred to, tubercle bacilli have been found, and reaction results have been obtained. Bacilli, however, are found so rarely in these cutaneous tuberculous affections that almost every case has been published. For instance, Jacobi and AVolff have found tubercle bacilli in lichen scrofulosum, and Jacobi, Pellizari, and Haushalter have obtained positive animal tests in this disease. Philipson and Macleo, and Ormsby have found tubercle bacilli in the case of folliculitis ; Thibierge and Eavaut, Colcott Fox, and Carle Lyon have observed animal pathogenesis in erythema induratum. These discoveries, however, are so isolated that the tubercle toxin theory may be maintained, which declares that the skin affections in question are caused by the secretions of toxins which the tubercle bacilli have secreted in another part of the body. The principal exponents of this theory are Delbanco, Hallopeau, and others. Finally, Klingmiiller thought to have definitely settled the question when he obtained the same signs in people who were suffering with such affections of the skin by means of tuberculin passed through Berkefeld candles. It is possible that our investigations may solve the question as far as the toxin theory is concerned, especially if one considers the fact that tubercle-like changes in the skin can be obtained, not once but repeatedly by employing, tuberculin. But after the results obtained in our experiments we are of the same opinion as Jadassohn in replying to the experiments of Klingmiiller, viz., that tuberculin, even after having been filtered through Berkefeld candles, may still contain particles of tubercle bacilli, perhaps of ultra-microscopic size. From our point of view there are no essential differences between living and dead tubercle bacilli ; and those differences that undoubtedly do exist find their explanation in the fact that the living tubercle bacilli propagate, and thus exercise a quantitatively stronger action of longer duration. In the case of scrofulides the most probable explanation is that tubercle bacilli are drawn haemotogenously from another tuberculous centre of the body, and undergo bacteriolysis in the skin. This view, w"hich is in harmony with the action exercised by tuberculin on the skin of tuberculous individuals, explains at DIAGNOSIS OF TUHEKCULOSIS 87 the same time the construction of tlie cutaneous reaction witli its histological changes reminding us of tuberculosis, and the fact that living bacteria, or those capable of life, are, com- paratively speaking, found so rarely in these centres of the disease. It is interesting to find that the Finsen therapy is no longer regarded as acting specifically, but — corre- sponding with our explanations concerning the treatment of tuberculosis — is looked upon as an inflammation which is slowly progressing from the surface to the interior, and which naturally supports and increases the natural healing tendency, the elective character of which is based on the processes of sclerosis which are already preparing in the diseased tissue centres. (Jadassohn.) The variegated nature of tuberculous changes in the skin causes Alexander (who, by the way, has supplied us with an excellent collective referendum on scrofulides in the Berl. klin. Wochenschr., 1907, Nos. 11-18) to exclaim that, as far as tuberculosis of the skin is concerned, dermatologists continu- ally meet with inexplicable symptoms, not only in regard to localization, but also with reference to the different ways in which the skin reacts when the same virus is implanted. The employment of the local tuberculin reactions may do away with these problems ; in a diagnostic sense, as has already been fully explained ; and in a theoretical sense, b}^ showing us that the reaction in the body does not take an arbitrary course, but follows certain laws of hypersensibility. which we have elucidated in another part of this work, and which only need be applied by expert knowledge to the mani- festation of tuberculosis of the skin, especially to that of scrofulide, in order to do away at once with practicall}' all problems. (Added during revision.) Pick has expressed himself in a similar way at a meeting of the Yerein fur innere Medizin on January 20, 1908. THE APPLICATION OF THE REACTION IN OPHTHALMOLOGY. The cutaneous reaction has not yet been employed very much in ophthalmology, although its employment here is very much to be recommended, for reasons to be dealt with further on. The conjunctival reaction, on the other hand, has been extensively used. Calmette has mentioned affections of the 88 THE OPHTHALMIC AND CUTANEOUS conjunctiva, . especially conjunctividse, as contraindications. In another part of this book we have already mentioned that we do not recognize these contraindications, since we have never in these affections been able to observe any undesirable secondary symptoms caused by the use of the conjunctival reaction. Calmette, however, recommends the free use of the reaction in affections of the interior of the eye, also in the case of tuberculosis, in view of the good service it may render to the ophthalmologist. And, indeed, the method has already been applied in a good many cases. Petit has collected the different results obtained by this method, and he distinctly says that it has never produced any harmful results. In fact Painblan and Brunetiere maintain that the local instillation of tuberculin in the eye makes us expect therapeutic effects. Stephenson, by means of this method, diagnosed the existence of tuberculous centres in cases of eye disease, and praises the reaction for enabling us to ascertain the tuberculous nature of affections of the eye (iridocyclitis, scleritis, chorioiditis) the etiology of which would otherwise have been somewhat ob- scure. Furthermore, Aubaret and Lafon have had favourable results in isolated tubercles of the choroid, in enchymatous keratitis in suspected affections of the nasal lachrymal ducts, in kerato-conjunctivitis phlyctsenulosa, in interstitial kera- titis, and in chronic iridocyclitis. In two of Stephenson's cases the results of the reaction corresponded with the results m cases of subcutaneous tuberculin injection. Painblan's positive results in cases of conjunctival tuberculosis and of phlyctaBnge are worth mentioning. Morax {Soc. des Hop., December 6, 1907) is of opinion that if tuberculin is instilled, if there is conjunctival tuberculosis, graver symptoms maij occur ; but in the case of uveal tuberculosis this method, according to the unanimous opinion of ophthalmologists, is quite harmless. In our chapter on Contra-Indications we had occasion to mention a few cases in which the employment of the con- junctival reaction has given rise to unfavourable and even grave results. I will refer here to this part and adduce a case of Barbier (Soc. des Hop., Paris, December 6, 1907) in which the instillation of a drop of tuberculin in keratitis led to a grave affection on both sides, resulting in ulceration and the loss of one eye. With reference to cases like these I strongly advise not to employ the instillation in the ordinary way in cases of affections of the interior of the eye, especially if there is a possibility of a tuberculous affection of the uvea. DIAGNOSIS OF TUBERCULOSIS B9 One should begin with very weak solutions, 1 in 100,000 and weaker, and repeat the reaction with solutions gradually increasing in strength. I am not contradicting myself by having recommended caution in diagnostically judging re- peated instillations, since I am not of quite the same opinion as Levy, who altogether rejects the value of repeated instilla- tions for diagnostic purposes. French authors point to the therapeutic action of the local instillation of tuberculin ; there is more likelihood of obtaining therapeutic results if these measures of caution are taken, and grave results may be regarded as being excluded.^ It would be very desirable if this method were commonly adopted in ophthalmology in the manner I have indicated : for it would be deplorable if this were only done after having caused damage to this most sensitive organ in several people. At the same time I believe that the proper application in ophthalmology of the conjunctival reaction has also a dia- gnostic use, since the views concerning the frequency of tuberculosis of the eye are still greatly confused. According to Groenow (" Grafe-Siimisch," 2nd ed., vol. ii., part 1, p. 67*2), one case of tuberculosis of the eye occurs in 4,600 ophthal- mic patients. According to Horner it is one in 4,000, and he specially refers to tuberculosis of the iris ; according to Stephenson it is one in 1,500, and it is asserted that con- junctival tuberculosis is the most frequent in occurrence. Helborn (Berl. klin. Wochenschr., 1907, p. 898) finds one case of tuberculosis of the eye in every 200 ophthalmic patients, and believes that it occurs even more frequently. Such differences cannot be explained by the difference in the material. Even non-ophthalmologists can see that great difficulties are attached to the proper diagnosis of tuber- culosis of the eye, and that a great deal may yet be expected from our method. It is impossible to say offhand whether in the case of conjunctival tuberculosis the reaction may be applied in the ordinary way. It is true that there is less danger in such a case than there would be in one of uveal tuberculosis, but the transmission of the disease to the cornea is not always ' (Added during revision.) In the Paris letter of the Berl. Jdin. Wochcu- scJir., 1908, No. 2, it is mentioned that Lapersonne warns against the em- ployment of the conjunctival reaction if the eye has not been examined beforehand and found to be free from tuberculosis. He therefore represents the same views expressed by us in the chapter on Contx'aindications. 90 THE OPHTHALMIC AND CUTANEOUS excluded. It is therefore best in such cases to apply the reaction cautiously. (Added during revision.) During the meeting of the Berlin Ophthalmolog. Soc. of January 16, 1908, several speakers drew attention to the fact that the reactions are always very strong in the case of follicular catarrh and trachoma. Since the action of the tuberculin causes an accumulation of lymphocytes, we can understand why the reactions must be a strong in diseases in which there is already hj'pertrophied lymphatic tissue. For this reason it is desirable to use caution in applying the reaction in such cases. THE APPLICATION OF THE REACTION IN GYNyECOLOCrY. It is generally maintained that tuberculosis in specific g5^n8ecological diseases is comparatively rare. Of course tuberculosis in other parts of the body will be found as fre- quently as usual, but this only in a secondary sense, requiring no further consideration when dealing with diseases of women. In the Berlin University Hospital for Women, Bumm and Martin have conducted experiments with a view of determin- ing the influence on the course of tuberculosis in artificial premature labour. They confirm the old maxim that preg- nancy and the secretion of milk (lactation) intensify the disease. For all other than vital cases (oedema of the lung) their advice is^ not to cause premature birth after the seventh month of pregnancy. Sterilization is frequently combined with artificial premature labour in order to take advantage, in the struggle against tuberculosis, of the change which takes place in the metabolism and which takes the form of an increase in weight. Now it is remarkable that, in cases of pregnancy in which the conjunctival reaction had produced a negative result on account of the very unfavourable physical condition of the individual concerned, the reaction made its appearance after the operation had been performed : that is to say. after the physical condition had imjiroved. This fact constitutes the first authoritative confirmation of the correctness of our ' See Bumm, Demonstr. in Ver. f. inn. Med. of Berlin, February 24, 1908: Cent): f. Gyn., 1908, vol. iii., p. 75; Zeitschr. f. Gebiirtsh.. vol. xli., p. 428. DIA(4N0SIS OF TUBERCULOSIS Ul conception concerning the prognostic significance of the re- action. It is of special vakie, because in this case it is possible to experimentally change the physical condition, which in all other cases can only be tried along therapeutic lines, though often unsuccessfully. Landau, at a meeting of the Berl. mediz. Gesellschaft on May 8, 1907, has pointed to the fact that tuberculosis — some- times it is true, in an early stage— was found in about 18 per cent, of the preparations obtained after operation. If we are assured from so competent a side that genital tuberculosis is of such frequent occurrence, greater significance may be attached to the local reactions even in the case of gynajcology. Eecently Birnbaum, in an extensive monograph, has shown the importance of the subcutaneous tuberculin injection : his arguments would also hold good in regard to the application of the local reactions in gynaecology. THE APPLICATION OF THE REACTIONS IN PSYCHIATRY. The reactions have also found a wide field of application in psychiatry, especially in France, and a series of interesting results have been obtained. We will not stop to mention any facts which are m harmony with the chnical observations dealt with above, and which, therefore, are not of special interest here. Lepine and Charpenel arrive at the conclusion that the local reactions in the case of dementia praecox indicate the presence of tuberculosis to the same extent as the one shown to exist by former investigators who worked with other methods. Eaviart welcomes these new methods of diagnosing in psychiatry, because the}^ facilitate the diagnosis of tuber- culosis, so difficult in the demented, and also because it has been found that tuberculosis plays some part in the etiology of certain forms of dementia. He has also observed that in the case of children, as well as in the case of adults, the number of cases of tuberculosis increases with the length of their stay in the asjdum. In the case of demented children this increase corresponds with that in normal children, increas- ing with the age. In adults, however, the increase is a very great one. Raviart has obtained reactions in dementia in 74 per cent., in idiocy in 64 per cent., and in dementia praecox in 60 per cent. These figures are remarkable in view of the fact that out of 628 demented adults only 43 per cent. 92 THE OPHTHALMIC AND CUTANEOUS gave a conjunctival reaction. fOut of 66 demented children, 64 per cent, reacted ; out of 38 idiotic children, 28, which is practically the same percentage.) Period of Residence. Tears 80 70 60 SO hO 30 ZO 10 0^0 0-1 1-5 5-10 10-20 ^*^ — T 1 ^ \ \ 1 / \ / \ 1 / 1 / I / \ / \ / \ W \ \ Is ^ ^ y Period of Residence. Years liO % 50 "/o 't0% ■30% ZO^/o W/o 0°o 0-1 1-5 D-10 10-20 More than 20 V \ \ ^ / ^ y' ^ -MJ / ^^ / / + / In children. In adults. Increase of tuberculosis during stay in a lunatic asylum according to Raviart. Positive Conjunctival Reactions per cent. Negative Conjunctival Reactions per cent. THE APPLICATION OF THE REACTIONS IN VETERINARY SCIENCE AND THE RESULTS OBTAINED IN ANIMALS WITH THE REACTIONS. The reactions have often been applied in animals, but here the results were of a much less distinct character than in man. A short time after my communication with regard to the conjunctival reaction, Vallee {Ac. des Sciences, June 17, 1907) obtained positive results with the conjunctival reaction in the case of tuberculous oxen and goats which had taken the infection in a natural way. His reactions were a little too strong with a 10 per cent, tuberculin solution, and he therefore recommends the use of weaker solutions. The results obtamed by Vallee are confirmed by Moussu. Arloing opened an extensive controversy in reference to the results DIAGNOSIS OF TUBERCULOSIS 93 obtained by Vallee. He had investigated oxen, goats, dogs, rabbits, and guinea-pigs which had reacted on the sub- cutaneous tuberculin injection. He could not determine the occurrence of a specific cutaneous reaction, and he obtained similar results when trying other kinds of tuberculin. Vallee replied that Arloing had in twelve out of nineteen cases made use of other species of animals than he, and that the reaction was positive only in naturally infected, and not in experi- mentall}^ infected animals. This reply, however, does not by any means settle the question, since there is no reason why the reaction should be different in artificially inoculated animals and in cases of natural infection. There is more importance, however, in his reference to the technical difference in the manner the experiments have been conducted. Arloing inoculated very superficially, whereas Vallee was very careful, trying to obtain a slight suggillation. Lignieres, again, has not met with one failure in the 200 tuberculous oxen in which he has applied the cutaneous and conjunctival reactions ; in health}^ animals it was never posi- tive. The conjunctival reaction was the more sensitive (Sem. mecl, 1907, No. 40). The other communications found in the literature are also contradictory. Levy (Ver. f. innere Medizin, Berlin, Decem- ber 16, 1907) did not obtain any conjunctival reactions in tuberculous dogs. Arloing, in four artificially infected dogs {Soc. d. Biol., November 28, 1907), obtained in one case only a slight reaction ; also only in one out of four calves. However, these were merely feeding experiments, and in most cases no macroscopic tuberculous changes were to be found post mortem. An increase in the temperature after the sub- cutaneous injection of tuberculin cannot, for our purpose, be regarded as sufficient proof of any existing tuberculous infection. Nobecourt and Mantoux have obtained positive conjunc- tival reactions in rabbits if there was slight infection with tuberculosis ; negative reactions in severe cases after in- travenous infection and after introducing tubercle bacilli into the stomach. They never obtained a positive cutaneous reaction in rabbits, though Lignieres speaks of having obtained such reactions. Also in the case of guinea-pigs the conditions seem to be rather complicated. Joannovics and Kapsammer were 94 THE OPHTHALMIC AND CUTANEOUS never able to detect definite tuberculous affections by means of the cutaneous and conjunctival reactions in a great number of »uinea-pigs. Out of forty animals under observation only two gave a slight reaction. Slatineau obtained very distinct reactions in tuberculous guinea-pigs. Bournet must also have obtained positive results in the case of guinea-pigs, since he speaks of having investigated pieces of the skin of guinea-pigs vs^here a reaction had occurred. The change, however, did not seem to him to have been specific, as the affection gave him the impression of a burn, and also because he found a great accumulation of leucocytes around an eschar formation. Levy and myself, again, have obtained negative results in guinea-pigs ; whereas Klemperer had positive re- actions (Ver. f. innere Medizin, Berlin, January (5, 1908). Professor Klemperer himself showed me one case of a guinea- pig, infected by means of sputum about eight days previously, which gave a distinct, although not very intense, reaction. In a few other cases, again, in which he had himself per- formed the inoculation, the reaction was a negative one. According to the observations of Klemperer, at least three days must elapse after the infection before the reaction can be judged. He has not yet found out at what later time it disappears. To judge by the material at hand, the reaction in the »uinea-pig seems to be somewhat vacillating, and often, it seems, there is a rapid reaction (see under that heading). If it could be definitely confirmed that in the guinea-pig a reaction can always be obtained in the first stages, it would pj-QYe — and this would be corroborated by other observations — that also in the guinea-pig we have a certain, though not a hicrh, degree of reactive capacity ; that is to say, an impulse resisting the infection with tuberculosis. This is also evident from the fact that the period between the infection and the death of the animal varies from six weeks to nine months, according to the virulence of the infection. But this does not in any way impair the conclusions drawn by us from the insensibility of the guinea-pig to tuberculin, since in the case of subcutaneous injection it has been practically estab- lished that the non-infected guinea-pig may be regarded as insensible to the action of tuberculin. The indefiniteness of the observations concerning the appearance of the reactions in animals is very deplorable, because it prevents the practical application of the local reac- DIAGNOSIS OF TUBEliCULOSIS 95 tions in veterinary science, though it is probable that also in the case of animals we may finally arrive at uniform results governed by biological laws. It is not impossible that animals which seem to be extremely sensitive to infection, such as the guinea-pig, present other conditions of reaction than those found in man. It is, however, very improbable that in animals such as the dog and the ox — which, as far as their sensitiveness with regard to tuberculosis is concerned, are not very distant from man — a reaction should not be obtained. From observation in man we know how complicated the course of the reaction may be. In this case we may learn from what takes place in man (about whom we now have definite information to utilize) what happens in the case of animals. The experiments in animals carried out by Calmette, as also our own observations in man, show how well the con- clusions we drew from the apparent unreliability of the reaction in man have been justified, and have been corrobo- rated by statements by the writers on the subject. In the case of animals it must also be taken into account that the thick skin greatly impedes the process of the reaction. On several occasions we have given as our opinion that the vascular arrangement and the capacity of absorption play a very great part in the production of the reaction. It would be necessary, therefore, in the case of dogs and oxen, to introduce the tuberculin into much deeper layers than in the case of man : that is to say, to penetrate deeper into the dermis. Klemperer also calls attention to the fact that it is necessary for the penetration of the tuberculin to thoroughly cleanse the skin by means of alcohol and ether in order to remove the sebaceous matter, and that it would also be of advantage to press a wad of cotton-wool soaked in tuberculin on to the wound, and to leave it there for some little tnue. According to Klemperer's directions it would then be necessary to thoroughly shave the reaction spot, and, in the case of guinea-pigs, to perforin the reaction on such spots as are covered with white hairs, since the pigment present at dark spots would impede the observation of the reaction. Finally, we should make mention of the animal experi- ments carried out by Calmette, Breton, and Petit. The first points out that by introducing tuberculin into the peritoneum the relation of the opsonic index to the normal may increase or decrease, according to whether a small or large amount 96 THE OPHTHALMIC AND CUTANEOUS of tuberculin is inoculated (normal index 0'07 ; after 1 milli- gramme of tuberculin, 0"38 to 0'52 ; after large doses [10 to 50 milligrammes], 0'05 to 0"03). Further investigations show that in rabbits, after injecting tuberculin into a vein, a distinct conjunctival reaction is obtained after sixteen hours if the amount injected has been small (2 to 10 milligrammes), and no reaction at all when large masses were introduced (100 to 200 milligrammes). If rabbits are infected by means of living tubercle bacilli, a conjunctival reaction can be obtained from the third day onwards, increasing in the days following, and disappearing after fifteen to eighteen days, simul- taneously with the arrival of the period at which the weight begins to decrease, and more severe tuberculous changes appear. The most recent communications from a veterinary side which have been published since the appearance of the German edition fully confirm these views. (90 to 98 per cent, of positive conjunctival reactions in bovine tuberculosis.) Through the kind assistance afforded to me by the Prussian Ministry of Agriculture, I am now in a position to form a personal opinion concerning the conditions in question. Our attention will be directed first of all to the problem whether we are enabled, by means of the conjunctival test, to recognize the presence of tuberculosis in such animals as have been previously injected with tuberculin. It is well known that this method is frequently employed to deceive the German quarantine officers, and thus to introduce into Germany tuberculous oxen which did not react with fever to the sub- cutaneous tuberculin injection on account of the previous treatment with tuberculin. DIAGNOSIS OF TUBERCULOSIS 97 C— THEORETICAL PART. THE COMPARATIVE VALUE AND SIGNIFICANCE OF THE CUTANEOUS AND CONJUNCTIVAL KEACTIONS. In the clinical part we have already dealt with the clinical significance of the reactions, and we only wish to supple- ment the observations made there by a few more theoretical points, which, however, will prove of great practical value. We have seen that in the case of tuberculosis no great difference exists between the results obtained by either reaction, and that great differences are found in those individuals who, on the basis of clinical observation, are not regarded as suffering from tuberculosis or as suspected. To put the matter concisely, according to our own investigations one-half, and according to the investigations of others three-quarters or more, of those clinically unsuspected of tuberculosis react cutaneously, whilst conjunctival reactions according to our own investigations occur in one-sixth, and according to those of others in an even smaller number of clinically unsuspected individuals. One is almost tempted to discuss the question whether in adults the cutaneous reaction is to be regarded as specific. Its specific property is disputed by many, and for this reason only the conjunctival reaction is employed by them. In the clinical part I have left no doubt as to my opinion, having clearly stated that for purposes of clinical diagnosis I distinctly prefer the conjunctival reaction lo the cutaneous, especially on account of the results obtained in health}' in- dividuals. It is quite evident that it is much more likely that one-twentieth to one-sixth of those thought to be non-tuber- culous should have tubercular centres in their bodies, than one-half, three-fourths, or even more. If all human beings should happen to be tuberculous, the recognition of this fact would certainly be of scientific interest, but clinically it would not have any practical and diagnostic value. Yet I believe the cutaneous reaction to be specific also in the case of adults. All that has lately been put forward against the specific nature of the cutaneous tuberculin reaction (Entz, Wien. klin. Wochenschr., 1908, No. 12) is based on false theoretical conceptions. As soon however as it is assumed 7 98 THE OPHTHALMIC AND CUTANEOUS that the cutaneous reaction is specific, it must at once he ad- mitted that it is by far the more sensitive of the two. This sensitivity of the skin with regard to tubercuHn is certainly somewhat embarrassing, because, as we have seen before, the basis of the reaction lies in the so-called hyper-sensibility, i.e., the rapid but very strong capacity on the part of a tissue to react against the slightest irritation. By reason of its rich blood supply and the thinness of the epithelium, the conjunctiva is certainly more capable of reaction than the skin, and in those who are sensitive to pollen, the con- junctiva is highly sensitive, whereas the skin in this case is very slow in reacting. Therefore in the case of tuber- culin either special conditions prevail, or in the adult there is a widely distributed reactive capacity on the part of the integument with regard to derivatives of tubercle bacilli. Such hyper-sensibility may sometimes have a very deep signi- ficance. And now we have to consider whether this hyper- sensibility is naturally present in an individual, or whether it only occurs by reason of the body having previously been in contact with tubercle bacilli or their products. Both these possibilities may be regarded as reasonable. If during the experiment the body for the first time comes ni contact with typhoid or cholera bacilli, the antitoxin which is present in the serum (without it being necessary for the body to have come in contact with these bacteria before) dissolves the bacteria, and the action of the poison thus liberated brings about the signs of inflammation. The oitlij changes which take place in the body by coming in con- tact with the bacteria consist in the fact that (1) the hyper- sensibility is increased, and (2) the process of dissolving becomes more rapid. Similarly, it is possible to regard the reaction of the skin to tuberculin as virginal ; that is to say, the first contact of the skin with the products of tubercle bacilli. This hypothesis explains the fact that in clinically healthy individuals the reaction in a great number of cases is a late reaction. This late reaction would correspond with the slow bacteriolysis and the slow onset of the inflammation in the above-mentioned first typhoid infection, resulting in a long period of incubation, lasting eight days up to the appearance of the first symptoms in serum disease, which ni the case of re-injection appear after twenty-four hours or even less. Accordingly, the late reaction would be the reaction of an individual who had never before come in contact with DIAGNOSIS OF TUBERCULOSIS 99 tubercle bacilli ; the early reaction, on the other hand, would have to be regarded as the reaction of a body which formerly had come in contact with tubercle bacilli, corresponding to the behaviour of the body in the case of re-injection. Now it is very remarkable that at the first application of the reaction a great many individuals do either show no reaction at all or a late one, whilst, if repeated, they give a rapid reaction, that is to say, they give a prompt reaction of hyper-sensibility. (See Pirquet, Wien. klifi. Wochemchr. 1907, No. 38.) The theory that in the period between the two reactions the body has become infected with tuberculosis may be altogether dis- regarded for the majority of cases, despite the fact that Pirquet refers to an observation in which he regards this as having been the case. If this assumption on the part of Pirquet should prove to be correct, his case would have been one in which the infection had occurred in a hospital, which would involve us in a discussion concerning hygiene in hospitals. On the basis of experiments with guinea-pigs which were put with advanced tuberculous patients, I believe it to be a great mistake to place initial, and especially suspected, cases in one room with advanced tuberculous individuals, as is the case at any rate in German sanatoria, and even in the best appointed hospitals. The clinician may say he has never observed infections originating in the hospital, but, considering the nature of tuberculosis, this really does not prove anything. In discussing this question it must be mentioned that it is actually possible to produce hyper-sensibility by inoculating a minute amount of foreign albumin substance. We know from the investigations of Rosenau and Anderson that in animals a condition of hyper-sensibility may be induced by the injection of very minute quantities of seram. If we were to assume the same thing to occur in the case of tuberculin (the production of a condition of hyper-sensibihty by only one cutaneous inoculation), we would still have to explain why a comparatively very considerable percentage of individuals who had not given a positive result at the first application of the reaction do not show any hyper-sensibility when the reaction is repeated. The percentage referred to is from 25 to 50 per cent. The reply to this question would be that also in the case of serum disease a number of individuals prove to be insensible to repeated injection, and that especially in the case of hay-fever it is not the majority of 100 THE OPHTHALMIC AND CUTANEOUS individuals coming in contact with vegetable pollen who become hyper-sensitive, but only a comparativel}^ small per- centage of such individuals. So far our train of thought allows us to be fully in favour of the theory that the late reaction is to be regarded as a virginal reaction of the body to the first cutaneous contact witJi tuber- culin. The wide distribution of tuberculosis, however, makes it little likely that there are many people who have never come in contact with tubercle bacilli. But even if this be admitted, the observation of Pirquet, confirmed by all other writers, is discounted by the fact that from yen.v to year a greater percentage of children react cutaneousiy. Newly born infants do not react ; from the first to the third year of age the per- centage of reactions increases to such an extent that at the age of from 5 to 10 years most children already react on tuber- culin. According to v. Pirquet {Wien. klin. Wocheiischr., 1907, p. 38) the post-mortem results are given of 100 children in whom the reaction had been applied. Half of those who were absolutely without tuberculosis were in tlieir first year of life ; out of thirty-three autopsies of children above the age of 3, only six v/ere without tuberculous changes. Therefore, the capacity to react on tuberculin cannot be innate in man, but must be acquired. In what way, then, has it been acquired ? It is hardly possible to ascribe the origin of hyper- sensibility to anything but the fact that the body has pre- viously come in contact with tubercle bacilli. Since a late reaction is to be regarded as a slowed-down and diminished liyper-sensibility reaction, it would indicate the presence in the body of an inactive, perhaps a long since healed-up tuber- cular centre, and that the cells of the body for a long period have had no occasion to react on tubercle bacilli or their products, but that, under the renewed stimulus of toxin of tubercle bacilli, they had re-assumed their former reactive capacity. When the reaction is repeated, therefore, a hyper- sensibility reaction promptly occurs. Our view that late reactions are due to inactive tuberculosis is supported by our post-mortem investigations, positive cutaneous reactions, especially late reactions, without conjunctival reactions, always showing latent tuberculous changes. It is of very great clinical importance if so simple a process (comparing the results of the cutaneous with those of the conjunctival reactions) enables us to diagnose correctly either a healed-up tubercular centre or active tuberculosis. DIAGNOSIS OF TUBERCULOSIS 10 L In the case of the conjunctival reaction, late reactions are of very rare occurrence. Also in healthy individuals they are very rare. This probably is due to the blood supply of the conjunctiva as referred to above. Summarizing, we will say that the greater clinical value must be attached to the conjunctival reaction. If the cutaneous reaction be positive and the conjunctival negative, we may diagnose healed-up or inactive tuberculous centres. This is the reason w4iy it is desirable not to confine one's self to one method, but to employ both reactions together, because they complement one another to a very considerable extent. Furthermore, the cutaneous reaction may also be regarded as a valuable control of technical mistakes in the application of the conjunctival reaction. Since the cutaneous reaction rather tends in a positive direction, its absence constitutes a proof that the conjunctival reaction remains negative on account of the condition of the patient, and not on account of any technical mistake that may have been made. That we have been able to arrive at these interesting conclusions by observing our material is due to the fact that we have always employed both reactions simultaneousl}^ and I find satisfaction in the consideration that my conjunctival tuberculin reaction has not done away with the need for Pirquet's cutaneous method. THE REACTION IN OTHER INFECTIOUS DISEASES. After the favourable results obtained with the local reactions in tuberculosis, I naturally felt inclined to extend my experiments to other infectious diseases, von Pirquet has already, I understand, tried to apply his discovery to other infectious diseases, or at any rate he has conveyed his intention by word of mouth. My conjunctival reaction would be even more adapted to such a proceeding by reason of the greater reactive capacity of the conjunctiva, and in reality I have carried out experiments in typhoid patients in Friedrichshain long before Chantemesse, and even before I gave publicity to the conjunctival reaction with tuberculin. For this purpose I made use of Picker's typhoid " dia- gnosticum," which is analogous to tuberculin, but I did not succeed in obtaining a distinct eruption on the skin and on 102 THE OPHTHALMIC AND CUTANEOUS the conjunctiva. I attribute my failure to the unsatisfactory bacterial extract represented by Ficker's reagent. A work of Waele (Soc. d. Biol., vol. Ixi., No. 28) deals with the theoretical basis underlying the transmission of the reaction to other infectious diseases. From a practical point of view this transmission is so obvious that Chantemesse does not even mention this work, and it is probable that he has begun his investigations without a theoretical basis, and merely on the ground of my communications. Waele started from the following suppositions : If small collodium capsules filled with a bacteria culture are inserted under the skin or into the peritoneum of animals in accordance with the well-known French method, its diffusible products are incor- porated in the animal body in question. Toxins which are affected by heat, and which produce immunity, do not spread by means of collodium capsules. The diffusible products here referred to are (comparatively) not affected by heat ; they circulate and produce no immunity, but hyper-sensibility. Animals with such capsules introduced underneath the skin are brought into conditions parallel to those prevailing in the case of infection with tuberculosis. The substances spread slowly and continuously, and produce a lasting effect in the animal. If, for instance, the capsule is filled with tubercle bacilli, a tuberculin reaction is obtained in from two to three weeks, ^ if a diphtheria, typhoid, or pyogenic culture is brought into the capsule, and if then the animals are subcutaneously injected with the homologous products formed in vitro, a temperature reaction is obtained in every way similar to the tuberculin reaction. It could not be decided at once whether, under natural con- ditions in the case of acute infectious diseases, the time it would take the bacteria to act would be long enough to produce such a result, or whether this time for the bacteria to act would have to be prolonged, although in the latter case, of course, smaller doses are absolutely necessary. Although this funda- mental question so far had not been finally settled, it was yet easy to carry out experiments on the conjunctiva of patients, analogous to the conjunctival tuberculin reaction. It is hardly necessary to mention that this form of diagnosis would be of the utmost value in cases of typhoid, as also in the treatment * Analogous observations have quite recently been recorded by Moussu {Bern, med., 1907, No. 49). DIAGNOSIS OF TUBERCULOSIS 108 of other infectious diseases, and, if possible, of still greater value in combating the plague. Chantemesse was the first to publish his excellent results obtained in diagnosing typhoid by means of the conjunctival method. He found that it was superior to the serum method, which up to then had been the only one at his disposal. In sixty-nine cases of typhoid, both the serum and the conjunctival reactions produced positive results, and in forty-eight cases of other diseases both these reactions proved negative ; in eight cases the conjunctival method proved superior to the serum diagnosis, the conjunctival reaction giving a positive result three to thirteen days before the serum reaction. " J'attache le plus de prix, parce que je ne I'ai jamais vu manquer chez les vrais typhiques " (Chantemesse). The diagnostic substance consisted of an extract of typhoid bacilli, the preparation of which rests on the principle that at a temperature of 60° the bacteria are killed off, and then undergo various processes of precipitation and purification. (For further particulars see the Congress number of the Deutsch. med. Wochenschr., September, 1907, No. 39, and " Serotherapie de la fievre typho'ide," Paris Cour. d'appel, 1907.) But very few other authors have so far carried out experi- ments in this direction. Kraus, Lusenberger, and Euss have tested the substance in some epidemic of typhoid {Wien. klin. Wochenschr., 1907, No. 45), and have arrived at uncertain and conflicting conclusions. Kraus maintains, with Pibram, that at a temperature of 60*^ the poisons suppressing the reactions are annihilated, and that the production of the reaction is dependent on certain bodies of a toxin-like character. Follow- ing up all these precepts, be obtained no more than 60 per cent, reactions, and he much prefers the serum diagnosis. Cohn (Ver. f. innere Med., Berlin, January, 1908) obtained positive conjunctional reactions in healthy individuals by means of typhoid extracts, whilst Entz obtained cutaneous reactions by the same means in healthy individuals {Wien. klin. Wochenschr., No. 13, 1908). This opens up one of the most difficult problems concern- ing the science of immunity. Whereas the school of Pfeiffer does not admit the occurrence of toxins, especially in typhoid and cholera, Kraus and others now have published experiments which go to show that a formation of toxin results from these bacteria. One might, perhaps, combine the two views by 104 THE OPHTHALMIC AND CUTANEOUS assuming the possibility that antitoxins may be produced against a few bacterial constituents changed by the method, whereas this would not be possible with regard to the un- changed bacterium albumin (li. Pfeiffer). Analogous to this theory we have the fact that no antitoxin can be produced by means of animal albumin ; although this may again be possible in a certain sense by means of reduction (reduction toxin, fatigue toxin ; see the works of Weichardt, and also a collective referendum prepared by myself for the Zentralhl. /. Bakt., vol. xl., part 5, on the basis of personal observations). Eisenberg takes the view that it may be possible that apart from bacteriolysis the living bacteria secrete a poison, an action not unhke the throwing off of plasmatic molecules by the substance of the body. Furthermore, the investigations of Grassberger and Schattenfroh {Wien. klin. Wochenschr., 1907) show that the production of anti-toxin does not under ail conditions constitute a protection even wdth regard to the kind of bacteria producing the toxin (Kauschbrand). This probably explains why, in the experiments of Kraus with regard to bacterial infection, the activity of the antitoxin was merely a moderate one. Whatever be the actual development of this theory, the difficulties involved in the application of the conjunctival reaction in other infectious diseases are due to the difficulty attached to the technical production of the appropriate diagnos- ing medium — that is to say, an appropriate extract of bacteria — and also to other theoretical difficulties, which are brought about by the presence of bacteriolysine substances, which so far have never had occasion to go through the disease in question. Personally, I have often experimented in this direc- tion without having arrived at a satisfactory conclusion. The filtrations obtained by means of the Berkefeld filter contain, or at all events contained in my own experiments, too little active substance, which could not be increased to a sufficient extent even by shaking the preparation for several days (before filtration) in the shaking apparatus. In making use of the bacterial bodies, they have to be heated to 60° for an hour in the interest of the patient, in order to kill off the bacteria : and if the suggestion of Kraus, to the effect that at a tempera- ture of 60° all active substances are annihilated — which I do not beheve to be the fact in all cases — were confirmed, this way of proceeding would never lead to the end in view. In all these experiments it has to be considered that the great DIAGNOSIS OF TUBERCULOSIS 10') instability of the substance of the bacterial bodies (tiie endotoxin) only makes it possible to work for at most twenty- four hours with a newly prepared extract. Recently I have tried to use, as a medium for diagnosis, agressin exudations which had been exposed to a temperature of 60° for an houi'. containing dissolved substances of bacteria bodies, the solu- tion being prepared in such a manner that the endotoxic action of the substances in question is not disturbed (see note on agressin, Zentralhl. f. Bakt., vol. xxxviii., ref. Nos. 21-28). Although I do not by any means attach too little import- ance to the difficulties which are in the way of applying the conjunctival reaction in other infectious diseases, yet these difficulties are to a great extent of a technical nature, and will certainly be overcome as time goes on, considering the great progress that has been made of late in the production of unstable bacterial products. In the case of the tuberculosis diagnosis, the conditions were extremely simple. By a comparatively very simple process — and, after all, it is of very little importance which of the numerous methods at our disposal we employ — the sub- stance of the body of the tubercle bacillus can be obtained, and can be preserved for years. Putzeys and Stiennon must also be mentioned when deahng with experiments of this nature {Soc. de Biol., 1907, No. 27). Starting from the principle that in animals with glanders the subcutaneous injection with virus of glanders produced local symptoms much more distinct than those obtained with tuberculin in tuberculosis, and at the same time considering the morphologic relationship between glanders and tuberculosis, these authors employed in a glanders epi- demic old and also fresh virus of glanders for the application of the cutaneous and conjunctival reactions. The reactions were either altogether negative, or only very weak and ephemeral ; yet I feel convinced that, especially in glanders, good results are to be be obtained with this method. Mean- while Machel {Berlin, klin. Wochensclir., 1908, No. 9) has obtained very good results in the case of glanders. The differences in the results obtained may be due to the vari- able material on the market. THE PATHOLOGICAL ANATOMY OF THE CUTANEOUS AND CONJUNCTIVAL REACTIONS. The Cutaneous Beaction. — So far we have only been able to observe rather advanced forms of reaction, viz., papules : 10() THE OPHTHALMIC AND CUTANEOUS that is to say, products with a fairly strong infiltration. An investigation of early reaction forms, therefore, would be very desirable. In the cutaneous reactions that have been investi- gated, great accumulations of lymphocytes are found around the more deeply situated blood-vessels of the skin, especially around those vessels situated in the adipose layer, close to the corium, and also around those surrounding the sweat glands. Bandler and Kreibich have in a few cases observed distinct giant-cells in these accumulations of lymphocytes, but not of the Langhans type, so that the authors do not regard these formations as tubercles, but merely refer to them as tuber- culosis-hke formations. Daels (Med. Klinik, 1908, No. 2) in investigating papules m our material, has found, besides ordinary giant- cells, also those of the Langhans type, and there- fore regards the pathological-anatomic correspondence between the formations produced by tuberculin and a tubercle as proven. Our own view (which further on we wall develop more fully) is that no essential difference exists betw^een tuberculin and tubercle bacilli, but that the action of tuberculin is solely due to the bacterial bodies it contains. Therefore, according to this view, any differences observed between the action of tuberculin and that of tubercle bacilli can only be due to the endotoxic substances present in the tuberculin being impaired in the course of preparation, whilst all other differences of action can onl}' be explained by assuming the existence of quantitative differences. Now the harm done to the endotoxic substance of the tubercle bacilli in preparing tuberculin seems to be but extremely minute, and for this reason the difference between the action of tubercle bacilli and that of tuberculin is, as a matter of fact, merely of a quantitative nature, caused by the fact that the living bacilli — a virus capable of multiplying — can always create fresh toxin to replace the old, in this way renewing and prolonging the action of the poison. The action of this prolonged influence of the toxin causes necrosis, and v. Baumgarten actually demands the presence of necrotic changes before diagnosing tuberculosis, not because the tubercle bacilli could not, in the beginning of their activity, produce formations without necrosis, but by reason of the fact that also other harmful substances can produce tubercle-like formations, and that it is only the presence of necrosis which pathologic-anatomically enables us to definitely conclude that living tubercle bacilli are present. In the different forms of the giant-cells analogous con- DIAGNOSIS OF TUBERCULOSIS 107 ditions seem to prevail. Without intending to discuss the voluminous arguments concerning the genesis of these much talked-of giant-cells, they seem to occur in tuberculosis only at such places at which a slow hacteriohjsis takes place in which the toxins are set free. The actual reaction of the body to toxins consists in the accumulation of lymphocytes. These correspond to the specific irritation, and the presence of giant cells seems only to refer to quantitative variations in the action of toxins. Again, it seems that it only depends on quantitative influences of the toxic action whether we find simple giant-cells or those of the Langhans type. Tuberculin with its fixed toxin can- not, as a rule, bring the tubercles it produces to necrosis ; we might regard the cells of the Langhans type as the very com- mencement of a necrotic process. When looked at from this point of view the old theory of Weigert regarding the giant- cells as cells with a commencement of necrosis, seems to have many points in its favour, and is supported by the microscopic picture of the cell in the centre of which we still find un- disturbed tubercle bacilli. The importance of the quantitative conditions of toxic action in the causation of necrosis may be seen from the fact that by inoculation with tuberculin an ulcer may be produced, that is to say, a loss of substance through necrosis, if the inoculation is effected in the lupous tissue (Nagelschmidt, Deutsch. mecl. WochenscJir., 1007, No. 40). This conception of the giant-cells as products of the reaction on toxic action is all the more important in view of the unreasonable phagocytic theories. Daels, for instance (Ver. f. inn. Med., December 16, 1907), injected dead tubercle into the carotid, and afterwards found tubercle bacilli (dead ones, of course, as the experiment would show) in giant-cells (in the eye). On this ground he concludes that the giant-ceils play a protective part in the battle against the tubercle bacilli. This, of course, simply plays into the hands of the opponents of the phagocyte theory, since a protective function on the part of the cells in question is assumed on the basis of their taking in dead bacterial substance, whereas the phagocytes are sup- posed to attack only the living bacilli. These giant cells are not protective cells formed in the body, but they have been formed out of other body cells as a product of reaction on the part of these cells against the action of the toxin. They disappear if the toxic action continues for some tune. 108 THE OPHTHALMIC AND CUTANEOUS From their point of view the pathologists are quite right in using the giant cells of the Langhans type for diagnosing a tubercle, since they cannot diagnose tuberculosis on the basis of an accumulation of lymphocytes ; but even a very careful pathological diagnosis based on morphological factors cannot do away with the indisputable fact that the primary activity of the tubercle bacilli essentially consists of the forma- tion of a lymphoma resulting from an emigration of lympho- cytes from the vessels. Philipsohn has shown in an analogous manner that in the case of folliculitis the initial alteration is represented by a small-celled infiltration of the vascular wall of the smaller vessels. In this connection we will point to the interesting constant relationship existing between the tubercle bacilli and tlie lymphocytes. In the tuberculous exudations of the pleura on a tuberculous basis, lymphocytes are found either exclu- sively, or at all events in preponderating masses, and the same I have found to be the case in the sputum of tuberculous patients, as indicated in another part of this work. Together with V. Torday I have obtained, in experiments with tuber- culin, lymphocyte exudations containing a percentage of 80, 90, or even higher, of lymphocytes. In the meningeal exudations Kronig found a preponderating number of lym- phocytes in tuberculous individuals ; and we have only to add that a transformation of lymphocytic forms into poly- nuclear formations enables us to unhesitatingly pronounce an unfavourable prognosis. Thus we find lymphocytes in all cases m which tubercle bacilli have come in contact with the animal body, in the tubercle ; in the exudations of a tuberculous etiology as a result of the action of tuberculin in experimentally produced exuda- tions ; and in the skin papules obtained with tuberculin. These conditions are all the more interesting biologically by reason of the fact that lately Bartel and Neumann have vindicated the theory that lymphocytes exercise a protective function with regard to these tubercle bacilli, by demonstrating that the virulence of living tubercle bacilli is distinctly lessened by the bacilli being brought in contact with lymphatic tissue, which corresponds with the observations of Brieger, Kitasato, and Wassermann with regard to other bacteria (cited aftt-r Bartel, Wien. Min. Wochenschr., 1907, Xo. 44). They not only assume the existence of connections between tuberculous infection and lymphocytes in animals, but also in DIAGNOSIS OF TUBERCULOSIS 109 the case of man ; to support this view they adduce the latency of tubercle bacilh in not specifically changed lymphatic tissues, as also the fact that in children lymph passages are found which are slightly reticulated, whilst in adults there is a denser lymph filter ; and further the decrease in the resistance to tuberculosis in the status lymphaticus, observed by Paltauf, and which is said to distinguish itself by a strongly destruc- tive growth of the supporting stroma and the almost com- plete disappearance of the specific parenchyma. I believe less in the harm done by the lymphocytes to the tubercle bacilli than in the role these former play in the formation of connective tissue and the encapsulation of the tubercle bacilli. Maciesca Jelenska {Brauer's Beitrdge, vol. viii., p. 1), again, interprets these observations as the formations of antitoxic substances, which of course cannot be the case. The purpose of the work is to show that in cases of tuberculous pneumonia numerous plasma cells are found in the perivascular, peri- bronchial, interlobular, interalveolar and subpleural connective tissues, as also in the border regions of the tubercles ; the author is of opinion that these plasma cells are the product of haematogenous lymphocytes found in the respective places. Nicolai A. Bibbin ("■ Inaugural Dissertation," Berlin, 1907) has found lymphocytes in the walls and capillaries of tuber- culous glands. Despite the fact that phagocytosis can be demonstrated, the polynuclear leucocytes do not possess any virulence- weakening influence with regard to tubercle bacilli. As mentioned above, fresh forms of reaction of the skin to tuberculin have so far not pathologic-anatomically been investigated. Clinical observation already shows that in the formation of the papule, exudative and infiltrative processes play a part, varying in intensity and strength. The original exudation is very tense, and after some time it changes into infiltration. The infiltrative forms are those in which the reaction remains a long time. They can be at once recognized as the form pathologic-anatomically described above, and represent the most favourable reaction of the body to tubercle bacilli toxins. They are those forms which bring about the localiza- tion of the tuberculous processes and prevent the propagation of tuberculosis. Cunjiuictlval Reaction. — The conjunctival reaction has so 110 THE OPHTHALMIC AND CUTANEOUS far not been investigated from a pathologic-anatomical point of view, and we cannot say, therefore, whether the changes m the vessels and in the interstitial tissues correspond with the cutaneous changes, as is to be expected. The investiga- tions carried out have to do with the cytological conditions. Subraxis and Duperie have discovered mucus and polynuclear leucocytes, and have demonstrated the absence of bacteria ; so also Mongour and Brandeis, who also mention the hbrin formations which mostly occur m the stronger reactions. ISIobody seems to mention the presence of lymphocytes in the exudations. (Added during revision : Fritz Levy informs me verbally that he has discovered accumulations of lympho- cytes in the conjunctivae in the reaction stage.) THE PROGNOSTIC SIGNIFICANCE OP THE REACTION. Following on this we will propound our theoretical views concerning the prognostic significance of the reaction in so far as this has not been done in the clinical part, to which we refer. I will only repeat here in a few sentences the principal results obtained by Stadelmann and myself in the clinical investigations we have carried out together. Out of fifteen individuals in the third stage of tuberculosis who did not react cutaneously (out of these thirteen did not react con- junctivally either), fourteen died within the very short period of six weeks ; the results in other stages are even more remarkable. Of those in the second stage who did not react two died, and one rapidly passed into the third stage ; whilst from non-reacting individuals in the first stage one died, and post mortem it was found that the simple reaction had given a better prognosis than the very thorough clinical investigation, the autopsy showing that the clinical diagnosis assuming the first stage of the disease was a mistaken one, and that in reality the patient was in the third stage. It would be a mistake to think that the reaction indicates the pathologic-anatomical condition of the lungs. Thus, if the reaction is negative, it does not mean that we have to do with a case of tuberculosis in the third stage, since a reac- tion may be negative even in the first stage of tuberculosis. But a negative result indicates that the body is no longer capable of counteracting the toxins of the tubercle bacilli, and therefore has to be interpreted unfavourably from a pro- gnostical point of view. A positive reaction, on the other hand, DIAGNOSIS OF TUBERCULOSIS 1 1 1 indicates reactive capacity on the part of the body, that is to say, the capacity on the part of the body to offer resistance to infection. Whereas a negative reaction may always be regarded as of unfavourable prognosis, the opposite conclusion may not be drawn from a positive reaction. The positive reaction indicates that a battle is being fought which as yet is undecided, and tlje result of which cannot yet be foreseen. He who would base a favourable prognosis on the result of the reaction would make the mistake of resardincr battle and victory as synonymous. These remarks will explain why sometimes in very advanced cases the reaction is positive. Experience has taught me that these are the cases in which many years of a diseased condition have produced great pulmonary changes, despite which, death is still far off. These important diagnostic and prognostic conclusions may be gained by simply observing the course of the reaction during four consecutive days. Stadelmann, who is well known as a careful and critical clinician, has definitely fallen in with the prognostic conclusions the reaction enables us to draw, and he attaches an even greater significance to the prognostic than to the diagnostic value of the reaction. These observations would be sufficient to regard the expecta- tions I expressed on May 15, 1907, at the Berliner medizinische Gesellschaft, as fulfilled. But I was not content with the results. I was of opinion that the results thus far obtained merely represented the foundations of a structure, the develop- ment of which in time would prove of the very greatest value. So far, only the crudest diagnostic means had been employed, and the quantitative differences of sensibility to different con- centrations, already fully demonstrated by Pirquet and myself, had hardly been considered. 8o as not to complicate matters too much in the investigation of great numbers, we did not take into consideration that distinct differences are found in each particular case according to the preparation of tubercle bacilli employed ; old tubercuHn, fresh tuberculin, tuberculol, emulsion of bacilli, tuberculin fi'om tuberculous oxen, S:c., all produced different results, and their observation and appli- cation promises further diagnostic and prognostic success. (See the chapter on the differences of tuberculin reactions.) The prognostic significance of the reaction seemed to me to be the most important, and starting from the principle that a representation of the course of the reaction by curves would 112 THE OPHTHALMIC AND CUTANEOUS lead to important and decisive discoveries along this line, I made a careful study, in co-operation with my colleague, Mr. Teichmann, assistant of the hospital in Friedrichshain, of the reaction as represented by most accurately drawn curves, com- paring the results obtained with clinically investigated cases, and I believe we have succeeded in obtaining excellent evidence in favour of the prognostic significance of the reaction. THE PROGNOSTIC SIGNIFICANCE OF THE CUTANEOUS AND CONJUNCTIVAL REACTIONS. The following observations were made when Koch's old tuberculin (tested by order of the State by the Kuete- Enoch method) was used in a 1 per cent, solution for the conjunctival reaction, and in a 25 per cent, solution for the cutaneous reaction. In order to obtain results of prognostic value and which may be compared, it is not only essential to employ the same concentration, but also to use tubercuhn of the same origin, and we advise the ordering of a large amount of tuberculin at one time, because even if the origin of preparations be the same there may easily be some differ- ence in the power of action in tuberculin of different consign- ments. The fact that the reaction depends on the degree of concentration of the solution employed shows that there is a difference in the influence of solutions containing different amounts of tuberculin, although Pirquet may not be quite correct in assuming that the strength of the action is pro- portionate to the strength of the solution employed. (See Curve fig. No. 1, p. 114.) The differences, however, are only of the intensity of reactions, and do not refer to the course of the curve. Now from our observations it is clear that the form taken by the diseases typically condi- tions the law governing the course of the reaction. In order to allow the judgment concerning the course of the reaction to be entirely independent of the subjective obser- vation of the investigator, and also to make the results very plain, I and my colleague Teichmann have drawn up a scheme which served us in drawing up the curve figures (see the curve tables). As a rule the reaction was applied in the morning", so as to enable us to make observations six or seven tnnes in the course of the day, and also in order not to overlook the rapidly fading type of reaction, as once occurred. The pomphous, DIAGNOSIS OF TUBERCULOSIS 113 or papule, which came up was measured by means of a tape measure, measuring the infiltration and the reddish halo separately, since they can almost always be separated. The curves representing the infiltration are as a rule homologous, •^-nd the separate measuring has the advantage that artificial irritation, such as is produced by scratching, can easily be distinguished from the curves themselves. For the rest the infiltration during the first few days consists of a strong exudation which, if palpated by the finger, gives almost the same impression as an infiltration proper. Despite all our experiments in this direction we have not been able to clinically differentiate between infiltrative and exudative pro- cesses. This can only be done by means of the pathologic- anatomical investigation. The length of the pomphus is then registered on the ordinate of the curve, whilst the abscissa indicates the time which has elapsed after the application of the reaction. It is more difficult to represent by a curve the course of the conjunctival reaction, since its values cannot be expressed numerically. But the use of the indications "trace," "first degree," " second degree," and " third degree " of the re- action suffices to represent the course of the conjunctival reaction. The curve figures, Nos. 1 to 11, refer to typical cases chosen according to clinical indications. To prevent miscon- ceptions I will lay special stress on the fact that they do not represent cases that have come under observation in a haphazard way. On the contrary, they show by exact curves what could only be expected after the investigations Professor Stadelmann and myself were carrying out ; they prove to a most exact degree the correctness of the funda- mental principles deduced from our investigations of great numbers as dealt with in the Clinical Part, and which refer to the local application of tuberculin reaction. The following types of curves are to be recognized : — I. The normal type, representing the specific normal reaction of patients infected with tuberculosis. — This is a vivid reaction which fairly quickly appears. It becomes apparent four to six hours after the application of the reaction, and reaches its maximum after twenty to twenty-four hours. This maximum remains at the same level on the second day and pales down in the course of the third, or, at the latest, during the fourth day. In the place of the pomphus there remains 8 114 THE OPHTHALMIC AND CUTANEOUS a slight pigment stain of the same size for one to two weeks, which then absolutely disappears. One may in this normal type still distinguish between a strong course and a medium strong course, which may become clear from the curve tables. Curve, fig. 1 : strong reaction. ) ,, .,-, , . . '- See p. 114. Curve, fig. 2 : medium strong reaction.] ^" As above mentioned, this normal type is the type of the specific normal reaction. This course is found in the great majority of cases of initial tuberculosis, and also in those cases of the first and second stage which show a favourable, that is to say a slow, progress. There are two departures from this normal type • — (1) The rapid hut very weak and ephemeral 7-eaction in the case of manifest tuberculosis, which is frequently overlooked, and (2) the late or lasting reaction. A special clinical sig- nificance is to be attached to both these forms. We will first deal with the weak and ephemeral rapid reaction of manifestly tuberculous cases. This reaction is distinguished by its rapid course and its very superficial nature. Like the normal type, it already appears six hours after the application of the reaction ; it soon reaches its maximum point, sometimes as early as ten hours after application, and it disappears just as quickly, at the latest in the course of the second day. It does not leave any traces behind. This reaction frequently takes a course so rapid, and its intensity, even at its maximum, is so slight, that it is very often overlooked, unless, indeed, the observations are made every two hours. We are sure that in our clinical observa- tion of great numbers, in which the result of the reaction was taken up after twenty to twenty four hours, it has been over- looked, since after twenty-four hours nothing is left of it. (This form of reaction is frequently found in cases of tuberculosis of the third stage, more rarely in those of the first and second stages of pulmonary tuberculosis, that is to say, in those cases which are distinguished by a specially unfavourable course of the reaction). (See curve figures 3, 4 and b, p. 116.) Such a form of reaction, therefore, indicates (as is evident from these remarks) an unfavourable prognosis, viz., a rapid course of the tuberculosis. If the reaction is altogether absent the chances are even more unfavourable, and this has only been noticed in patients who died within a short period of time. Curve Table I. rst Stage. J'oAn Bale, Son: &' Danielssor , Ltd., London. ^ t 5i f 1 --- -- ■— L , •- , ' •• 1 ! ■■ c 1 i I j ! 1 -- - --• ! i i 1 i riot 1 1 X — c — > 1 — J— X— x-f 1 -X-tX X— ; — X X — X — X- -X — X— -X — X— <— X X — X— X— IV )sis D «?'• Tal V. cen in Ci Dismi fav VI V Octob irve ol ssed ii ourabl 77. V er anc 1 gooc 3 local 'III . I 8th D ecemb condi appej IX. X ecemb er '07. tion ai irances . Day. er 1907 ter According to Wolff-EUner %nd Teichmann. Cutaneou 3 1 ^eact ior 1 in a case of Tubercu osis in l\ e Firs Staj ;e. yoh„ b aU. Sn -- b- Do nicU„.. «A, , ^d«,. 1 - -H ... ... [- ..J 1 ... 1 !; 1 1 1^' \ ^ K* s ^ ■♦l ^+-* .K- frr! — St -*, « s ... ..!. ... ■ ' 1 1" r tr- ■»- jCJ, i 1 -- \ N 1 i 1"^ I 1'^ f », ■ 1^ \ ! 1 Tla. We f" ii nA. izn. r- ;. - - s \ » \ fe / K ^*, ' N te / K f-J^ Ui \ f « " y 1 ''x ^ f^ii \ '-' !a ^ 1 \ N \ * J y ^' *- s\ - - c ^ L / \ S ■^« / / \ s ^ !: s / \ ^*^ 1% - - - r - / J ^V ^ \ •x - - S ^ / '^ V '' s . ; iS* / / / V. \ \ Si ._. >■ / ji .■^ • \| \ •s, to ,+ /' 'V \ \ ■^ kV • % \ \ ^ 1 * 1 i J r / 1 /' \ '*--N *^ \ s " /' ' /■ \ -*. i ^" ^^ " ^ d t: - - - - — 1 • - - - - - - IT -:k * d a g? ^ a-— Reddening) lower 1 o/J ^^ercuUn. Wolff-Eisner, The Ophthalmic and Cutaneous Diagnoses of Tuberculosis. K VI VII VIII IX. X. Day. Taken in October and 8t>' December 1907, Curve of St^- December '07. Dismissed in good condition after favourable local appearances. ¥Q Tuberculosis. John Curve Table II. Bale, Sons &• Daniehson, Ltd., London. 1 I 1 39 1 ?«.,^ ^^ ! V ^^-a;. 1 1 ! i — j 1 1 — i 1 — i 1 1 1 1 K — 1 t S ^ 1 cj 1 1 ^ 1 °Uk ■*'■> '1 N o^ — .-— ■ .-:^ — ' > t-i-x* ieZ3 ^I >^ V i T n*=^-T>< =-, Xi e IV TCUl( : D )sis ay 20'! V. > Jan. VI. P raken '08: cc 71. I' on 8th )nditioi Ill J Dec. 1 very rx. X '07. satisf. . Day. aotory. Curve Table II. Conjunctival and Cutaneous Reactions in a case of Early Tuberculosis. According to Wolff-EUner ar i Teichmann y,h,i „l>. So „b-D „iW LU.. ..»*,,. -'1 ... -■ ... ...... "t s /^ ... -- ... ...... ... _.. ll ^- / / \ < - r' N r •- 1 "" JN / / / — J — — — — P ^ ^ 1 li / / ~~x. / / V -*-* ; 1 ? ; 5 » - i , fe 1 1 \ fc \ ' '• g \ 1 !S ■ ^ % :n^_ t ' t Y* '/" s ^ k 1 KS f^ , ^ \ H« /^ 1' *1 I 1 *^ t / y \ \ I ^S / ,>' / t, ^o. ^ ! J*^ - i)f / \ ~ ;°- § c5% / s >^ ^ j ^- ^ i te -- '- 1 \ •■>-. r ■§ •i ! \ \ "~ •^*^ -= ^ ♦ % 1 -^„ 1 fl *5 ji- «-* «_ , -< 1 .^- —5 '. - \ i \ \ r , 1- ■»-l«^-4- •»+>4=:^rji~'rH°^ i s - - M 1 I II', \ \ ; ; ^ - ^ ^ "x ¥ € S 10 12 i^ 16 IS 20 22 Z^Haurs. I. Day Specific normal reaction. °— o— Infiltration " ^~ Reddening Wo Iff- Eisner, The Ophthalmic and Cutaneous Diagnoses of Tuberculosis. V. VI. Vn. VIIL JX. X. Day. Taken on 8''' Dec. '07. 20'i' Jan. '08: condition very satisfactory. DIAGNOSIS OF TUBERCULOSIS 115 The fact that in the case of oi'dinary ohservation after twenty-four hours the rapid and superficial reaction may be overlooked is not of very great importance for all practical purposes, since this form of reaction indicates, like the fully negative reaction, an unfavourable prognosis, and the mistake may safely be regarded as nil. For scientific purposes, however, these conditions have to be taken into account. The second departure from the normal type, the late reaction (or rather the late and lasting reaction), has an exactly opposite clinical significance. Its course is as follows : The first evidence of the reaction occurs after about the same space of time as in the case of the others ; the maxi- mum point, however , is not reached quickly, but only very gradually, at the end of the second day, or even later. In sharp contrast to the quick reaction of the manifestly tuber- culous cases, it remains unchanged for days at its maximum point, on an average persisting like that for a whole week. But it has sometimes been found unchanged even after three and more weeks. This course of the reaction is illustrated by curve figures 6, 7, 8 and 9 on p. 118. This type of curve is of profound diagnostic significance. This reaction is only found in those cases in which clinically there are no indications whatever of active tuberculosis. We may maintain with practically absolute certainty that the late reaction represents the type of reaction of cases of perfectly inactive tuberculosis.^ The prognostic conclusions drawn from the course of the reaction, therefore, may be synthesized as follows : If a patient loith initial tuberculosis reacts according to the normal type, toe 7nay make a favourable prognosis, increasing in propitious- ness icith the strength of the reaction within the limits of this normal type. Weak rapid reactions, or even altogether nega- tive ones, point to an unfavourable course of the disease. Strong reactions according to the normal type, in ana- tomically advanced cases, indicate a certain healing tendency, and therefore signify stasis or slow progression. In advanced cases a rapid reaction indicates quick intensification of the disease ; no reaction predicts an early death. ' On this point I differ from Stadehnann, who regards the late reaction as that of healthy individuals, whereas I regard it merely as that of clinic- all y healthy persons. From a practical point of view the diflerence is of no great importance : tlie theoretical reasons for my opinion are expounded in various places. 116 THE OPHTHALMIC AND CUTANEOUS It may be objected to our view of the matter that it is very remarkable that the majority of initial cases should give a strong reaction, and that, on the other hand, most of the advanced cases give an altogether negative reaction or one of but slight intensity. If we think the matter over, this frequent correspondence with the anatomical condition will not appear so strange. Tuberculosis is an infectious disease, which leads to death within a short time if the body offers no resistance (miliary tuberculosis). The great majority of cases in which the apex is affected do not show a rapid course, but one stretching over years, and this resisting capacity in affections of the apex is shown by the reaction of the normal type. In pulmonary tuberculosis of the third stage the resisting capacity of the body very frequently is exhausted — as may be judged by purel}' clinical means — and as an exterior sign of their condition these persons respond nega- tively to the reaction, or give a rapid reaction. But this can be said with absolute certainty : the form the reaction takes is not in any way connected with the anatomical character of the lesion, and it is just those cases which anatomically deviate from the expected type of reaction that are clinically of the greatest interest. And the observation of such cases has shown that our method of prognosis, based on the reaction, is of great value. I wish to call special attention to one point. The strong- normal reaction in cases of tuberculosis does not, of course, in itself guarantee a favourable course of the disease, and cannot definitely indicate the further course. It only shows that at the time at which the reaction is applied the body is still in full possession of all its bodily energy, combating the infec- tion. Combat indicates the possibility of victory, not victory itself. And this capacity to react shows that in such a case the donning of the whole of our therapeutic armour may not be in vain. The limitation referred to lies in the nature of the case : it would be an exaggeration (and every able medical man would think it improbable) to say that any reaction would definitely indicate a favourable course of the disease, especially since we know the influence of external measures on the course of tuberculosis. Moreover, if this were the case, the whole of the importance of the dietetic therapy would fall away, and this would certainly not be a progressive, but rather a retro- gressive step in combating tuberculosis. But even within this limitation, inherent in the nature of sis in the Second Stag Curve Table III. JfoAn Bale, Sons b' Daniehson, Ltd., London. 1 1 L 1 : 1 1 1 1 1 .-.1__L.. r r - f — 1 r r 1 ! 1 ! I 1 i 1 — 1 i 1 f ! 1 Tul )er( II :ulo ^. z sis. ^ay. V. VI I y/. I Taker Die 777. 1 on 2 d on 1 IX, A 2nd De 3th Jar '. Day. c. 1907 1. '08. Conjunctival and Cutaneous Reaction in a case of Tuberculosis in the Second Stage. According to ^olff-Eisner and Teichmann. 5'oA" BaU. s •'-. f= == 1= ?= = r= p= = -t-:- ... N ^; s. ^• .^'-- ._. _.j... ._. ..-, ■»' ■• - r ^ - -" "~ [- h — Y- - r- ■ ... F= 1 — 1 h- - - - — - — — i •§ 1 ^ ! «- -o. ■o- o- .^^ " "*~ ■X- X- ij >». ^ L 4 »- ->H "■^ 1 Schalk. 22"J Dec. 1907. Rapid //. Day in. Day. IV. D. — o — o — .o~ Infiltration /ith uniavourable prognosis. -»— «— x- Reddening Wolff- Eisner, The Ophthalmic and Cutaneous Diagnoses of Tuberculosis. K (••/. V/l. VIII. IX. .V. Da,: Taken on 22""i Dec. 1907. Died on 13* Jan. '08. r John Bale, Sons &■ Da Curve Table IV. nielsson, Ltd., London. -- - -- -- — — --- --- -4— "1 n i 1 ' 1 — 1 — 1 1 — — — - — i t 1 1 i -o- -X- Tu — X ber n cul In R( osis Day filtr atic enii )n V. VI. I Y/. V 'III. IX. X '. Day. Curve Table IV. Wolff-Eisner ana Teichmann. >*, B U, Sm. &■ D„ „;b..n. Ud., LoHdcn. - — 1 -^ [ 1 ' ' - L ! : 1 - 39" t r _^ - \" -■'■ 1 -. -- -- J - _"^ ^ - - - - ^.-. ~i .- -: ■ ■" 1 ... ..:.. 1 '" ^^ -t- ' ^ -V-- - ... ... .. .|.... H |n j i 1 '"1 ^ U - ! 1 11 - ^ I - - - - --i- - 4- - - - - p ^ a s ... -- t s ■ — s t - - .IS §« 1 t" I '^ \ '-'<. j - '- i - - •« / ^' j-k •- ■^ " 0-, * .' •-0 r^ *3 ■1 -f - •»r v^ - — -- N y- *- I- v. 4" ::j ~-*- ^ 1 ^ ! 1 1 1 i ^ LA ? Jl 1 ... -r- ^o ^ 1 1 i 1 1Z n to JS lO 22 Z9Ifonrs I. Day Mdller. Dementia alcoholica. Bronchitis. Tbc. Type II (Rapid reaction with unfavourable prognosis). Wolff- Eisner, The Ophthalmic and Cutaneous Diagnoses of Tuberculosis. V. VI. VU. VIII. IX. X. Day. ¥0' 39' SS' 37" .-JG" *« L .8 ?J (3^ 1^ M n the Second Curve Table V. Stage. John Bale, Sons &' Daniehson, Ltd., London _.. 1 1 - — ( - ■■- ^-_1.4 i i r 1 '— _iA \ \ \ \ \ ---- - 1 i 1 ' 1 1 1 ~i i ; 1 1 1 1 1 CUJ ^. z osi: '>ay. ve 3. of] Fo Dec llo\ . 4 vec 1£ 07 r he V. (minin ;ctic fe VI. I lal lun 'ver; \ -77. V g app( ^ery bj 7/7. jaranc id pro IX. ^ es sub gnosis '. Day. febrile) Curve Table V. Conjunctival and Cutaneous Reactions in a case of Tuberculosis in the Second Stage. According to Wolff-Eisner and Teichmann. Joff^ Bate, s^ns &■ DanirUton, i w — — — — ;= — — — — — — — — — — — " ■ ' r ~ ^ n ~ i ! ' 1 ' , ' , 1 ' ' ; i i 1 ! i I i 1 i i ^ s L^"!^ ' ' !J \ ' J . i" 1 "T --; f-;--^-|- -.4.- .U-y 1 ! t y r'- 1 N..4't';^-" 1" -4^- .- 1 I ! ! ■ -j-'-' •- r- ... - ... 1 -f-r-t- ....... ■{ . ) .i : ! ' ' - i .-4- - -j-|-j- ■ 4... 1 1 ■ ! 1 ■ .. . .. .j.. H i \ ■ ! 1 I- i 1 i 1 "^ / -' - r- - - - - -— - '- ■~ ., : ! 1 1 1 to ■~ - --^ i ~ -' -' " - -- 1 - - - - -- - j- - s ' - - - - 1 ■ i a s , 1 --1- - is ■ i S 1 ! . s - - - -- - -j^ " ' ! __ -- - - — 5: i r" 1 -^ - - - ^« i ' "■ ft :: 4- - — !- i 1 -- - ^s - J 1 i- -i •I' 1 '\ 11 \ 1 -- — / / ^ \ 1 - - * V' / r ij V •' - ^ ^ H /' ^ - - - - '^ \ ■^ rV \ * \ __, - rJ 1 1 ^N ^ 1 1 ' i 2 'r /i 3 !0 IS fi f6 IS W 22 Z"! Hotu\S- /■ Day IL Day //I. Day. _o— o— o— Infiltration Murzyn, Stage II. Type II (rapid reaction with unfavourable prognosis). —* — ^ — >« — Reddening V. VI. VII. VIIL IX, X. Day Curve of Dec. 4. 1907 (minimal lung appearances subfebrile). Followed fay hectic fever; very bad prognosis. Wolff- Eisner, The Ophthalmic and Cutaneous Diagnoses of Tuberculosis. DIAGNOSIS OF TUBERCULOSIS 117 the progress of an infectious disease, the importance of the fact that our analysis of the reaction enables us to make a prognosis based on biological changes needs no elaboration. These important practical results of the prognostic signi- ficance of the reaction are in need of a theoretical founda- tion. We will at once proceed to discuss such a foundation in accordance with modern views, especially with a view of explaining the principles underlying the reaction to the clinician. The complicated theoretical problems which in their turn spring up on the basis of such an attempt, and to which an answer has been found, will be dealt with in the course of this explanation. According to the prevailing view, introduced into science and practice by Wassermann, the tuberculin reaction depends on the combination of the anti-tuberculin substances which are present in the bodies of the tuberculous affected, against the tuberculin introduced. According to this theory, there- fore, one might expect the reaction in the skin and on the conjunctiva to be most distinct and strongest in those cases in which the body is still in full possession of its defences : that is to say, in which the body is best able to produce the assumed anti-tuberculin substances, therefore in the majority of incipient, and in those of the more ad- vanced cases in which the body may successfully offer resis- tance to the progress of the infection. If one assumes that with the piogress of the process of infection the resisting capacity of the body decreases, and that the resisting substances are no longer produced, one can understand why the reaction should become weaker in the progressing cases and in those which take an unfavour- able course, until finally the body of the consumptive, which is altogether incapable of reaction and severely cachectic, in the final stage cannot possibly respond to the irritation pro- duced under ordinary circumstances by a '25 per cent, tuberculin solution. It cannot be denied that these extremely simple theoretical explanations of the course of the reaction would satisfy both the practitioner and the theoretical scientist to the same extent. However, more precise analysis has shown that highly complicated processes take part in the origination of the reaction, and although this complication of apparently simple conditions is to be regretted, yet a simplification is introduced by this very fact, showing that the tubercle bacillus no longer 118 THE OPHTHALMIC AND CUTANEOUS occupies a peculiar position, but t,akes its natural place with reference to the great biological law governing the intro- duction of foreign albuminoid substances. THEOKETICAL PART OF THE PROGNOSTIC SIGNIFICANCE OF THE REACTION. In order to explain these conditions we shall have to go back a little further, to the question relating to the specific nature of the reaction. Everyone will, of course, be some- what startled by the statement that a number of manifestly tuberculous individuals do not respond to either a cutaneous or conjunctival reaction (or rather, as shown above, give so rapid and ephemeral a reaction that it can easily be over- looked). One is very apt to conclude (especially if one dis- regards our latest discoveries) that for this reason the reaction cannot he specific. Clinically speaking, this absence of the reaction is no disadvantage. That the reaction really has a prognostic significance is shown by the fact that, as mentioned above, a clinician like Stadelmann accepts this view, and defends it with great enthusiasm. The prognostic significance of the reaction cannot any longer be disregarded to the same extent as was the case after I first communicated it ; it will become as generally recognised as the diagnostic. Thus it has been proved that the absence of the reaction is of advantage for clinical purposes ; this fact, however, cannot fully do away with all doubt as to its specific nature. In this connection it must be mentioned that also the sub- cutaneously injected tuberculin does not produce a reaction in advanced cases. The question concerning the specific nature of the reaction, therefore, is identical with the question as to the specific nature of tuberculin itself. And that tuberculin has a specific action has been proved by numerous experi- ments, so that it is hardly necessary to further discuss this problem here. To this is added that our investigations have shown that progressive tuberculosis does not give an altogether negative reaction, but that the reaction is of another type, a rapid type, and that it is only absolutely absent in cases of very severe cachexia and in the last days of life. We have already made reference to the Wassermann theory, which is the one that has found the widest acceptance DIAGNOSIS OF TUBERCULOSIS 119 amongst practitioners, maintaining that the tubercuhn reaction is brought about when tuberculin encounters anti- tubercuhn. The meeting of both substances in the tuber- culous centre causes the reaction. The fact that the reaction does not appear after the application of tuberculin may be explained by the fact that anti-tuberculin freely circulates in the blood, where it already meets the injected tuberculin. The tuberculous centre is to some extent surrounded by the anti-tuberculin circulating in the serum, so that no tuberculin can enter it. It is very interesting to know that some time ago Pirquet maintained a similar theory with regard to serum disease, which now' he has accepted with reference to the tuberculin reaction. The reaction in this case, the serum disease, is brought about by the fact that the albumin encounters its anti-substance in the skin, and here it produces the symptoms of the disease. It has always seemed to me that this theory was improbable in several respects ; I could not believe that toxin meeting with an antitoxin could produce a reaction. My experiments rather pointed to the probability that the difference between the clinical phenomena in the repeated injection and those in the first injection is due to the changed conditions of absorption. In this way I regarded the foreign albuminoid substances as analogous to the bacterial albumin, and the bacteriolysine, which appeared after intro- ducing bacterial albumin, I regarded as analogous to the processes appearing after the introduction of albumin, and in this way I formed the theory of the alhuminolysincB, which assumes that, under the influence of the first introduction of albuminoid substances into the body, reaction substances are formed which destroy the albumin in a way analogous to the bacteriolysines, changing the conditions of absorption and liberating poisonous (endotoxin) substances from the albumen. (See A. Wolff-Eisner, Zentr.f. Bakter., vol. 40, No. 3.) These explanations are not irrelevant to the subject, because the same problems have to be discussed in regard to the action of tuberculin. In reference to the Wassermann theory I must mention that, despite many experiments, I have never succeeded in finding an anti-tuberculin in the sense of a substance which is capable of neutralizing tuberculin. The theory concerning the existence of anti-tuberculin as accepted by Wassermann is known to be based on the assumption that tuberculin is capable of forming a complement in unison with 120 THE OPHTHALMIC AND CUTANEOUS anti-tuberculin ; thus, indirectly, it was concluded that anti- tuberculin was present. Now our investigations enable us to fully maintain my former view of the question concerning the action of tuber- culin, or rather to adopt it without alteration with reference to the tuberculin reactions, and yet to fully explain all observa- tions made by Wassermann which in the first instance led to the doctrine of anti-tuberculin. Notwithstanding the fact that our investigations compel us to modify the theory con- cerning the action of tuberculin, we must acknowledge that Wassermann, in his excellent experiments and their ingenious interpretation, was on the right way and has made it possible for us to attempt a solution of these complicated processes. I have the impression that the toxin theory in the science of tuberculosis, as also in that of other infectious diseases, has given rise to numerous diagnostic and theoretical misconcep- tions.^ Especially in dermatology a voluminous discussion has arisen around the question whether certain affections of the skin are due to the tubercle bacillus or to toxins secreted by tubercle bacilli. All these discussions will prove to have been fruitless if our discoveries be confirmed. These show that the action of tuberculin is due to the particles of tubercle bacilli ; in fact, that tuberculin only consists of very minute particles of tubercle bacilli. Tuberculin represents the type of preservable endotoxin. The endotoxin doctrine becomes more and more accepted in clinical science, and the times have passed in which the word "endotoxin" was not very pleasant to hear, because now our knowledge concerning bacterial and albuminoid poisons enables us to clear up many dark problems, whilst at the same time this doctrine con- cerning the indications for treatment affords a basis in the case of the bacteriocide serum therapy. In the case of tuber- culin our curve figures prove that we have to do with the action of dissolving substances. A coincidence however, afforded a direct proof that we have simply to do with bacteriolysine. It is a well-known fact that in times gone by tuberculin contained great masses of tubercle bacilli. Most of the manufacturers of the present day, however, filter the tuber- culin through Chamberland or Berkefeld candles, and in these cases we may regard as excluded the probability that tubercle 1 See A. Wolff-Eisner, Berl klin. Wochenschr., 1907, No. 38. lealthy individual . Curve Table VI. ^oAn Bale, Sons &* Daniehson, Ltd., London. : . -l. - -■ - ~ - ; ' ' ' ■ ' • 1 ' ' 1 : f ■ i 1 1 _ !_ ^_ _;_ _,' _____!_ 1 i 1 ' ! ' 1 1 1 ; 1 I 1 1 1 1 ^ , 1 ! T ; j 1 1 , j 1 1 1 1 ' ! ^ 1 ; i 1 1 1 1 ; i ! ! ' i -I ' - 1 — ' — i — 1 — i — I — i i i 1 ; 1 1 1 ' 1 - - ! 1 l~^ i 1 ! 1 1 1 i t-i-n 1 i ! ' ! 1 i i ' ; 1 1 i 1 H , - ■ i 1 Li ^J 1 1 1 1 i ! i 1 ] 1 1 fr'"T"^—h"f"i .... .... — —•••••• ...■••— •■ — — "♦— . — ,^ ■|- o-ift*^ 0- id- ^ o -O Lo^o. -o- •o ^o« .o.«-o • •o-**o ""-•^ N i ! i ! i •^o 'O \ 1 1 i ' o 1 1 1 1 o ^ 1 V 1 i j i '•^vg^ i 1 1 1 ! 1 i ] >? i 1 1 1 ' 1 ■: ! n 'ubercul osij Day i. V. VI I ^11 I VII IX. ,^ '. Day. Curve Table VI. Cutaneous Reaction with 25% Tuberculin in a clinically healthy individual. Accor iing to Wolff-Eisner a nd Tet chn an n. John B tU. S^ 7 ^ 1 1 j 1 ! ^o ^^ If » i 1 1 ] 1 ■^ / r , ^\ — 7^ 1 _i 1 * r bull ire uit My o' 1 ■ i 1 i i ' 1 1 1 - i,-l <^ , ■ J 1 j _i 1 i i - ^ ». i\ f /I 1 1 ; 1 , ! ■ ' 1 ' - , ^tV / 1 . 1 1 ' 1 1 1 . 1 1 1 ! i , , , 1 1 ' 1 , 1 V. VI. VII. VIII IX. X. Day. Type III. (lasting reaction). Wolff- Eisner, The Ophthalmic and Cutaneous Diagnoses of Tuberculosis. ials. Curve Table VII. John Bale, Sons &- Daniehson, Ltd., Ivndon. fO ( --- — - "- ^ 39 < \_ i ^38 1^ 36 i 1' r 3 ^ t 1 .8 1^ J r" ( . 1 — o— o— o-— 3 — O — o— o-> •o— . o~«~o~ •o-« ^. ^. p«// *"'Ss i i en / :ulo sis. Day V. VI. I ///. I ^III. IX. J\ r. Day. Curve Table VII. Wolff-Eisner and Ttichm .„„ Cutaneous Reactior in clinically healthy individuals >*. a cU.S^ • &-i>.. ..itan. i(rf., / mJn. ---- "V ... - 1 - ....... \'T" '"" " t 1 i "' f: -: ^ ^' '^\ ■ ... ... '" |j7" /, N >-■ '■' ~; -■■-:: "- -H --. -T-- -J-.- " ....p. ^ -,.. N — 1 1 1^ 1 1 1'^ ll . 1 . _j . r- r — — - s - a fe fc s s S: - - - cK - I** -- -- ?^ J l=> -- i cS% •- - A ^i- ■■^ - J / .o- o" J!, ■■-n *, " -o^ ■^ > > *^ \ i°J '* '*■ >, *S =^ <^- s< ^ ir- •* ^ Pi JX- H^ -» -».-«- "— ~* » w— ' '■ ■^ " " ■ --^^ •= ' \ ,/ r v" 1 c. / V r / ~ i /■ ^ ^ e aw 12 n fs , I. Day 21 Zt Hears III. Day V. VI. VII. VIII. IX. X. Dav. Szagum (lysol poisoning), (unsuspected.) n. Cutaneous inoculation. !■' — 5* day. Type III. (Lasting reaction). (I, Tr.. I, I, I.) W o I f f - £ i s n e r, The Ophthalmic and Cutaneous Diagnoses of Tuberculosis. lals • Curve Table VHI. yoin Bale, Sons &- DanieUson, Ltd., London. ■ .. 3i 1 1 . p-|-- ^ 1 «* 1 ^^ a K .. 1' 1 J 8* . . ro^ c ?" "^ ^ J ^ S ST 5; ^ ^ * i^ i b . V -o— o— o— *- D— D— oH o— j7 • '^ cv > n culo sis. ^ay. V. Dismis VI. sed w VII. ithout nil clinica IX. } 1 sym Sr. Day. ptoms. Curve Table VIII. Cutaneous Reaction in clinically healthy individuals. Wolff-Eisner an i Teichmann 3,h„B oU. Son I fr D« „>to„ i(,<., LmJm. »" -. ^" • . 1 -1 ... " .-- "rf la?" ^- ..-■--■^ ... '-■-- - -■ -.: ^^' -,-- '- ''.. ?'" -..+.., .- ,-^ N r 1 '^ >i 1 i J- ni (0 ; aS^ CO ) Mr (0) tii^ COJ f ■ ' -^ ? ^ ft 6 % SB * t .s* N i^ - ■Sw "^■» / •o> '>~^ <- / P ^ y '' ■~o * ^ - - — ~ h^- 5 — — _ :! -^ .o. r o- o- ^^ 5(= o— tz nf; °^ ■^ t>— ^J-, ^ -i- r — — - — — — ^ A - K ^ 3i =T ■* -^ -^ ~ — u - — - - - — * t ' " \ f^ N :-* p.-y i- '- L- _j _j ^ «! ifJ^- __^ . Z ¥ 6 8 10 iZ n I. Day Mittag, Neurasthe //. Day Late and lasting reaction. Wolff-Ei V. VI. VII. VIII. IX. X. Day. Dismissed without clinical symptoms. r, The Ophthalmic and Cutaneous Diagnoses of Tuberculoas. *st Stage. Join BaUy Son Curve Table IX. J b' Daniehson, Ltd., London. H .... — - .._J.... 1 — - M 1 I- ! -- --- -- -- ^•^ b. ;x: 1 [ ■" J s 1 "*2 t si 1 1 5 c ;t-^ ■o^ '*-«. 'O , '"•X ^x- -X. -x^ ■X-, ";?H xSj^ \--^ ^z> '^^nt^/. ot 1 **■ ^- 1 n :ul< ^. 1 >sis V. . E :nd of Jar V. Cur 1. '08 1 VI. / ve of ew lo no ni. I 10* D cal ap] sputun ^III. ec. 19 aearan 1. IX. X 37. ces, su '. Day. bfebri] Curve Table IX. Cutaneous Reaction in a case of Tuberculosis in the First Stage. Wolff- Eisner and Teichmann JoMnS aU. Sm t fs- Da „Vto. UJ., Lmd„n. ' i -i -i- .._!,_. ■ - j - ■-4- - -' ...L.. ..J..L-. L--I--J. -4 - i i 1 - (--^-.-,...U-J...i... 1 ' 1 ( 1 -[-■-- "^ h \ - ^ -v; L.- - \ --1 \-'- 1 '^ -.- *\ \ „ h- -', ""i", /- . - .4'.: :-.J ... ... .... ,., ....... i "T"'" -■ - - " - ■*! 1 - / .. • ... • 1 N 1 1 fcq 1 1 ; i 1 t 1 - . k ^ ~ __ i 1 ' _-J -" 1 " 1 1 i " h - - -^ — 1 1 ^ [ !~ 1 S 1 e - ~ - fe - - fc s - - s * s tj fc 1 ,o' s 1 -o V * 1 \ -o -o -V ■ ' =. -+ 7* " ^ V ' "A " f' -* .■#^ ■■■% * / .+' \ ». t---* ^>. t- - I / / o. o. ... ... ._. -o- --. fca__ «> / ^:°- =- „ ■^ / (" ' ^' - '- -X. -X. « i''' f .^ '"-"^t'^*^ _^9"au-^, ~=. 1 \ \ } 1 1 "~"* N 1 ? / 1 1 V \/ 1 1 1 1 1 1 » « » 16 M iO ZZ Zf Jfom-S I. Day c li t. Stage I. Specific normal //. Day in. Day. going over into a lasting reaction. Wolff- Eisner, The Ophthalmic and Cutaneous Diagnoses of Tuberculosis. V. V!. VII. VIII. IX. .V. Day. Curve of 10'l> Dec. 1907. End of Jan. '08 few local appearances, subfebrile, no sputum. DIAGNOSIS OF TUBERCULOSIS 121 bacilli are to be found microscopically in the tuberculin, except in the case that some filter or other had been some- what defective. The Kuete-Enoch tubercuhn which we employed has, according to the assurances of the manu- facturer, only been filtered through paper filters, and after having centrifuged the product in my own laboratory Daels discovered a great many tubercle bacilli and particles of tubercle bacilh. {Med. KL, 1908, No. 2.) Now the theoretical importance attached to this discovery lies in the fact that this tuberculin has been tested, by order of the Prussian Government, at the Institute for Experi- mental Therapy at Frankfort a. M., and for this reason is identical in action with the other kinds of tuberculin. When, subsequently, our tuberculin was passed through the Berke- feld filter, and in this way the bodies of the bacilli, in as far as they were not of ultra microscopical size, were removed, we found that the action of the tuberculin had decreased as far as quantity was concerned, but had retained the same qualitative properties. From this it is evident that the action of tuberculin is connected with the bodies of the bacilli, but that probably a part of these bodies are present in particles of so ultra-microscopical a size that they pass through the candles of the filter. Already Jadassohn has expressed a similar suggestion in criticizing the theory of the tubercle bacilli toxin of Klingmiiller. The action of tuberculin, tlierefore, is due to the bodies of the bacilli it contains, whether they are present whole, or, on the other hand, are crushed to such an extent that they pass the pores of the filter. In order to understand the action of tuberculin it is more simple to apply the fundamental doctrines of bacteriolysis which have practically been proved, than that one should have to seek for an explanation in the action of albuniino- lysine ; this would, however, not lead to very important differences, since, according to our explanations, it is highly probable that an analogy exists between bacteriolysis and albuminolysis. The particles of tuberculin in themselves do not become active in the animal body, especially because we must assume that to judge by its chemical structure the tubercle bacillus is not easily absorbed, and probably may resist bacteriolysis to a very great extent. It is only the lysine, in this case the bacteriolysine, which brings into activity the substances 122 THE OPHTHALMIC AND CUTANEOUS contained in the particles of the tubercle bacilli, in accord- ance with the ancient medical phrase corpora noyi agiint nisi soluta. It is in accordance with this postulate that a body which has never before come in contact with the products of tubercle bacilli does not react on tuberculin as long as the doses applied are not inany times the size of those which are effective in the case of tuberculous bodies. The bacterio- lysine substances which may be assumed to exist m the normal body are not sufficiently abundant to create an amount great enough for producing effects. Furthermore, when tubercle bacilli substance is introduced for the first time there is no hyper-sensibility whatever, which seems to be necessary before a reaction can occur. From the curve figures it may be seen to what extent the substances, which bring into activity the toxins contained in the fragments of the bacteria, are present in each particular case. As mentioned before, it is probable that in individuals who have never before come in contact with the tubercle bacilli, e.g., in the infant, the tubercle bacilli which are introduced b)' means of the tuberculin remain where they are. At any rate it does not produce a reaction, so that in such cases no reaction bodies — the bacteriolysine substances, the existence of which we have assumed — are present. These bacteriolysine substances are apparently present in different quantities in different stages of tuberculosis. This may be concluded from the fact that the maximum point of the reaction curve is reached in each case after a different period of time. The differences in the amount of bacteriolysine present — in as far as this may be judged from the course of the curve — are not very great in the case of actual tuberculosis, because in all tuberculous individuals the reaction commences after a fairly equal space of time, and only shows differences in the reaching of the maximum point. The difference is, bacterio- lysine between non-infected and infected individuals corre- sponds in all essential points with the experiences obtained in other ways as to the amount of bacteriolysine in infected and non-infected cases. We will here say a few words as to whether the symptoms to be observed clinically in tuberculosis can be explained by the presence of bacteriolysine substances (for as such we regard the much-discussed anti-tubercular substances). We 5. inn. John Rah , Sons K 5:~ Dam Curve Table ■I'sso/i, Ltd., I. on X. don. i i i < 1 \ 1 _ 1 ' 1 1 1 1 1 1 1 1 1 showed exacerbations like shoots, analogous ranees which I, for instance, had found in se- rum disease. 1 1 i 1 , , 1 1 1 1 , I 1 1 1 1 i ! 1 _ . 1 |>t— X -X- ■X- K—t -X- -X— , :-x- -X- X^ ■x-i [-XH -X-) ">v I i ~^ t \ k, •0^ 1 \ \ i 1 °s \ i >. \^ 1 \ ll ! 1 ' t ■ 1 1 j : 1 1 i 1 I : 1 V. D losis 01' V VI. l ni. I \ - -- - -- — ; ' \ ^ _j i- _ " - - / .-4-_ -1 -- ■ij- - s \ - / i .i- .... ^ ,■■ .. ■- ,^ \ \ - "1 "'" / i , N. \ %, \ j i 1 : - ' / x \ ^ , / .'■ V \ \ _L / •f \ \ i -i: -- - / •■., \ - - / ■■■■ ■-.. ^. \ 1 >• ■^ . -■ ; i ■■•■ \ "r _ ^'■^ ;t °v vx -»- '-»r«4,-xt-x-j«. ^ ■!'lr A ;, t ' ♦-■J .- " °";-°r°t°i°* .— 1^ 0.^ .'- .-, _• ul+o- H- -1 -.iv —. -..-.+.-1.4.- ■±:rz -K. «-, ,-x. h«- "-" -X- -'- ■K-1 '"% J V "• 0^ --•J^ ^^/ -■ ! 1 / u 1 \l 'J- / i r 1 _1j 1 I'll — 1 ' '" i 1 ! 1 i 1 i\ - /f\ '- ! ^ r I ' 1< •*x xV / I 1 1 1 1 i ■ ! 1 ' i 1 1 i ... J .■ t e « 70 B It 7f rs so a St-mmiv. I. D„y Hyper-sensibility and lasting reaction curvi //- rioy HI. Day IV. Da} (Dr. T.) Wolff- Eisner, The Ophthalmic and Cutaneous Diagnoses of Tuberculosis. V. I'l VIL 17//. /.v. -V. Day. ds of Tuberculin. Curve Table XI. John Bale, Sons s.^"'<. w " <.,.-»i»JS ......i—> ' 1 1 ' 1 ! •Vv •^, L^^ ^ ! \ ^-•'"'-^ ?^*T. : 1 1 1 ^ L^l-' --'^- S"'°v > JU ^ efe s= •" ^•' , rM 1 i ' 1 ' , , ; , , 1 ; 1 , ; y", Bouchard serum of goats, Auclair serum of fowls, and Maffucci and De Vestea serum of sheep. However, this therapy has not been heard much of recently. Then we come to the long series of investigators who endeavoured to cure tuberculosis by means of tuberculin, and who alternately employed weak and strong doses of tiiis sub- stance. Amongst these, Pfuhl, Kitasato, Bujwid, Spengler, Petruschky, are amongst the most important. Some of the 138 THE OPHTHALMIC AND CUTANEOUS writers on this subject regarded this treatment as conducive to life, whilst others have viewed it in an opposite light. The latter view was especially held by Metschnikoff and Roux and Buchner, who plainly declare that weak doses have no effect whatever, whilst large ones kill the animals sooner than the infection alone. The treatment by means of tuberculin has frequently been modified. Beraneck employs a specific tuberculin prepared with tubercle bacilli obtained from the patients themselves (comp. Krausse, Hanover, Zeitschrift filr Tuherkulose, vol. 2, No. 5). Klebs has carried out most elaborate experiments for the purpose of preparing perfectly innocuous tuberculin, and Landmann worked with a kind of very concentrated tuberculin prepared in a special manner, the so-called ttiberculol. Then we come to the numerous experiments carried out for combating tuberculous infection by means of antitoxic serum. Various ways have been tried for obtaining such sera. Predisposed and other animals were treated partly with tuber- culin, partly with tubercle bacilli. Auclair endeavoured to find antitoxin in the blood of tuberculous fowls, Viquerat in the serum of mules infected with tuberculosis. The latter maintain that they obtained favourable results by their treat- ment ; Eutkowski, on the other hand, obtained unfavourable results. Tizzoni and Cantani immunized guinea-pigs with tuberculin, also fish and horses by means of the new Koch tuberculin (T.-R.). Maragliano's favourable result, as Maxu- tow expresses it, proved nothing more than that guinea-pigs that had been mfected with tuhercuUn could be cured by means of his curative serum, not, however, guinea-pigs into which living tubercle bacilli had been introduced. His experiments {Ann. 1st. MaragL, 1904), at all events, are worth consideration. A great deal of interest attaches to those experiments carried out with a view of obtaining active immunity to tubercle bacilli ; they seem to have the greatest chance of a successful issue. Maffucci and Di Vestea first injected dead, and then living tubercle bacilli, and found that the infected guinea-pigs lived twenty-four days longer than the control animals (seventy-four and fifty days respectively). Babes and Broca combined the treatment by tuberculin with that by tubercle bacilli. They first injected tuberculin prepared from tubercle bacilli of birds, then living" tubercle bacilli of birds, DIAGNOSIS OF TUBERCULOSIS 139 then human tubercuhn, and finally living tubercle bacilli of men. Lastly, Rodet and Granier, and Redon and Chenot, employed tubercular organs of immunization, Bertarelli {Ze7i- tralb. f. Bald., vol. 55, p. 1) subsequently supplying them with the theoretical basis, apparently proving by means of a technique (precipitin technique) — which 1 believe to lead to false conclusions — that the substances of the tubercle can produce independent substances from the anti-substances of the tubercle bacilli. The fact that so great a number of unsuccessful results have thus far been obtained in the specific treatment of tuberculosis is not surprising, in view of the theories developed by us in regard to the bad prospects of the production of antitoxin, and that all these methods only produce hacteriolysin, or at any rate stimulate the formation of hacteriolysin suijstances. From our experiments it is evident that this is not conducive to any important curative results, and on this basis we dis- agree with the view of AVassermann and Citron, to the effect that hacteriolysin substances are present in all cases and in all stages of tuberculosis, but that these substances alone — for reasons fully expounded in their proper place — are not capable of deciding the battle against the tubercle bacilli in our favour. But it becomes all the more clear why there are so many unsuccessful therapeutic results obtained and injuries l)rought about by the tuberculin treatment of tuberculosis. Our experiments show very clearly that injury is done at once when tuberculi)i or derivatives of tubercle bacilli are employed in those cases in which the hyper-sensibility conditions are changed for the worse by their injection. As shown above, an undesirable result may easily be brought about in the treatment with tuberculin ; we do not as yet master the conditions of hyper-sensibility to such an extent that they can be used therapeutically with advantage. In addition, we must mention that the newly-discovered local reactions can also be employed for curative purposes. Nagelschmidt {Deutsche rued. Wochenschr., 1907, No. 40) has employed the cutaneous reaction for curative purposes in cases of local tuberculosis of the skin, and he compares the results obtained by this method with the Finsen treatment, over which it has the advantage of being universally applicable and cheaper. On the place of application a strong reaction occurs which may lead to ulceration, and, healing up, forms a smooth scar. So also Bandler (Verein D.A. in Prague, Milnch. med. 140 THE OPHTHALMIC AND CUTANEOUS Wochenschr., 1907, Ko. 52), who has employed ten drops of tubercuhn locally in lapiis witiiout injurious secondary symp- toms. A local swelling was obtained, the little lupus nodules partly ulcerated, and healed up, forming a smooth scar. In the chapter referring to ophthalmology I have pointed out the possibility of apphnng tuberculin in cases of ophthal- mic tuberculosis for therapeutic purposes. Further results have not yet come in, but it is probable that the possibilities for the therapeutic application of the local reactions are not yet exhausted. When objections were made with regard to the repeated application of the reaction in the Friedrichshain Hospital, I expressed the conviction that the repeated application of the cutaneous and the conjunctival reactions represented the mildest form of tuberculin therapy, and that, if the tuberculin therapy was at all justified, the repeated application of the reaction would not in any way injure the patient, but, on the other hand, might be of therapeutic advantage. The repeated reactions which were applied to the same patients caused our colleague, Miinzer, an assistant of Pro- fessor Stadelmann in the Friedrichshain Hospital, to carry out systematic curative experiments with repeated cutaneous inoculations in patients, based on the view that in this way the production of the anti-substances could be stimulated. Although we think that the production of anti-substances in the sense of antitoxins is not very probable, or at any rate has so far remained without any proof, yet we believe that it is quite possible that bacteriolysins may be produced ; it certainly is a good idea to regard the process of the produc- tion of these bacteriolysins as taking place in an altogether unaffected tissue at some distance from the tubercular centre. As far as systematic experiments in this direction are con- cerned, which only after a long time can give reliable results, I should like herewith to ensure Miinzer's priority. Personally, I am engaged just now in finding a practical therapeutic application of the percutaneous introduction of tuberculin in pulmonar}- tuberculosis and in lupus. This method represents the least harmful tuberculin therapy in existence at the present time. DIAGNOSIS OF TUBERCULOSIS 141 THE HISTORIC DEVELOPMENT OF THE DOCTRINE OF HYPER- SENSIBILITY, OF HYPER-SENSIBILITY DISEASES, OF ALLERGY, AND OF THE LOCAL REACTIONS. The local tuberculin reaction has not been invented by accident, so to speak, but a long history is attached to it. Its discovery only became possible after medical attention had been directed to the enormous clinical significance of the hyper-sensibility phenomenon by the works of Pirquet, Schick, and Wolff-Eisner. First of all, Richet {Soc. de Biol., 1902, p. 170, 1905, 21, i.) discovered a peculiar substance, to be classified with the medus£e (actiniae), the injection of which did not, as had been believed for a long time, produce immunity, but left the body in a state of hyper-sensibility. This condition was called anaphylaxia. For several years this was the only case on record, only supplemented by the discovery of Arthus, to the effect that repeated injections with the serum of horses brought about a condition of hyper- sensibility. Then AVolff-Eisner, after numerous experiments, put the doctrine of hyper-sensibility on a broad basis and published his " fundamental law of immunitj'," maintaining tJiat the injection of any kind of albumin (with the exception of the specially characteristic toxin) produces a condition of hijper-senslhllltij . The fundamental principle observed by him consisted in the fact that all kinds of albumin behave in exactly the same way, inducing a condition of hyper- sensibility, whether the albumin be of animal, vegetable, or bacterial origin. The differences between the various kinds of albumin, according to him, are due to then* purely quanti- tative nature, in the sense that, for instance, serum albumin is to be regarded as comparativelj' harmless, whereas spernia albumin is poisonous to a very high degree, whilst organ albumin takes a place between these two. In an analogous way he differentiated between the poisonous character of the body albumin of the pathogenic bacteria, as contrasted witli the comparatively very insignificant poison of the so called saprophytes {Zentralh. f. Bakt., vol. 87.). The substantial injections of albumin substances in animals was not conducive to a proper study of the clinical conditions prevailing in cases of albumin injections ; apart from all other difficulties, by reason of the fact that the reactions take so rapid a course that sometimes they cause the death of the animal within a few minutes. This naturally stands in the wav of thOroucrh clinical observation. 142 THE OPHTHALMIC AND CUTANEOUS Yet this work, which appeared in 1904, and was severely criticized, and which proclaimed hyper-sensi])ility to be the general principle underlying immunity, constitutes the theo- retical basis of all further clinical work in this direction (the creation of new forms of disease). At the same time Pirquet began to work with the hyper- sensibility reaction, the significance of which (allergy) he recognized from the beginning, and for which he is entitled to merit. As early as 1902 he published a work on incuba- tion. Those who will take the trouble to compare both works will be convinced of the absolute independence of both inves- tigators. To put it concisely, Pirquet has paid most attention to the local symptoms and to man, whereas Wolff-Eisner was more concerned with the more complicated phenomena of hyper-sensibility in animals. This way of putting the matter does not, of course, claim to express the idea fully. There are a good many important points upon which Pirquet and myself differ greatly, and which have to do with the interpretation of the discoveries. On the basis of these investigations new clinical forms of disease, such as those having to do with the absorption of foreign albumin substances, were formed : — 1903, 1904. Serum disease (Pirquet and Schick). 1905. Hay fever (Wolff-Eisner) and Urticaria (Wolff- Eisner). 1907. Re-vaccination (Pirquet). In the case of serum disease^ we have to do with a some- what natural experimental disease which occurs after the introduction of the diphtheria curative serum, as a rule obtained from horses, the clinical appearances of which, after the injection has been repeated, produce very interesting changes in the degree of rapidity and intensity of the reaction, according to definite rules. Hay fever, on the other hand, is a disease which occurs spontaneously, and which is not caused by somewhat experimental measures, and as such forms the prototype so far of a clinical form of disease : the hyper-sensibility of the organism with regard to the introduc- tion and absorption of foreign albumin substance. - Both these hyper-sensibility diseases — the serum disease and hay fever — have been studied clinically so thoroughly that 'Von Pirquet u. Schick, " Die Serumkmnkheit," Deuticke, Vienna, 1905. - A. Wolff-Eisner, " Das Heufieber,'" Lehniann, Municli, 1906. DIAGNOSIS OF TUBERCULOSIS 143 they may be regarded as belonging to the clinically best known forms of disease (the former by Pirquet and Schick, the latter by Wolff-Eisner). Let me point out here that the historical representation of the matter by Dorr in the Wien. klin. Wochenschr., 1908, No. 13, gives an altogether wrong concep- tion of the actual state of things. The thorough study of hyper-sensibility, in its turn, was conducive to a better understanding of the processes going on in cases of bacterial infection. As observed above, Wolff- Eisner based his investigations on the analogous conditions of albumin and bacterial albumin. Hyper-sensibility is unfavour- able for the body in regard to albumin, which is not capable of propagation. Now Pirquet and Wolff-Eisner direct our attention to the fact that under natural conditions the harmful hyper-sensibility of the body, with regard to albumin substance which has no propagating capacity, is of very rare occurrence. Much more frequent is the hyper-sensibility with regard to bacterial infection, and in this case it has all the significance of an extremely important reaction, since it gives a stimulus to the mobilization of the bacteriolytic forces of the body (one of the active factors in combating infection). In this view I differ to some extent from Pirquet, who (" Serumkrankheit," p. 135)) regards the capacity for quicken- ing the reaction as the sole form of immunity in cases of chicken-pox, variola, measles, German measles, whilst I attach to it a general significance, at the same time also believing it to play a part in bringing about bacteriolytic immunity. As regards this latter point, Pirquet, in his work on re-vaccina- tion, has approached my view, being induced to do so by an absolutely indisputable proof adduced by himself, concern- ing the simultaneous appearance of vaccinations applied on different days. Von Pirquet informs me by letter that in a communication he gave in the early part of 1903 to the Academy of Sciences in Vienna, he included (besides the diseases just mentioned) relapsing fever and the repeated injection of streptococci. I have not seen the communication referred to : perhaps it did not appear in print. But it would seem as if Pirquet, in 190.5, in his work on serum disease, had again reduced the number of diseases which he regards as coming into con- sideration in respect to immunity from hyper-sensibility, since here he appears to accept other conditions as governing the bacterial infections. 144 THE OPHTHALMIC AND CUTANEOUS In my publication on this subject iBerl. kiln. Wochenschr., 1907, No. 38) I pointed to the fact that Pirquet's newly- discovered cutaneous, as also my own conjunctival, tuberculin reaction represented a farther proof of the significance of hyper-sensibility in the case of immunity. The proofs — which I believe to be absolutely incontrovertible, being fur- nished by the reaction — I have treated more fully in the theoretical part. It w^as especially the study of the serum disease which induced Pirquet to analyze the symptoms occurring in re- vaccination in the same thoroughgoing manner from a clinical standpoint as also from that of the immunity doc- trine.-^ The observations made on this occasion were the direct cause of the cutaneous application of tuberculin and of the entire development w'hich follow^ed this discovery. It took place in the following manner : In re-vaccination experi- ments he observed a very peculiar condition of hyper-sensi- bility of the skin, quite similar to that occurring in the case of tuberculosis wuth regard to tuberculin. This could not have been foreseen, because in the case of, for example, hyper- sensibility of the organism to pollen albumin the hyper-sensi- bility does )wt affect the skin. On the other hand, after this cutaneous hyper-sensibility had been demonstrated for tuberculosis by so simple an experiment, my thoughts tended in the direction of testing the sensibility of the conjunctiva to tuberculin in a w^ay which would be analogous to that pursued in the case of hay fever. The result has justified the experiment ; the prevailing conditions, however, only allow of a result in cases of tuber- culosis and allied diseases. It was due to these conditions that no satisfactory results were obtained with the cutaneous and the conjunctival reactions in typhoid and other infectious diseases, though they might have been expected according to Pirquet's theories, which I regard as erroneous ; already some years ago I openly declared them false. On account of the non-realization of Pirquet's theories on the matter, outcry arose also against regarding the local tuberculin reactions as specific. The objections raised against these reactions are based upon the following principles : — (1) That tuberculous individuals do not react. (2) That diphtheria, typhoid, coli, and other poisons produce Von Pirquet u. Schick, " Die Seruiakrankheit," Deuticke, Vienna, 1905. DIAGNOSIS OF TITHKRCULOSIS J 45 reactions also in healthy individuals, and tliat, therefore, such a reaction cannot possibly be specific. Now the first objection also holds fi[ood in the case of the subcutaneous tuberculin injection ; it has sufficiently clearly been dealt with in the course of this work. Klieneberger (Deutsche med. Wochenschr., April, 1908) has neither troubled to read my explanations concerning my own method, which he ventures to criticize, nor has he acquainted himself— as he might have done even without studying the literature to any extent — with the fact that a single ophthalmic instillation of tuberculin can only reasonably be compared with a single subcutaneous tuberculin injection. It is a well-known fact that in tuberculous individuals the conjunctival re-instilla- tion of tuberculin produced IQQ per cent, of positive reactions. I will not further trouble the English reader with this matter, which is merely of local significance. I will turn now to the only objection raised against the specific nature of the reaction which can seriously be discussed. It is obvious that a toxin like that of the diphtheria disease produces a local reaction, the poison, when bi'ought in active contact with the skin, exercising a toxic action. It is different, however, with the endotoxin substances of typhoid and coli, which also frequently produce a reaction in healthy individuals. These endotoxin substances, to become active, must first be liberated from the bacteria which have been introduced. This, as we know, happens by the aid of bacteriolysin substances, and these — as has been demonstrated experimentally (Pfeiffer's experiment) — are present to a suffi- cient extent with regard to typhoid and coli. It is therefore possible that in the healthy individual a reaction may occur after the first introduction, increasing the difficulty of the diagnostic application, although it is quite possible that the more exact study of the hyper-sensibility conditions in infected individuals may also in this case prove to be of diagnostic advantage ; the fact that the reaction occurs, however, does not in itself allow us to draw diagnostic conclusions. In the case of the tubercle bacillus the normal bacterio- lysin substances present are not sufficient to effect a solution of bacterial substance, which can occur only in the infected body immunized by contact with the bacteria. This fact corre- sponds with all our other observations concerning the lysis of the tubercle bacillus, proving that the tuberculin reactions are characteristic and specific for tuberculous individuals, and 10 146 THE OPHTHALMIC AND CUTANEOUS for these only. The different kinds of reaction have their origin in the different forms of hyper-sensibiHty. Thus it was simply a direct transmission of experience gathered whilst studying re-vaccination, after Pirquet had to some extent succeeded in localizing the tuberculin reaction, with one blow removing all disadvantages which are attached to the subcutaneous injection of tuberculin. {Added during Bevision. — I must insert here that Max Wolff, at a meeting of the Berliner Ophthalm. Gesellschaft on January 16, 1908, as also in an article in the Berl. Idin. Woch., January, 1908, suggested that it would not be advisable to revive the old fairy tale concerning the injuriousness of the subcutaneous tuberculin injection. One cannot discuss so biassed a point of view. If he reverts to authority to maintain this standpoint, we can adduce more numerous clinicians of probably equal authority who have arrived at different conclusions by the observation of clinical material. Polyclinical material seems to me to be inappropriate for settling the question concerning the subcutaneous and con- junctival reactions.) These disadvantages were done away with by simply intro- ducing minute amounts of tuberculin into the skin (cuta- neously, as opposed to subcutaneously), thus forcing the process to take its course in the skin itself.^ With a caution ' As a matter of fact, the first cutaneous reaction was obtained by the well-known dermatologist KUngmiiller, and by Escherich, who would have been the discoverer of the cutaneous reaction had he not proceeded from an erroneous tlieoretical basis. He was of opinion that the reaction in question could only be obtained in individuals with distinct manifestations of cutaneous tuberculosis. Otherwise Klingmitller ought to have been the discoverer of the cutaneous reaction, in the same way as the ophthalmologist ought to have discovered the conjunctival reaction. The puncture reaction would not have been forgotten even without Pirquet's interference, since Dely employed the puncture reaction for differential diagnostic purposes, and Hamburger, at the same hospital as Pirquet, carried out extensive experiments with puncture reactions. For this reason it is remarkable that Pirquet has never referred to the cutaneous puncture reaction and its close relationship with his cutaneous reaction. As mentioned before, Pirquet had communicated his discoveries to me, and already on May 15 I was able, in conjunction with Baginsky, to confirm the results of his observations. My publications concerning hyper-sensibility, and especially my experi- ences in cases of hay fever, caused me to employ tuberculin for diagnostic purposes. On May 15 I had already a fairly large amount of material at my disposal (more than 100 cases) which had been treated conjunc- DIAGNOSIS OP TUBERCULOSIS 147 in harmony with the difficult material in hand I maintained that further conclusions could onl}^ be arrived at after the investigation of a very large amount of material, the more so since Pirquet, in making a diagnostic use of his cutaneous re- action, had advised caution in treating adults, whilst at the same time I was enabled to observe that differences exist between the cutaneous and the conjunctival reaction, the full explanation of which was not as 3'et forthcoming. Calmette only began to experiment after Pirquet's publication — towards the end of May — and published bis experiments on June 17, as the " ophthalmic reaction " (Acad, des Sciences) ; at the same time he puts my publication down to June 3, thus approaching the period of his own publication, which is not consistent with the actual facts. We have at other places referred to Calmette's tuberculin test, Ike. It is due to Calmette's work that the reaction has been so rapidly introduced into clinical practice. It is an abso- lutely untenable assertion to maintain that I had not made clinical use of this method, and had not recommended it for clinical purposes. (See Petit, p. 26 : " Wolff-Eisner declarait du reste, a ce moment, ne rien pouvoir aftirmer au sujet de sa valeur diagnostique, n'ayant pas encore experimente cette methode.") The second part of the sentence is altogether beside the truth, the first part distorted. A communication still less consistent with the truth of the matter has been given by Citron at the meeting of the Berliner Ophthalin. Gesellschaft of January 16, 1908. I refer to to the report of the meeting in question in the Bed. hlin. Wochenschr., 1908, and my final tivally with 10 per cent, tuberculin. I declared the reaction to be specific, and assigned to it a diagnostic and prognostic significance. I will explain here in a very few words how my experiments with hay fever induced me to at once apply tuberculin to the conjunctiva, as soon as I had heard of the cutaneous reaction. In disagreement with Dunbar, I represent the view that in hay fever we have not to do with the action of toxic substances, but with that of a poison of an albuminous character. The poisons, under the action of lytic substances, are liberated from the pollen. Therefore, in order that hay fever symptoms should appear, it is necessary : — (1) That lytic substances should be present, quickening the absorption by liberating the poison (present in all hay fever patients). (2) That there should be hyper-sensibilit\-, which is produced for pollen albumin in the same way as for all other albuminous substances. The hyper-sensibility differs quantitatively in each case of pollen-sensi- bility. This co7iception of haij fever enabled me to at once transmit the conjunctival reaction ivith tuhercuUn to tuberculosis. 148 THE OPHTHALMIC AND CUTANEOUS reply in the Ver. f. inn. Med., January, 1908. Calmette started his experiments after having read the results obtained by Pirquet (" aussitot qu'il eut connaissance de son [Pirquet's] premier travail"). Since I published my conjunctival reaction together and in the same paper with Pirquet's cutaneous reaction, Calmette ought to have had knowledge of the method, as also Valiee, who on the very same day com- municated to the Acad, des Sciences of Paris his application of the conjunctival reaction in oxen. Calmette maintains he has discovered the ophthalmic re- action independently of me ; but even in this case I have the undoubted right of priority, and at any rate the same right as Calmette, quite apart from the role I have played in developing the reaction practically and theoretically. THE OUTLOOK. The fact that a great number of healthy individuals react to tuberculin — that is to say, show hyper-sensibility with regard to tubercle bacilli derivatives, whicb, according to our view, in the case of re-infection makes it easier to combat the new infection — causes us to critically review the measures taken in reference to tuberculosis prophylaxis. Modern prophylaxis endeavours to prevent infection by keeping the tubercle bacilli at bay. Parallel to this we have the attempts to produce active and passive immunization, and to eliminate, if possible, the diseased individuals in order to obtain by selec- tion a race free from tuberculosis. We can pass over the active and passive immunization in a few words, since the attempts in this direction have so far been without any remarkable results (see Eber, 79, Naturforschervers. ref. intern. Zentralh. f. Tuberk., 1907, p. 115). In the human race a natural selection must take place by the dying off of tuber- culous families, if only partially. In the case of cattle- breeding such endeavours can be carried out on a larger scale. Thus, as mentioned above, the prophylaxis as far as man is concerned can only consist in attempts to diminish the chance of infection. This is done by retaining the sputum, and by teaching hygiene in sanatoria and hospitals, showing the patient how to deal with his sputum, &c. It cannot be denied that in this way a great deal of good has been done, since some of the patients bring out such an enormous number of bacilli that even the men with the most resistance DIAGNOSIS OF TUBERCULOSIS 149 cannot withstand an infection of this sort, and from this point of view we can only welcome the prohibition to spit in waiting-rooms, railway carriages, and other public places. As a matter of fact, it would be a highly desirable measure to absolutely prohibit spitting in all public buildings and resi- dences. I make a special point of these things in order to avoid the misconception that I pooh-pooh these hygienic principles, because I really doubt whether the decrease of tuberculosis observed in recent years is to he attributed to these measures. If we consider that the immunity of man, that is to say the absence of active progressive tuberculosis in most people, depends on an increase in the reactive capacity, i.e., on a state of hyper-sensibility, it would seem to us that if man no longer encountered tubercle bacilli a race would appear which would at once succumb to such an encounter. Veterinary scientists have frequently observed this condition of things in uninfected animal bodies, and also in man such states have been met with. One might mention measles in the case of the Faroe Islanders. We will confine ourselves, however, to tuberculosis. The negro, for instance, does not meet with tubercle bacilli in Africa, and therefore in his native land he is free from tuberculosis ; in other parts of tlie world, however, he exhibits an extraordinary tendency to contract tuberculosis. Ziemann himself, who is one of the best authorities on Africa, assured me that this was actually due to the fact that the negro in Africa does not come in contact with tubercle bacilli. Ziemann has never found tuberculosis in cattle in Africa, and altogether only one case of tuberculosis in a fowl ; he has also observed that, as soon as there is a tuberculous European living in Africa, the negroes living in his vicinity very easily take up the tubercle bacilh. The hygienic measures for destroying the sputum contain- ing the tubercle bacilli are very justifiable ; every civilized man has sufficient opportunity to meet with tubercle bacilli in other ways, to overcome them, and to establish a relative immunity with regard to tuberculosis with hyper-sensibility. Thus if we see that numerous people possess latent centres which may become active under certain conditions of life, and if at the same time we consider that an infection with tubercle bacilli can only be overcome if the outer conditions are very favourable, we find that it is necessary — without depreciating the hygienic measures taken to decrease the 150 THE OPHTHALMIC AND CUTANEOUS danger of infection — to provide hygienic conditions of life in order to combat and put an end to tuberculosis as a disease of the people : good air, good food, good housing. Especially the hygiene of the work contemplated has to be developed ; this has been maintained for a long time past by all practical hygienists in combatmg tuberculosis. Of all these hygienic problems the one concerning diet, in a sense, is the factor which it IS easiest to regulate and enforce. I regard the problem of popular diet as one of the chief factors in the battle against tuberculosis. Since, under the prevailing conditions in Ger- many, this question is not merely one of hygiene, but at the same time one of politics, I will content myself with this reference. Thus the well-known saying of Grancher becomes quite intelligible from the scientific point of view: "Each country combats tuberculosis in its own way ; Germany by means of sanatoria, England with beefsteak and tennis. Personally I prefer the English method ; it is more pleasant and more efficacious." (Quoted from the Deutsche vied. Wochenschr., No. 40, 1907; Schober, Paris letter.) The methods allow of a broad outlook also from a scientific point of view. One of the most important theoretical ques- tions touched by these investigations we have dealt with in the course of this work. A perfect solution of these problems has not been expected up to the present ; further investiga- tion will bring about unexpected results. But these problems which we have been discussing to some extent do not by any means represent all scientific aspects which will be elucidated by this reaction, and clear up most questions of long standing. We will discuss the question of congenital infection with tuberculosis in this outlook, because we are entirely dependent in this respect on statements in the literature on the subject, we ourselves not having been able to experiment on children and infant material.^ After autopsies in children the problem concerning the occurrence of the congenital infection with tuberculosis turned against Baumgarten's views, and even Friednch Franz Fried- mann could only prove the ^possibility of the occurrence of ' Dr. Julius Eitter has promised to carry out experiments in his infant home with the cutaneous and conjunctival reactions simultaneously applied, and will communicate on the subject before long. DIAGNOSIS OF TUBERCULOSIS 151 congenital tuberculous infection under very special experi- mental conditions. The results of the reaction may be regarded as settling this question, by showing that for prac- tical purposes the occurrence of a congenital infection is to be ignored. Prouff, Bue, Sonneville (Petit, p. 75) together have applied the method to fifty-seven newly born infants and have not once obtained a positive conjunctival reaction, although fifteen of the mothers had given a positive reaction. The results obtained with the reaction throw interesting side-lights on the well-known and controversial theory of Behring, to the effect that in infancy tubercle bacilli are taken up by the intestine of the infant which is accessible to the bacillus, and after having been latent for years may cause an outbreak of tuberculosis and consumption. This remarkable theory had no definite clinical basis ; it has not been proved that children who in infancy received any nourishment con- taining tubercle bacilli, being fed with mother's milk (from healthy mothers) or cow's milk boiled, showed any great difference from children fed in any other manner. In reference to this theory the result of the reaction proves that as a rule no tuberculosis — not even a latent one — is produced during the infant stage, and also shows the correctness of the pedia- trists' view that tuberculosis contracted during early childhood always indicates death, and never a latent centre. On the other hand, however, the reaction has shown that Behring was right when maintaining that in an exceedingly large number of individuals tuberculosis is present in a latent condition. (This could only be demonstrated by a form of the reaction which only showed that latent centres were present. Bee our earlier chapters.) After the well-known post-mortem results these facts might have found us more prepared, but we must not lose sight of the different impressions produced by finding old tubercular centres on autopsy, and by finding latent centres m living people who had so far been regarded as perfectly healthy, and perhaps even in ourselves. Clinical importance is attached only to latent tuberculosis in the living individual, not to the post-mortem result. It is likely that the reaction may, at all events to some extent, decide the question concerning the difference between bovine and human tuberculosis, if only after the most preci;ie representation of the observations made in the shape of curves. 152 THE OPHTHALMIC AND CUTANEOUS It is to be foreseen that both kinds of bacteria possess ingre- dients in common (see Detre and Levy, Orvosi Hetilap, 1907, No. 46), so that human tuberculosis will give a reaction on bovine tuberculosis and vice versa. AVhen, by the application of our curve method, we shall at last be able to distinguish between human and bovine tuberculosis, the reaction would enable us to obtain a large amount of material concerning the number and kind of bovine infection, and to decide the question to what extent infection occurs with bovine tuber- culosis by drinking cow's milk unboiled. (See also the chapter on the Differentiating Cutaneous Reaction.) Thus we see that the application of the reaction provides a large field of scientific investigation, and promises even more. The solution of scientific problems is directly or in- directly connected with practical results. The application of the reaction in practice is of still greater practical utility. First of all a golden rule is brought home to us by the reaction, a rule which, from the point of view of the medical man and the national economist, can hardly be emphasized suhiciently ; the so-called receptive individuals in reality are tuberculous already, and are such persons as are combating tuberculosis. Tbe German language [or the English, for that matter — Translator] is hardly pliable epough to reproduce the neatness of Calmette-Petit's way of putting the matter. " Les pre- disposes sont les tuberculeux et non les tuberculisables." Calmette (Petit, p. 120) gives a splendid representation of the action of the generally employed local reaction. It is too tempting a passage not to be given here in full : — " On comprend tout de suite I'interet qu'offre une methode aussi simple et aussi sure pour depister la contagion tubercu- leuse dans les families. " Actuellement, lorsqu'un medecin est appele a donner ses soins a un malade, a quelque classe sociale que ce dernier appartienne, s'il reste entoure des siens, il est extremement diihcile, pour ne pas dire impossible, d'eviter a son entourage toutes les occasions de contagion. Tot ou tard, celle-ci, quoi ([n'on fasse, trouverait a s'exercer, soit par le linge, soit par les particules salivaires emises pendant la toux, soit par les mains ou par les aliments souilles de bacilles. " Les enfants y sont plus particuHerement exposes a cause de leur insouciance et de leur ignorance du danger. Comment reconnaitre chez eux le debut d'une infection qui peut rester latente pendant des annees, ou qui ne s'aggravera bientot qu'a la faveur de reinfections successives et multiples ? DIAGNOSIS OF TUBERCULOSIS 153 " Sans doute, dans quelques cas, f^race ci une surveillance etroite et a I'eniploi judicieux de la methode d'auscultation fine de Grancher, on reussira a saisir les premieres manifesta- tions apparentes du mal. Mais les desordres organiques seront alors trop souvent irremediables. " L'ophtalmo-reaction, periodiquemenL essayee sur cbacun des membres de la famille, peut trouver ici son application la plus utile peut-etre et la plus feconde. Elle donne au medecin le moyen de s'eclairer ; elle lui fournit le meilleur des arguments pour exiger, soit I'isolement du malade, soit, lorsqu'il en est temps encore, I'envoi a la campagne ou le placement en sanatorium de ceux des siens qu'il a deja contamines. " Tous ceux d'entre nous qui s'occupent de lutte sociale antituberculeuse sentent, combien la nouvelle metbode de diagnostic precoce dont il s'agit, va leur etre precieuse pour selectionner ces enfants, nes de parents tuberculeux, mais encore sains eux-memes, qu'il est si urgent de soustraire a la contagion, en les placant a la campagne chez les paysans indemnes de toute tare ! Topbtalmo-rcaction foumira, a cette oeuvre admirable fondee par le regrettc Grancher, ce qui lu, manquait : elle lui permettra d'affirmer la non-existence de lesions tuberculeuses chez ses pupilles. Elle permettra auss, de diriger tout de suite vers les sanatoriums marins ou vers 1 etablissement d'Ormesson les jeunes enfants chez lesquels elle aura revele I'existence de lesions latentes ou insoupconnees cliniquement. On comprend sans peine I'immense benefice social qui resultera forcement de la promptitude avec laquelle des soins eclaires les entoureront ! " Also these apparently extravagant hopes will be fulfilled as soon as humanity shall be able to afford the financial cost of actively combating tuberculosis. The correct interpreta- tion of the reaction holds out a possibility of an early diagnosis to an extent never before conceived, and therefore also the possibility of successes never before attained. It is hardly a point of discussion nowadays that the cure of tuberculosis depends on tbe early diagnosis of the disease, and that the comparatively unsuccessful results obtained in the sanatoria are due to the fact that obvious cases of tuberculosis, wbich are indicated by the appearance of tubercle bacilli in the sputum, can only very rarely be cured. But there is no doubt that, when properly managed, the sanatoria can still be of very great use in combating tuberculosis. One must, however. 154 THE OPHTHALMIC AND CUTANEOUS make up one's mind to admit all individuals giving a positive reaction into the sanatoria, even if no clinical symptoms of any description be found. The specific reaction indicates the presence of tuberculosis in the body, which may be cured with practical certainty if the individual in question is provided with favourable external conditions. Already nearly every hospital patient is examined by means of the local reactions. We may predict a time in which the reaction will accompany man through life. In the schools the reacting pupils will be isolated, and brought to forest schools, and sanatoria, and especially to sea hospitals ; thus will tuberculosis be combated in the schools. It will play a part in industrial hygiene, and it will also give im- portant results in the army, indicating the distribution of tuberculosis among the people generally, at the same time preventing tuberculous individuals from entering the army or the navy. In this way tuberculosis may be further com- bated in the army, although to some extent this occurs already by means of the prevailing measures. The renewed applica- tion of the reaction when the soldier leaves the ranks will show whether tuberculosis has made its appearance during military service. The splendid medical organization of the German Army has already frequently — for purposes of combating epidemics, — made use of a remark emphasized by Schjerning in his opening speech to the last Congress on Hygiene. The same organization can be of the greatest value by simply handing the results of the reaction, together with the clinical con- dition, to the civil registrars and hospital authorities in a manner which has been recently recommended by Pannwitz at the fourth Conference of Tuberculosis Specialists at Berlin. Objections have been raised against the application of the reaction, to the effect that the very sensitive cutaneous reaction would cause fear of tuberculosis with every little cold or ailment, and that hypochondria would be more prevalent than ever. I regret to say that notions of that sort have even appeared in newspapers and have remained uncontradicted, since I did not regard the daily press as an appropriate medium for the discussion of scientific problems. Such objections I do not regard as reasonable, since the same might be maintained with reference to the publication of Nageli's statistics concerning the distribution of tuber- culosis. If it becomes common knowledge that most human DIAGNOSIS OF TUBERCULOSIS 155 beings have successfully resisted tuberculous infection, the recognition of this fact would not by any means produce a genev&l fear of tuberculosis, but would rather tend to put into the proper light the advantages of taking hygienic measures. Bliimel and Clarus reply to these fears that they cannot at any rate be applicable to the cutaneous reaction, since it is not more sensitive than the subcutaneous injection method. At the same time it would seem, they say, as if the public — medical as well as lay — had become more optimistic with regard to pulmonary tuberculosis. They were in a position to judge, being doctors in sanatoria (in Gobersdorf), and could enumerate a good many examples. This optimism is, at all events, just as significant as the pessimism which formerly prevailed, when every lung patient was regarded as a death candidate. The importance of the reaction to life insurance com- panies for forming diagnostic and prognostic opinions is too evident to need further consideration on my part. The reaction becomes of very great importance when we come to deal with the gravest problem of prison hygiene, viz., that concerning prison tuberculosis. The well-known fact that so very many cases of tuberculosis occurred in prison was very puzzling : it was hardly conceivable that the infection with tubercle bacilli could have taken place in the prison itself. Our discoveries concerning the extremely frequent occurrence of latent tuberculosis at once throw light on this problem. Under the influence of mental depression, the latent tuberculosis becomes patent and acute. It would be of very great value if prison doctors applied the reaction to the prisoners on their arrival, and every six months after- wards. Such a course of action would still more decidedly prove the accuracy of our theories, and the influence of external conditions on the course of the tuberculosis. In any case, it would lead to an improvement in prison hygiene, which is often regarded as being already far too good as it is. THE CLINICAL METHODS OF EARLY DIAGNOSIS. The Sputum Investigation. Since the discovery by Koch of the tubercle bacillus in 1882, the presence of tubercle bacilli in the sputum has been regarded as the deciding symptom, not only in the case of sputum investigation, but generally in the examination of 156 THE OPHTHALMIC AND CUTANEOUS incipient pulmonary tuberculosis. We have already men- tioned that the importance of the tubercle bacillus is not so gi'eat as was originally suspected, since its presence does not indicate the very early cases. The Part Played by the Tubercle Bacillus in Early Dia- gnosis. — The importance of the tubercle bacillus for the purpose of an early diagnosis of tuberculosis has become so small that it would seem advisable, as an antithesis to its under-valuation, to point to the extremely great significance of the tubercle bacillus even for the purposes of early diagnosis, and, at any rate, to call attention to the facts recognized by all those who have a large amount of material at their disposal in which tubercle bacilli are to be found ; although it must be confessed that, notwithstanding all experiments in this direc- tion, physical examination has not produced any distinct results, quite apart from the fact that for the numerous general clinicians who, by their great amount of work along surgical and gynaecological lines, do not master the finest methods of percussion to the greatest extent, the discovery of tubercle bacilli will always remain an important comple- ment to their diagnosis, which not even the discoverer of the apical percussion would think it safe to do without. Technique. — The acid-resisting properties of the tubercle bacilli were long regarded as absolutely characteristic. This view cannot any longer be maintained, since numerous other acid-resisting bacteria have become known (Eabinowitsch, Deutsche med. Wochenschr., 1900, No. 16), which may some- times be present in the lungs, or, according to A. Frankel, ("Diseases of the Lungs"), Milchner, and others, in the bronchial cavities. However, this discovery seems to be so rare that it may be ignored for all practical purposes. As far as the technique itself is concerned, I recommend the method of Baumgarten ; that of Gabbet I regard as less appro- priate, since the distance cannot be controlled, and also because the method of Baumgarten enables us to separate a part of the acid-resisting bacteria (smegma) from the tubercle bacilli by means of the 2 per cent, hydrochloric acid alcoholic solution for removing the stain (A. Pappenheim, Berl. klin. Wochenschr., 1898, No. 87). Staining MetJtods of Kronig. — Kronig has recently (Mediz. Klinik., 1907, No. 24, Deutsche Klinik, 1907) recommended a method for staining the sputum, which he regards as pro- ducing the maximum amount of staining. It consists in !)IAGNOSIS OF TUBERCULOSIS 157 heatin^^ the carbolic fuchsin not only to evaporation, but allowing it to boil up several times, in exactly the same way as is done in staining spores. He recommends the use of a special apparatus in which the boiling may take place. The result is a thickening of the tubercle bacilli as usually seen under the microscope. He maintains that the principal object, the increase of the acid-resisting capacity in the tubercle bacilli, is attained by this method. The acid-resisting capacity of the spores frequently remains comparatively small, notwith- standing maximum staining. To those who have some diffi- culty in observing the tubercle bacilli by reason of their thinness, this method will be an invaluable acquisition. Magnifying Dish of Kronig. — I heartily recommend the so-called magnifying dish of Kronig (p. fjlO of the DeidscJie Klinik, I.e.). It consists of a glass dish which is divided into three sections of different transparency (black, milky-opaque, and glass), and a mirror which, in a microscope, has a light underneath. This dish assists considera])ly in finding those particles in tlie spiituiii in which the tubercle l)acilli are most likely to be contained. Furthermore, it is of importance to know how, if there is no sputum, appropriate material for examination of tubercle bacilli can be produced. Sticker {Zentralh. f. hi. Med., 1891, No. 3) has recommended the use of 0"2 gr. of iodide of potassium, in order to bring about a catarrh artificially. A. Frankel has not found that this process is productive of special results. Kronig believes this method to be not without danger, and effects an accumulation of secretion during the night by means of doses of morphine, applying auscultation early next morning. In this way he often obtains sputum containing bacilli. Michaelides and Much have communicated in Brauer's Beitrdge, 1907, vol. 8, Nos. 1 and 2, that a coloured variety of the tubercle bacillus may sometimes be found which, how- ever, can only be detected by means of a modified Gram staining. This species of tubercle bacillus differs only in colour, but is otherwise absolutely analogous with the typical tubercle bacillus, in its cultured properties as well as in animal experiments. According to the proofs put forward^ the occurrence of a species of tubercle bacilli with totally different staining properties must be regarded as demonstrated. In fact, Much has succeeded in showing that the one species may sometimes go over into the other {Bed. klin. Wueliemichr., 158 THE OPHTHALMIC AND CUTANEOUS 1908, No. 14). We must therefore include it here, although it cannot be denied that this discovery will cause a good deal of confusion in the investigation of tuberculosis. We should not forget to mention the earlier studies of Spengler {Deutsche med. IVochenschr., 1907) who also describes various species with very little acid-resisting capacity, as also some very interesting staining methods. Processes of Increase. — Now we come to the processes of increase. This name in reality belongs only to those pro- cesses in which an increase actually takes place in a nutritive medium by the multiplication of some of the bacilli, such as, for instance, is the case when using the nutritive substance of Heyden. But as a rule this designation is applied in a broader sense, and is used for all attempts to demonstrate the presence of the tubercle bacilli by centrifugalizing, &c., the sputum. Very much, however, cannot be attained from a practical point of view with these very carefully thought- out methods, and this is the reason that they are not employed in practice to any great extent. The oldest method is con- nected with the name of Biedert {Berl. Min. Wochenschr., 1886, Nos. 42, 43). The modification of Kronig seems to be the best ; for dissolving the phlegm he uses an amount of O'l per cent, solution of caustic soda, six to twelve times as great as the amount of sputum, both being thoroughly shaken in an Erlmeyer bulb, and heated during the process of shaking. The grosser and as yet undissolved sputum particles, which do not contain any tubercle bacilli, are thrown out by shght centrifugalization, the remaining fluid being centrifugalized together with the tubercle bacilh and the other bacteria for some time, until they are discovered in the sediment. The method of inoscopy of Jousset (see Exudations) may be used in a similar manner in the case of sputum. Animal Experiments. — In all cases in which the presence of tubercle bacilli cannot easily be ascertained, the only method which cannot fail is to experiment on guinea-pigs. By repeatedly washing the sputum, in a way similar to the one pursued in forming an influenza culture, one endeavours to wash away the other bacteria, so as to prevent the animal from dying prematurely of sepsis. The method recently recommended by Bloch {Berl. hlin. Wochenschr., 1907, No. 17), of crushing the regional lymph gland, has provided myself and others with favourable results, so that it is no longer necessary to wait six or eight weeks for the result of the DIAGNOSIS OF TUBERCULOSIS 159 experiment, but so early as after a fortnight it is possible to give a definite conclusion by extirpating the regional glands and examining them for tubercle bacilli. This decrease in the time which must elapse before the diagnosis can be pro- nounced with certainty would naturally enhance the im- portance of the animal experiment. Other writers on the subject obtained results which were equally favourable (see Joannovics and Kapsammer, Berl. klin. Wocheiischr., 1007, No. 45 ; Weber, Med. Ges. in Giessen, ref., Deutsche med. Wochenschr., 1907, No. 47, p. 1976). We must also refer to the method of Marmoreck, who observed that the injection of tuberculin a short time after the injection of tuberculous material caused a considerable increase in the temperature ot the individual concerned (see the Theoretical Part). Lymphocyte Sputa, and their Significance for Early Dia- gnosis. — Recently I have called attention (Verein fiir innere Med., November 4, 1907) to a thus far unknown and very peculiar sputum discovery, viz., the occurrence of Ij^mphocyte sputa, and their significance for the early diagnosis of tuber- culosis. The discovery has been all the more remarkable by reason of the fact that it was so very obvious, and E. Klebs confessed he had never paid attention to the fact, but it came upon him with a flash, and he believes that I was right. Also Fritz Koch (in his referendum in the Milnch. med. Wochenschr.), who has been assistant for a good many years in the clinic for pulmonary diseases, and as such has had sufficient opportunity to observe sputum, declared that the sputa demonstrated by me were those which he had always suspected of containing tubercle bacilli, although he and his colleagues did not know on what basis to explain this instinc- tive view. The fact of the matter is that lymphocytes occur in very large numbers in commencing and still also in advanced tuberculosis. The lymphocytes sometimes constitute as much as 33 to 90 per cent, of the sputum cells. x\nd if the epithelial cells — which arrive in the sputum by desquaniation, and are indicative of quite different things — are regarded as leucocytes and lymphocytes migrating from the vascular passage, the percentage of lymphocytes becomes much larger still. This observation and the frequency of its occurrence are very remark- able, especially if one takes into account the extent to which the sputum from the tubercular centre is mixed with the secre- tions from the nose and throat, and with those from the merely catarrhal and non-tuberculous parts of the lungs. 160 THE OPHTHALMIC AND CUTANEOUS Now these lymphocyte sputa are found in cases clinically suspected of incipient tuberculosis without tubercle bacilli, in cases with few tubercle bacilli, and frequently in those with numerous tubercle bacilli. These last-mentioned cases to some extent prove an existing connection between the lymphocytes and the tubercle bacilli. After this connection has been proved to exist their presence becomes of importance, even in such cases in which tubercle bacilli have not yet been found, thus apparentl}^ representing an early symptom of tuberculosis. I regard the presence of lymphocytes as a direct indication of tubercle bacilli, bound to appear in the sputum at a more advanced stage of the disease, in a manner similar to that of the geologist, who concludes that certain substances must exist where particular formations are found. Indeed, I have succeeded in finding one or two tubercle bacilli in lymphocyte sputa after a very long search, which I was induced to undertake by the lymphocyte character of the sputum. It is remarkable also that in so many advanced cases lymphocytes are found in so high a percentage, especially if one considers that in advanced cases we have to do, as a rule, with mixed infection, and that the bacteria of the mixed infection produce substances which have a positive chemo- tactic action on the leucocytes, and not, like the toxins of the tubercle bacilli, on the lymphocytes. In a discussion at a meeting of the Verein fiir innere Medi- zin, Arnheim mentioned an observation he had made in the case of whooping cough, in which he had found lymphocytosis — after a beginning of leucocytosis — in the sputum as well as m the blood, after employing Pappenheim's pyronin-methyl- green stain. My reply was to the effect that, in using Pappen- heim's stain, the epithelial cells — -w^iich in the advanced stages of whooping cough are desquamated in large numbers — might easily be taken for lymphocytes,^ and that especially in the sputum, which contains a great many degenerated cells, the same difficulties are encountered for distinguishing lympho- cytes from swollen or degenerated polynuclear leucocytes as in the exudations.- In correction of his statement he adds ' A. Wolff, "Lyinphoidzellenleukamie," Zeitsclir. f. Miii. Med., vol. 45, Nos. 5, 6. - A. Wolff, " Unters. iiber Pleuraergiisse," Berl, Min. II ocJiefscJir., 1901, Nos. 34, 45 ; 1902, No. 6. DIAGNOSIS OF TUBERCULOSIS 161 that L. INIicliaclis accounts for the appearance of lyiuphocytes in the sputum by the irritation of the lymphatic apparatus in chronic infectious diseases, the reply to which is that Michaelis is in concord with myself, and has only called attention to the fact that in chronic inflammations an accumulation of lym- phocytes is found in exudations, due to an emigration of lymphocytes. Chrome catarrhs which have not a bacterial cause probably occur very rarely in the lungs, and it is on account of this fact that the presence of lymphocytes in the sputum is of so great an importance as an early symptom of tuberculosis. Whether a chronic bronchitis of a non-bacterial character can produce a lymphocyte sputum (L. Michaelis, " Discussions," Verein f. innere Med., November 4, 1907) is a problem for later investigation, the solution of which will be considerably aided by the new diagnostic local tuberculin reaction in discovering whether or not the presence of lymphocytes in the sputum in cases of chronic catarrh points to a tubercular etiology. For relations existing between the tubercle bacilli toxins and the emigration of lymphocytes and their role as lym- phocyte producers I refer to the chapters on these subjects. The Pathological and Diagnostic Significance of the Pre- sence of Lymphocytes. — I will only mention here that in my opinion the emigration of lymphocytes is not only a character- istic of the toxins of the tubercle bacilli ; there is no doubt that it may also be caused by other processes. I have frequently adduced the investigations of Senator {Virchoto's Archiv, vol. 181) (see Schniitgen, Berl. Idin. Wochenschr., 1907, No. 45), according to which lymphocytes are found in the sediment in chronic nephritis ; I also refer to the discovery of lymphocytes in the lumbar fluid in tabes and syphihs, and to the discovery published by Torday and myself of lym- phocyte exudations, not only after peritoneal tuberculin injec- tions, but also after the injection of tetanus toxin, and to a few other communications. Here I am dealing with the undeniable discovery of lymphocytes in the walls of arteries, which I obtained after cerebral injections of tuberculin and endotoxins. But even if Arnheim were right in maintaining that lymphocyte sputa occur in whooping cough, this would not in the least impair the diagnostic significance of the 1 A. Wolff and v. Torday, Exper. Erzeugung von Lyniphozytenexsudaten Berl. Min. WochcnscJir.:' 'l90i. No. 49. 11 162 THE OPHTHALMIC AND CUTANEOUS discovery, because clinically it does if whooping cough and tuberculosis are interchanged ; besides, accumulations of lymphocytes are of so rare an occurrence that differential diagnostic conclusions may be drawn : that is to say, the presence of all ordinary bacteria and causes of inflammation may be excluded. Since it is just in the sputum that we need not take into consideration any confusion between the lymphocyte producers known so far (tetanus toxin, syphilis, tabes), I am of opinion that the presence of lymphocyte sputum is of very great significance for the early diagnosis of pul- monary tuberculosis, and for this reason I take the oppor- tunity of calling attention to this fact (see reproductions in tables 1 and 2). Technically speaking, no difficulties are attached to the demonstration of the presence of lymphocyte sputa apart from the morphological differential diagnosis. The lymphocytes can easily be recognized by means of the preparation stained by Loffler's methylene blue ; as a rule the simple method of double staining with methylene blue suffices. It is only necessary to consider the possibility of the presence of lym- phocytes, and not merely, in the examination, to rely on the presence of red rods. The Examination for Elastic Fibres. — This process used to be the only one by which a diagnosis could be formed from the sputum with regard to an ulcerating puhnonary phthisis. At the present day the presence of elastic fibres is still of great significance, enabling us to form a picture of the destruction of the pulmonary tissue. The process has passed into oblivion undeservedly, partly by reason of the fact that nowadays the examination of stained objects is preferred. I warmly recommend the modified method of L. Michaelis for the examination of elastic fibres {Ver. f. innere Med., 1901), which enables us to definitely ascertain the presence of even the most minute particles of elastic fibre. Considering the great significance of the results obtained by this very simple method, it is far too little employed. The sputum is spread out in the ordinary way on the slides (not too thin a layer). The slides, which should be dry, but otherwise unfixed, are put for a quarter of an hour into Weigert's solution for elastic fibres. After staining apply a short differentiation with a 1 per cent, hydrochloric acid alcohol solution. The preparations are dried, and the whole of the slide is given a thin cover of cedar oil, so as to avoid DIAGNOSIS OF TUBERCULOSIS 163 using four cover-glasses; then the preparation is thoroughly examined at suspected spots by means of objective No. >]. The elastic tibres present a beautiful steel-blue colour ; they cannot possibly be confused with other formations. THERMOMETRY. I will not dwell long on the subject of thermometry, bi,it in a work on the early diagnosis of tuberculosis thermometry may not be omitted. It is almost certain that a tuberculous- infected organism shows a special change of temperature, and an increase in the temperature may occur with very little cause. Although such increases most frequently occur at night, yet they are not limited to any particular time, and for this reason in such cases the temperature should be taken every two hours (compare Turban, Hess, Cornet, Penzoldt, Hochstetter, Biirgelin, Kronig). Although in other respects everything is done for the patients in Germany, one objects to taking the temperature every two hours. One dare not entrust this to the already over-worked assistants. As a matter of fact the additional work would not be so very much, especially in view of the fact that the diagnosis based on it is of so very great importance. The cause for this objection lies chiefly in the technical difd- culties encountered in the use of the thermometers employed in Germany, which require ten or more minutes for exact measurement, notwithstanding the fact that they are sold as " minute thermometers " ; they are a torment to doctors, nurses, and patients alike. Advantages of the Americ(i)i Thermometer. — I have con- vinced myself of the fact that the American thermometers possess great advantages over those used in Germany, and that they actually allow of taking a certain and reliable measurement in one minute. This is due to the fact that in American thermometers of good manufacture the amount of mercury present in the reservoir is very small, and it takes very little time to become warm. Since the tube has very minute dimensions in consequence, the temperature is not read directly from it, but is viewed through a prism which enlarges the appearance of the mercury. The prism is con- tained in the thermometer itself, its anterior surface being cut prismatically. At first it is somewhat difficult to read these thermometers, but tliese (lilficulties are soon overcome, even 164 THE OPHTHALMIC AND CUTANEOUS by laymen, and then it becomes more easy to read the American thermometers than the German one. It is also much easier to shake them down. Where to take the Temperature. — The anus is the place most appropriate for takin99, vol. 35) and one by myself {Deutsche, med. WocJienschr., 12 178 THE OPHTHALMIC AND CUTANEOUS 1908, No. 6), " Erfahrungen mit der Perkussion der Lungen- spitzen nach Kronig," need to be mentioned. Oestreich had tested the \method experimentally in post-mortem material, and had obtained extremely favourable results. He has been able to determine the presence of tubercular centres the size of cherries : that is to say, if they were not too far removed from the surface. The essential part of Kronig's methods, it seems to me, is the observation that from the point of view of percussion it is not conducive to favourable results to imagine a border at the top of the apex, but that it is much better to project the sound departing from the apex as a broad band over the shoulder-girdle. This does away with the upper and mesial borders assumed by all former methods of percussion. The Kronig method shows very clearly that it is easier, more to the l^urpose, and productive of better results, if light percussion is Fig. 5.^ — RepresectiEg the cocditions existicg in strong and light percus- sion, modified according to Krunig. a, b, Low centre, which can be brought into the cone-shaped sound region only by means of strong percussion, and can be recognized by ear ; c, d, superficial apical centre, which can be recog- nized only by weak percussion — strong percussion would produce the resonance of too great an amount of healthy, full-toned pulmonary tissue. employed, avoiding the resonance of the deeper regions of the apices, and not endeavouring to define the upper border of the lung. It is quite possible to define this border : lately Gold- scheider has experimented with Kronig's special method of per- cussion, applying the softest and most defined percussion ; j^et it would seem to me to be wiser not to endeavour to fix the exact position of the apex, but to project the resonance of the lung to the region of the neck in front (bordered by the sternocleido, clavicle, and cucularis) and behind. DIAGNOSIS OF TUBERCULOSIS 179 The sound-waves in this manner are projected in a way not unlike the projection of Hght-rays by a projection lens, the actual conditions prevailing in the apex being trans- mitted on to an enlarged surface. Now it is true that the determining of the upper lung border constitutes a direct method, whilst percussion of the apex is merely an indirect method ; but this is not necessarily a disadvantage. Just as we frequently prefer the use of a lens to the naked eye, so also in this case it is of more advantage to enlarge the minute surface at our disposal in the direct method of investigation by the method of projection. Refractions — which in direct percussion of the apex, by determining the condition of its highest point, do not differ, or only to the extent of a few millimetres, or are not to be determined at all — showed, when the projection method was applied, a distinct narrowing of the band which passes over the shoulder in front and behind, representing the projection of the pulmonary sound. Outlining the apex, according to Krouig Determining the con- dition of the apex, according to Gold- scheider. Fig. 6. — The arrows indicate the direction cf the sound, showing the difference between Goldscheider's and Kronig's methods of percussion of the apex. Those of my readers who are acquainted with both methods are not in need of any further elucidation ; those who have not heard of them will ask with some surprise wherein con- sists the difference between the production of sound projec- tion and the method of determining the condition of the apex. In case the slight percussion is employed for both no real difference can exist, they will say, and both methods ou the principal efforts in examining the exudations were in the direction of discovering bacilH and of procuring sut'ticient evidence for diagnosing a pleuritic carcinoma. Furthermore, our investi- gations have shown that only in very exceptional cases it is possible to diagnose carcinoma from microscopic examinations of exudations, by reason of the extraordinary polymorphia in the epithelia of the serous cutis in exudations. At the same time very clever experiments were conducted to distinguish between exudations and transudations, espe- cially to decide the question which of the exudations have to be regarded as hsemorrhagic, because it was believed that the diagnosis of haemorrhagic exudations allowed of far-reaching diagnostic conclusions. I will give a short liistorical epitome of its further develop- ment on account of the great similarity between this and the history of the conjunctival reaction. In the year 1898 the Berlin Medical Faculty invited a prize essay on the microscopical, bacteriological, and chemical examination of pleura exudations. I participated in the com- petition, which had to be completed on May 1, 1899. In my investigations I had discovered, to my great astonishment, that in tuberculous exudations lymphocytes are constantly found ; that tuberculous exudations are characterized by this fact, and that they can be readily distinguished from other exuda- tions ; and that at the same time a diagnostic significance was to be attached to this discovery. I could hardly under- stand how it was that before me nobody seems to have made the same observations, and on research I found that these facts were not once referred to in the whole of the litera- ture. I showed the preparation at the time to Leonor Michaelis, who confirmed the accuracy of my observations. Nine years later I was equally surprised to find that the presence of lymphocytes in the sputum had in a similar way been overlooked by all investigators. The prize was awarded to another author, since Gerhardt seems to have regarded my observations as erroneous. Also Professor Franz Konig — in whose laboratory m}^ work had originated, though quite independent of his supervision — disregarded the diagnostic significance of the discovery, and after having originally accepted the work for a chirurgical journal (in a letter of August 30, 1899), he reversed his decision a vear later in the following communication : — 192 THE OPHTHALMIC AND CUTANEOUS " March 10, 1900. " Grave doubts have arisen in my mind concerning the accnracy of the whole theory, but in any case I must say that the cases on which you have been working have been chosen so unfortunately, or have been reported to you in such a way, that the whole matter appears in a different light, &c." Already on August 30, 1899, he had written as follows : — " You are in the extraordinary position of having had before you exudations containing but few bacteria, and I regret that you have not at the same time examined my empyema. . . . This, however, may not allow you to see in this the lack of result of your experiments." Professor Franz Konig has distinctly given me permission in writing (May 25, 1905) to make use of these letters. A year later (June, 1900) Vidal (and Kavaut) in Paris announced the same discoveries. Although at the very first notice of this publication which came to my knowledge I communicated with him on my own discoveries, the written reply to which is in my possession, he has never mentioned ray independent observations in a single instance, nor the theoretical conclusions concerning the active lymphocytosis, the movability of the lymphocytes, kc, based upon them. From 1900 to 1902 articles on the cyto-diagnosis filled the medical journals to the same extent as is the case nowadays with those on the ophthalmic diagnosis, and, being at the tmie a mere student without connections, it was impossible for me, till 1901, to publish the discoveries which I had made a year before Vidal. ■\ \'The Technique of tlie Examination of the Pleura Exuda- tions. — The examination of the exudations by means of the most refined hsematological methods presents some technical difficulties ; these may sometimes be so great that the examina- tion is altogether fruitless. The value of such investigations, of course, can only be judged in so far as they produce positive results. In the examination of the blood one always works with cells taken fresh from the body, and never with the generated cells, which are never found in the circulation in normal cases. The cells of the exudations, on the other hand, have frequently been outside the circulation for weeks ; they are dead, and subject to all kinds of harmful influences. Even if we assume that their presence in the exudation does not DIAGNOSIS OF TUBERCULOSIS 193 damage the cells, yet their physiological death takes place within the exudation instead of in the spleen, the marrow of the bones, &c. We know that the granules of the leucocytes, especially those that are neutrophile, represent the most labile part of the cells which it is most difficult to stain. We are, therefore, not surprised to find that the representation of the neutrophile granules in exudations is very often unsuccessful. The formation of the nucleus allows us to easily recognize the polynuclear leucocytes ; but in exudations there are tivo processes which complicate the diagnosis. The cells, where not fixed at once, may contract to such an extent that they assume the appearance of lymphocytes, the nucleus having a globular appearance on account of the contraction. Or, on the other hand, the nuclei may swell to such an extent that a similarity to lymphocytes is produced in this manner. A process in which the polynuclear cell really acquires more nuclei is more rare, but yet of importance. In that case the cell divides into several parts, the so-called pseudo-lymphocyte formation. It is important to know of the existence of these processes, since when diagnosing the presence of lymphocytes in the exudation very important diagnostic conclusions are incident- ally arrived at ; for this reason one must be acquainted with the kinds of cells with which they might be confused. The pseudo-lymphocytes may be confused with nucleated red blood- corpuscles ; even Vidal has fallen into this error; but in order to avoid this mistake one need only think that it is impossible, if the blood is normal, for nucleated red-blood corpuscles to be present in exudations. They may also be confused with lymphocytes. This is on account of the round nucleus, which is surrounded by a con- centric border of protoplasm ; in most cases, however, the greater breadth of this border in pseudo-lymphocytes prevents confusion. Fortunately the pseudo-lymphocytes, as is clear from their origin, are chiefly found in exudations of the poly- nuclear type, so that the character of the exudation is easily understood, since there are but few cells which could give rise to confusion. One has to note the presence of single neutrophile granules in the triacid method of staining, and also the result obtained with Pappenheim's pyronin-methyl green stain. Eosinophil cells are rarely found in exudations and have no diagnostic significance ; the same may be said of giant cells. 13 194 THE OPHTHALMIC AND CUTANEOUS According to Ehrlich, the lymphocytes are distinguished by the absence of granules. Although Michaelis and myself have described the presence of azure granules in lymphocytes, this does not impair the diagnostic significance of the absence of granulation in most other methods. Since the other leuco- cytes in exudations frequently do not any longer contain visible granules, and since degenerative processes occur in leucocytes (karyorrhexis and karyolysis, pycnosis), the diagnosis of lymphocytes is not always simple. It is v^'ell to employ a number of different methods simul- taneously in the investigation, especially in those cases in v^hich the nature of the cells in question is doubtful (hsematoxylin- eosine, triacid, Loffier's methylene blue, Pappenheim's pyronin- methyl green method, May-Griinwald's stain and the Homan- owski method ; occasionally the glycogen staining with iodium vapour and the vital staining may give further results). We have already mentioned that the epithelia in exudations have a very polymorph nature, and may assume the most peculiar shapes. According to Ehrlich, they too are frequently confused with the large mononuclear cells of the blood, as also by Vidal and a great number of his pupils. These cells are principally found in metapneumonic and rheumatic exudations. Vidal lias pointed out that in ordinary tubercular pleuritis these cells do not occur at all, or very rarely. In many cases the albumin contained in the exudations hinders a thorough investigation, causing an intense staining of the background. It would be well to get rid of the albumin by centrifugalizing the exudation and by pouring off the whole of the remaining liquid which contains the albumin; then a physiological solution of common salt is added, the sediment is disturbed, the centrifuge is again applied and the sediment is brought on to the cover-glasses. Since in a number of investigations, especially if giant cells be present, the addition of water should be avoided, it is well to always prepare a few cover-glasses with the exudation without the addition of water. In preparing the slides one should remember a frequently neglected rule which, for blood preparations, is essential. The cover-glasses should be provided with but a small amount of liquid, so that the cells may dry easily and no damage be done to their morphological structure by a slow process of drying, which would cause more difficulties in the diagnosis. Summary. — In summarizing the results of the examination of exudations, we may distinguish between acutely infectious DIAGNOSIS OF TUBERCULOSIS 195 and chronic exudations. If 50 per cent, or more lymphocytes he found, one may safely diagnose tiiberculosis. Whatever has been said on this subject in the hterature has not in the least affected the final results. As previously mentioned, the accumulation of lymphocytes in exudations is not charac- teristic for tuberculous processes alone. Lymphocytes are also found in spinal fluids in paralysis, tabes, &c. But these diseases can hardly be confused with tuberculosis, and especially in the pleura no process has become known in which lympho- cytes accumulated in the pleura to any extent. Strauss, in the Charite Annals of 1902, mentions the discovery of a lympho- cyte exudation in a patient who did not react on a diagnostic tuberculin injection. It is quite possible that this exudation was not tuberculous, notwithstanding the presence of the lymphocytes. On the other hand, it may also be possible that the cells were mistaken for lymphocytes in one of the ways described above, which may happen even to the expert. Finally, the individual in question might not have reacted on the tuberculin injection even if he had been tuberculous, the actual possibility of which has been frequently referred to in the course of this work. One word more concerning the manner in which these investigations — which, according to the description, may seem to be somewhat protracted and troublesome — are carried out. Ih order to obtain results of scientific value it is desirable that for each case a great number of different methods should be employed. As a rule, staining wdth Loffler's methylene blue suffices to form a perfectly decided diagnosis in cases of distinct lymphocyte exudations. In practice, this method at the same time is so simple that I can recommend its universal application. THE PROTEOLYTIC FERMENTATION OF EXUDATIONS, &C. We will refer to another method giving results almost identical with those obtained in the morphological examination of exudations. Although the results are somewhat less decisive, on the other hand, the application of this method is very much easier, and for this reason it may be employed for quickly discerning between tuberculous and other ulcerations. The process consists in bringing the exudations to be examined on to (coagulated) albumin (sheep serum, bovine serum, &c.) ; non-tuberculous exudations effect a dissolution 196 THE OPHTHALMIC AND CUTANEOUS of albumin through peptonization as a result of proteolytic fermentation, whilst tubercular exudations and tuberculous pus do not cause the albumin to liquefy. The test is carried out at a temperature of 55^ C, in order to prevent a growth of other l^acteria which may be contained in the exudation, so as to avoid liquefaction through bacterial enzymes. The ixiethod may be regarded as an indirect one, and is based on the different amounts of peptonizating ferments con- tained in the various types of leucocytes. We are dealing here with conditions analogous to those observed in staining with guaiacum blue. It is well know^n that the granulated leuco- cytes produce a bluish tinge, whilst the non-granulated cells, the lymphocytes, do not possess this propert3^ In the same way the bone marrow gives the guaiacum a blue tinge in myeloid leucocythaemia, whilst the cells of lymphatic leucocy- thfpmia do not have this effect. There are other methods for determining the contents of pus in proteolytic fermentations, and for distinguishing between tuberculous and other ulcerations. If tuberculous pus is brought into Millon's reagent, a strong thin membrane is formed, which does not give a colour reaction, whilst non- tuberculous pus produces a red tinge. This test corresponds with the biuret test, which is negative in tuberculous pus and positive in all other substances. (Miiller and Jochmann, Miinch. med. Wochenschr., 1906, No. 29; Kolaczek and Miiller, Deutsche med. Wochenschr., 1907, Nos. 7 and 17.) However, tuberculous pus produces a negative result in the biuret test onl}^ when the affected spot has not before been treated with iodoform. This statement at the same time constitutes a criticism on the process. From studies in cyto-diagnosis we know that in tuberculosis lymphocytes are predominant, and since, according to our experience, the lymphocytes are lacking in enzymes, we can understand why, in processes in which lympho- cytes govern the disease pictures, those reactions which are produced by ferments are negative. Now, if in a tubercular process, from whatever cause, an accumulation of leucocytes is produced — for instance, bj'' injecting iodoform — all these ferment reactions will be positive. The coagulated serum fluidifies ; Millon's reagent produces a red tinge, and the biuret test becomes positive. From this we may conclude that only the absence of ferment reactions has diagnostic DIAGNOSIS OF TUBERCULOSIS 197 significance, whilst a positive result may very well occui- in tuberculous affections if an ordinary leucocytosis is produced by chemical means or by mixed infection. Nor does the negative result prove anything definite, and it is obvious that a greater significance is to be attached to the morphological investigation, since the predominance of lymphocytes enables us to draw diagnostic conclusions, which cannot be obtained in the ferment reaction, because the fer- ment reactions are negative only when hardly any perfect leucocytes are present. At any rate, this very easy method may be employed as a control to other methods. INOSCOPY. The method of inoscopy^ is one which is intended to facilitate the ascertaining of the presence of bacteria in an exudation or a transudation, especially of tubercle bacilli, by making use of spontaneous fibrine coagulations. To ascertain whether tubercle bacilli are present in an exudation is some- times so difficult that in cliiiical science the following phrase has frequently been employed : " An exudation without bacilli points to a tubercular etiology." This was on account of the fact that bacilli, especially tubercle bacilli, were very rarely found in exudations. Recently, in a fairly large number oi cases, tubercle bacilli have been found in exudations in animal experiments on guinea- pigs ; negative results of the experiments, however, do not prove anything definite, because it is quite possible that the virulence of the tubercle bacilli may have become impaired. Apart from this the method in question is not very practicable, since results are obtained onlj' after some weeks. By methods of crushing the glands and injecting tuber- culin a few days after the injection of the suspected fluid, as previously described, more rapid results may be obtained in animal experiments, though they are not always absolutely reliable. The method of centrifugali^ing the exudations is frequently unsuccessful because the few bacilli are contained in a viscous ' This interesting method has lately been neglected because it did not provide investigators with safticiently favourable results. Its theoretical basis, hovi^ever, is good, and in scientific investigations in which all methods are applied to elucidate problematic disease pictures it should certainly be emploj'ed. 198 THE OPHTHALMIC AND CUTANEOUS fluid of a fairly high specific weight, so that in centrifugaliza- tion they are not precipitated. Coagulation, on the other hand, fully purifies the exudation of all morphotic constituents, which may be seen from the fact that Jousset, for instance, after injecting the coagulated substance into animals, obtained as many positive (infection) results as Dumany, who injected the coagulated substance together with the serum. Now Jousset gets at the bacilli by artificially dissolving the coagulated substance. After various experiments with alkalies, &c., he concluded that the best process was the artificial digestion, according to the following recipe : — Pepsine (s. d.) .. 1 — -2 gr.v This liquid substance keeps good from three Glycer. ) I to four weeks. It keeps intact the leuco- Acidi hydr. (s. d.) [ - cyte nuclei and the bacteria, and only par- Fluornatrium .. 3'0 | tially destroys the virulence of the tubercle Aqua dest. . . 1,000-0 ^ bacilli. The coagulated substance has to be thoroughly cleansed by means of water, because if serum is present the digestion is very much slower. (This must take place at a temperature of 38° C.) Shake every half-hour ; the process is terminated in two to three hours. Instead of bringing the mixture into the thermostat, one may heat it for a short time in a water bath at a temperature of 50" C. The mixture is now centri- fugalized, and the precipitation is used for streak preparations. For purposes of investigation it is best to employ coagulated substance of several hundred cubic centimetres of exudation. The inoscopy of the blood is performed in a similar way, the only difference being that directly after puncture the blood is brought into 150 to 200 c.cm. of sterile water. Fluid substances which cannot coagulate (urine, transuda- tion of ascites, &c.) are made to do so by adding the blood of horses. The results obtained by this method are as follows : — In twenty-three pleura exudations, seventeen of which were procured from patients not suspected of tuberculosis, Jousset found tubercle bacilli in each case. In twelve exudations of the peritoneum, the etiology of which was unknown, he found eight with tubercle bacilli, of which three, according to the clinical examination, &c., had been diagnosed as alcoholic cirrhosis of the liver. Autopsy revealed tuberculous changes in each of these last three cases. In two cases of acute tuberculosis and in three of subacute tuberculosis he found tubercle bacilli in the blood ; similarly, in two cases of arthritis he found gonococci. DIAGNOSIS OF TUBERCULOSIS 199 Already in the living mdividuiil a process analogous to coagulation is supposed to occur, and probably constitutes an automatic defence on the part of the organism. The exuda- tions containing the greatest amount of fibrine substance probably contain the fewest bacilli. The number of bacilli varies to such an extent that sometimes thousands of them are found, and sometimes only a few after a prolonged search. If the number of bacilli is not very large they are generally found free, although sometimes they are grouped round a leucocyte nucleus, round which they had been grouped also before the digestive process had taken place, and where they have remained, although the protoplasm had been dissolved in the process and only the nucleus had been left. As to their staining capacity, we may say that the bacilli of the exudations are easily damaged when unstained, and for this reason unstaining has to be carefully proceeded with. The author believes that this fact is not due to the digestive process, but is a result of their stay in the pleuritic exudation. The method has the advantage of directl}'^ showing the tubercle bacilli present. The direct bacteriological examina- tion of the exudations has been neglected up to the present only on account of the great number of unsuccessful results. The inoculation of guinea-pigs has this advantage over ino- scopy, that to some extent it gives an idea as to the degree of virulence of the bacilli in an exudation, while inoscopy does not. Under certain circumstances it ma}^ be very im- portant to know whether tuberculosis is very active or less so. For this reason the combination of inoscopy with the inoculation of guinea-pigs may be regarded as a good method for the examination of exudations as to the presence of tubercle bacilli. As far as the urine is concerned, the diffi- culty of distinguishing between pseudo-tubercle bacilli and tubercle bacilli remains, and therefore the method in question is of value only in some cases. 200 THE OPHTHALMIC AND CUTANEOUS LITERATURE. 1. Abrami, A., et Burnet, Et. "Reaction cutanee a la tuberculine chez rhomme adulte," Comptes rendus de la Soc. Biol., T. 62, 1907, No. 25. 2. Adam und Albers-Schoneberg, " Fortscbr. auf dem Gebiete der Ront- genstrahlen," Bd. 10, S. 182. 3. Alexander. Berl.klin. WochenscJir., 1901, 'i>ir. 11— 13. 4. Arloing. Soc. de Biol, 1907, No. 27. 5. Arloing, F. Soc. vied. Hop., 22 Juin, 1907. 6. Soc. de Biol, Nov. 23, 1907. 7. Arnsberger. Milncli. med. Wochenschr., 1907, Nr. 2. 8. Aronade. Med. Klinik., 1907, Nr. 51, "Die kutane Tuberkulinreaktion von Pirquet als Mittel zur Erkennung der Tuberkulose im Kindesalter." 9. Aubaret et Lafon. Gaz. liehd. demed. de Bordeaux, 4 Aout, 1907. 10. "L'Ophtalmo-reaction en Opbtalmologie," Aout, 1907. 11. Aubaret et IMagne. Jotirnal de Mi^decine de Bordeaux, 15 Aout, 1907. 12. Audeoud. Eev. vi^d. de la Suisse rom., Oct., 1907. 13. Axenfeld. " Bakteriologie des Auges," Jena, G. Fischer, S. 232. 14. Baginsky und Neumann. Berl. klin. Wochenschr., 1907, Nr. 9. 15. — — Berl med. Ges., Dec. 8, 1906. 16. Bandler. Wiener klin. Wochenschr., 1907, Nr. 48, "Die kutane Tuber- kulin-Impfung bei Hauttuberkulose" (Verein d. Arzte, Prag, Oct. 30, 1907). 17. "Ver. deutscher Arzte in Prag," il/»"nc/i. vuecZ. T^oc/ie7^sc/Jr., Nr. 52. 18. und Kreibich. Deutsche med. Wochenschr., 1907, Nr. 40, " Erfah- rungen iiber kutane Tuberkulinimpfungen (Pirquet) bei Erwachsenen." 19. Barbier. Soc.dcs Hop., Paris, Dec. 6, 1907, ref. Sem. med., 1907, No. 50. 20. Bartel. Wiener klin. Wochenschr., 1907, Nr. 38, "Der normale und abnormale Bau des lymphatischen Systems und seine Beziehungen zur Tuber- kulose." 21. und Neumann. Wiener klin. Wochenschr., 1907, Nr. 43 ; Zen- tralbl f. Bakter., Bd. 40, S. 4 u. 5, " Experimentaluntersucbuugen iiber den Einfluss von organischen Substanzen auf den Gang der Tuberkuloseinfektion beim Meerschweinchen." 22. Baxy. Bulletin Soc. dc Chirurgie, Paris, 6 Aoiit, 1907. " De I'opbtaJmo- r^action a la tuberculine en chirurgie." 23. Beck, Max. Deutsche med. TFoc7(C7isc7m-., 2. III., 1899, Nr. 9. " Uber die diagnostische Bedeutung des Kocbschen Tuberkulins." 24. Beraneck. " Sur les tuberculines," Comptes rendus de I'Acad. des Sciences, 23 Novembre, 1903. 25. Congr^s international de la tuberculose, Paris, 1905. 26. ■ Reunion des medecins suisses, Juillet, 1907. 27. Bertarelli. Zentralbl. f. Bakter. ,BA. 4.5, B.. 1. 28. Bertherand. "Diagnostic de la tuberculose pulmonaire." These dc Paris, 1899. 29. Birnbaum. " Das Kochsche Tuberkulin in der Gyniikologie und Geburtshilfe. " Gr. Literaturiibersicht. Berlin, Springer. 30. Blache. 4me Congres de la tuberculose, Paris, 1898. 31. Block. Berl. klin. Wochenschr., 1907, Nr. 17, " Ein rascher Nachweis des Tuberkelbazillus im Urin durch den Tierversucb." 32. Bliimel und Clarus. Med. Klinik, 1907, Nr. 50. 83. Boso (Paris). "Pediatric pratique," 1 Novembre, 1907. DIAGNOSIS OF TUBERCULOSIS 201 34. Braillon. Lettres a M. Calmette, Aout, 1907. 35. Brauer, L. " Der therapeutische Pneumothorax," Deutsche med. Wudmtschr., 1906, Nr. 17. 36. " Uber Pneumothorax," Marburg, 1906, Univ.-Buchdruckerei von Job. Aug. Koch. 37. Briihl. Intern. Zenlralbl.f. d. gesatnte Tuberkuloseforschung, 1907, S. 23. 38. Brunetiere. " Ophtalmo-reaction, son emploi en oculistique," Gaz. Iiebd. des Sc. m<^d., Bordeaux, Juillet, 1907, et " Ophtalmologie provinciale," Aoiit, 1907. 39. Buinet. Soc. de Biologic, 1907, " Eeaction cutanee de von Pirquet." 40. Soc. de Biologic, 1907, No. 22. 41. Burnet. Soc. de Biologic, 1907, No. 22. 42. Calmette, A. " Prophylaxie de la tuberculose par la recherche de I'ophtalmo- reaction a la tuberculine," La Clinique, 80 Aoiit, 1907. 43. Calmette, M., Breton, et G. Petit. "Influence des injections de tuber- culine sur la phagocytose 'in vivo' des bacilles tuberculeux," Soc. de Biol., 12 Aout, 1907. 44. Calmette. C. r. de VAcad. des Sciences, 17 Juiu, 1907, " Sur un uouveau precede de diagnostic de la tuberculose chez Thomme par Tophtalmo- reaction a la tuberculine. " 45. C. r. de VAcad. des Sciences, 29 Juillet, 1907, "Sur le diagnostic precoce de la tuberculose par I'ophtalmo-reaction a la tuberculine." 46. Breton, Paiublan, et G. Petit. Presse mM., 19 Juin, 1907, " Utili- sation pratique de I'ophtalmo-reaction." 47. Breton et G. Petit. " Etude experimentale de I'ophtalmo-reaction," Soc. de Biologic, 12 Oct., 1907, No. 28. 48. Carlier. Presse vuhi., 13 Juillet, 1907. 49. Chaillous. Soc. d' Ophtalmologie de Paris, Juillet, 1907. 50. " Prophylaxie de la tuberculose infantile par la recherche de I'oph- talmo-reaction a la tuberculine," La Cliniqice, 15 Aoiit, 1907. 51. Chantemasse. " Serotherapie de la fievre typhoide,'' Paris, Cour d'appel, 1907, Deutsche med. Wochenschr., 1907, Nr. 39. 52. Citron. Berl. med. Ges., 24. VII., 1907 ; Berl. klin. Wochenschr., 1907, Nr. 33, S. 1052. 53. " Uber Tuberkuloseantikorper und das Wesen der Tuberkuliu- reaktion," Berl. klin. Wochenschr., 1907, Nr. 36. 54. Cohn. Berl. klin. Wochenschr., 1907, Nr. 47, "Uber die Ophthalmo- reaktion auf Tuberkulin." 55. Combemale. Vme Cougres de Medecine interne, Lille, 1899. 56. Comby. Presse medicale, 10 Aout, 1907. 57. Bulletin medical, 20 Nov., 1907 ; Sem. viM., 1907, No. 31, p. 371, "Ophtalmo-reaction a la tuberculine chez les enfants," Soc. d'hop. ; Sem. mid., inf. 1907, No. 50. 58. Curschmaun, Hans. " Orthoperkussion," Deutsche med. Woclienschr., 1907, Nr. 52, 59. Delearde (Lille). Presse medicale, 13 Juillet, 1907. 60. Denys. Reinie internal ionale de la tuberculose, Oct., 1907. 61. Derscheid. Lettre a M. Calmette. 4 Juillet, 1907. 62. Desbonnets. Lettre.s a M. Calmette, 22 Juin et 17 Juillet, 1907. 63. Desplats (Lille). Lettre a M. Calmette, 27 Juin et 7 Juillet, 1907. 64. " Ophtalmo-reaction," Journal des Sciences mt'dicales de Lille, 10 et 27 Juillet, 1907. •202 THE OPHTHALMIC AND CUTANEOUS 65. Detre. Budap. kgl. mecl. Ges., 9. XI., 1907. ('>6. Dor, L., jr. 18me Congr. de la soc. fraiK;. d'ophtalm., 190G. 67. Dufour. Soc. VK^d. Hop., 19 Juillet, p. 576. 68. — — Sem. med., 1907, No. 30, " Ophtal mo-reaction a la tuberculiae chez les enfants." 69. Dujon. Lettre a M. Calmette, 30 Juin, 1907. 70. Eisen. " Uber die Tuberkulin-Ophthalmoreaktion," Beitr. z. Klinik d. Tuherkulose, Bd. 8. H. 4, Wiirzburg, Stuber's Yerlag, 1907. 71. Engel und Bauer. Berl. hlin. TFoc/jewscZtr. , 1907, Nr. 37. 72. Eppenstein. Med. Klinik, 1907, Nr. 36, " Uber die Reaktiou der Kon- junktiva auf lokale Anwendung von Tuberkulin." 73. d'Espine. "Tuberculosis," VI., 1907, Nr. 5. 74. Feer. " Die kutane Tuberkulinprobe im Kindesalter," Milnch. med. Wochenschr. , 1908, Nr. 1. 75. Perrand et Lemaire. " Etude clinique et histologique de la cuti- reaction chez I'enfant," Presse med. 70. Ferreira. Lettre a M. Calmette, 25 Aout, 1907. 77. Finger und Landsteiner. Wien. klin. Wochenschr, 1904-1907. 78. Forlaniui. " Zur Behandluug der Lungeuschwiudsucht durch kiinstlich erzeugten Pneumathorax," Deutsche med. Wochenschr., 1906, Nr. 35. 79. Le Fort. Presse med., 13 Juillet, 1907. 80. Francke. " Gefassstreifen, ein Erkenuungsmittel der beginnenden Schwindsucht," Milnch. mfd. Wochenschr., 1907. 81. Francke, Karl. " Die Orthodiagraphie,'" Miincheu, J. F. Lehmann, 1906. 82. Franke, E. " Uber Ophtbalmoreaktiou bei Tuberkulose," Deutsche med. Wochenschr., Nr. 48, 1907. 83. Praukel, A. " Spez. Patbologie und Therapie der Lungenkraukbeiten," Urban u. Scbwarzenberg, 1904. 84. Fuchs. " Lehrbuch der Augeuheilkunde." 85. Gaudier (Lille). Presse medicale, 13 Juillet, 1907. 86. Geugou. Compt. rend. Acad, des Sciences. Paris, 3 Aout, 1903. 87. Goldscheider. " Orthoperkussion,'" Deutsche med. Wochenschr., 1905, Nr. 9 u. 10. 88. Grashey. "Atlas typischer Rontgenbilder von normalen Menscben," Miinchen, Lehmann, 1905. 89. Grassberger und Schattenfroh. Wie7i. klin. Wochenschr., 1907. 90. Grasset et Rimbaud. " Ophtalmo-reaction a la tuberculine," Province medicale, 13 Juillet, 1907. 91. Grasset et Vedel. Acad, de Medecine, 25 Fevr., 1896. 92. "Discussion sur la reaction a la tuberculine ; I'emploi de la tuberculine pour le diagnostic de la tuberculose," Soc. vu'd. des Hapitaux, 7, 14, 21, 28 Juin, 1907. 93. Griffon. "Relations du purpura infantile a forme pseudo-peritoneale avec la scarlatiue," Soc. mi^d. des Hupitaux, 6 Dec, 1907. 94. Grillot. Lettres a M. Calmette, 27 Juin et 13 Juillet, 1907. 95. Gronow, Grafe-Siimisch. IL Aufl., 1 Abt. S. 672. 96. Guerin. Bee. de Med. veterinaire d'Alfort, Aout, 1907. 97. Guerin et Delattre, "Note sur I'opbtalmo-r^action fi la tuberculine," Bull, de la Soc. de Mid. vet., 18 Juillet, 1907. DIAGN(JSIS OF TUBEHCULOSIS ' '203 98. Halipre et Hue. Lettres a M. Calraette, 22 Juiu et 17 Juillet, 1907. 99. Hamburger. Wien. klin. WocJienschr., 1907, Nr. 3G. 100. Hellborn. Berl. klin. W'ochenschr., 1907, S. 893. 101. Holzkiiecht. '' Die riintgenologische Diagnostik der Erkrankungen der Brusteingeweide," Hamburg, 1901. 102. Hutinel. " Effet des injections sous-cutanees de tuberculine chez les enfants tuberculeux," Sem. medicale, 1895, et Congres de la tuberculose, Aout, 1898. 103. loannovics und Kapsammer. " Unters. iiber die Verwertbarkeit neuerer Methodeu zur Diagnose der Tuberkulose in Tierversuch," Berl. klin. Wochenschr., 1907. 104. Kalt. 7 Okt., 1907, Ophthalm. Ges. zu Paris. 105. Kentzler. " Uber differenzierende Kutan-Tuberkulinreaktion," Wien. kli7i. Wochenschr., 1908, Xr. 1. 106. Klemperer, F. Ver. f. inn Med., Berlin, 6. I., 1908. 107. Klieneberger. " Kritische Bemerkungen zur klin. Bedeutung der Ophthalmoreaktion auf Tuberkulose," Mihich-. med. Wochenschr., Nr. 52, 1907. 108. Kliugmiiller. •' Beitriige zur Tuberkulose der Haut," 4rcft. /. Dermal., 1904, Bd. 69, und Perl. klin. Wochenschr., 1903, Nr. 34. 109. Kolaczek u. Miiller. Deutsche med. Wochenschr., 1907, Nr. 7 u. 18. 110. Kraus. " Zur Kenntnis des Erythema induratum," Arch. f. Dermaf., 1905, Bd. 76. 111. Kraus, Lusenberger und Russ. Wien. klin. Wochenschr., 1907, Nr. 45. 112. Krause-Hannover. Zcitschr. f. Tuberk., Bd. 11, H. 5. 113. Kronig. " Uber masimale Fiirbung der Tuberkelbazillen im Auswurf und iiber das Anreicherungsverfahren," Med. Klinik, 1907, Nr. 24. 114. " Friihdiagnose der Lungeutuberkulose," Deutsche Klinik, 1907 (Urban u. Schwarzenberg). 115. " Zur Topographie der Lungenspitzen und ihre Perkussion "' Berl. klin. Woche7tschr., 1889, N. 34. 116. Labbe. " Tuberculino-diagnostic," Gaz. des H6pitaux, Juillet, 1907. 117. Landau. Berl. med. Ges., 8. V., 1907. 118. Lapersonne s. "Pariser Brief," Berl. klin. Wochenschr., 1908, Nr. 2. 119. Leber. "Die Zirktilation und Ernahrungsverhaltnisse des Auges," in Griife-Samisch, " Handb. d. Augenheilk,'" Bd. II., 2. 120. Lemaire (Jules). " La tuberculintest de Calmette et la tuberculine de rinstitut Pasteur, employees pour roculo-reaction. La cuti-reaction a la tuber- culine dans la tuberculose a marche rapide. Remarques sur deux cas de cuti- reaction," Comptes rendus hebd. de la Soc. de Biol., T. 63, No. 29. 121. "Note sur quelques points particuliers de la cutireaction a la tuberculine," Comptes rendus hebd. de la Soc. de Biol., T. 63, No. 28, 1907. 122. Lemerre. Soc. de Pediatric, 15 Oct., 1907. 123. Lenhartz. " Uber Erfahrungen mit der Calmetteschen Ophthalmo- reaktion und der von Pirquetschen kutanen Tuberkulinprobe," Arztl. Verein Hamburg, 29. X., 1907, Miinch. med. Wochenschr., 1907, Nr. 48. 124. Lepine (Jean). " Ophtalmo-reaction en Psychiatric," -Soc. de Biol., 27 Juillet, 1907, No. 27, 28, 29. 125. Lepine et R. Charpenel. " Nouvelles recherches sur I'ophtalmo- reaction chez les alienes," Soc. de Biol., 1907, No. 28. 126. Lesne. "Sur I'ophtalmo-reaction," S. d. H. 6. X., 1907. 127. Lesne et Slarre. La Clinique, 30 .\out, 1907. 204 THE OPHTHALMIC AND CUTANEOUS 128. Letulle (Paris). Soc. Mid. Hop., 28 Juin, 1907 ; Soc. de Biol, 1907. 129. Cit., Soc. de Biol., 1907 ; Sem. mi'd., 1907, No. 27. 130. Levy, Fritz. " Uber konjunktivale Tuberkulinrealition." Verein fur innere Medizin, 16. 12. 1907. Deutsche med. Wochenschr., 1908, Nr. 3. 131. Levy. " Orvosi Hetilap," 1907, No. 46. 132. Lignieres. " Sur un nouveau mode de reaction de la peau a la tuber- culine et son utilisation dans le diagnostic de la tuberculose." Acad, des Sciences, Paris, 28 Oct. 138. Lignieres u. Berger. Acad, des Sciences, 28. X., 1907. 134. Lowenstein. " tjber die intrazellulare Lagerung der Tuberkelbazillen im Sputum und ibre prognostische Bedeutung. " Deutsclie vied. Wochenschr., 1907, Nr. 43. 135. Zeitschr. f. Tnberk., Bd. 5, Heft 6. 136. Mainini. " Haut- und Opbthalmoreaktion auf Tuberkulin." Miinch. ed. Wochenschr., Nr. 52, 1907. 137. Arztl. Verein Miincben, 10. 7. 1907, Miinch. med. Wochenschr. 1907, Nr. 43. 138. Mantoux, Congres de Medecine de Paris, Octobre, 1907, ref. Miinch. med. Wochenschr., 1907, Nr. 49. 139. Marie et Bourillet. " Opbtalmo-reaction chez les alienes," Comptes rendtis hebd. de la Soc. deBioL, 1907, T. 62, No. 27. 140. Marzagalli und Figari, Ann. del Istituto Maragliano, 1904, H. 1. 141. Masenti. " Tuberkulinaugenreaktion," Rif. med., 1907, No. 46. 142. Metraux. Bev. mid. de la Suisse romande, 20 Aout, 1907. 143. Meisser. " tJber die friihe Erkennung der Lungentuberkulose," Therap. Monatsh., Nov., 1898. 144. Micbaelides, Bremer's Beitr. , 1907, Bd. 8, H. 1 . 144o. Beitrdge zur Klinik der Tuherkulosc , Bd. ix. H. 1. 145. Micbeli ed Omarelli. " lusensibilita locale provocata dalla tuber- culina," Acad, di medic, di Torino, 29. XI., 1907. 146. Moller. " Le traitement de la tuberculose dans les sanatoria," Con- gres de Paris, 1906. 147. Mongour et Brandeis (Bordeaux). "Cytologic de I'exsudat dans rOphtalmo-reaction a la tuberculine. " Bulletin medical, 6 Novembre, 1907. 148. Mongour et Lande (Bordeaux). Bulletin medical, 4 Septembre, 1907. 149. Montagnon. " L'Ophtalmo-reaction a la tuberculine," Prov. midicale, 20 Juillet, 1907. 150. Morax. Soc. des Hup., 6. XII., 1907. 151. Moro. Arztl. Verein in Miinchen, 10. 7. 1907, ref. in Wien. klin. Wochenschr., 1907. 152. Moro und Doganoff. "Zur Patbogenese gewisser Integumentverau- derungen bei Skrofulose," Wien. klin. Wochenschr., 1907, Nr. 31. 153. Moussu. " Cuti-reaction a la tuberculine," Bull. Soc. cent, de Med. vtit., 18 Juillet, 1907 ; Sem. mid., 1907, No. 49. 154. Mouton (Lissabon). A Medicina contemporanea, 4 Aout, 1907. 155. Mucb. " Uber die nicbt siiurefesten Formen des Kocbschen Tuberkel- bazillus. Beitr. z. Klinik d. Tuberkulose, Bd. 8, H. 4. Wxirzburg, Stuber's Verlag. 156. Miiller, E. " Proteolytifcbe Fermentwirkuugen der Leukozyteu : neue einfacbe Verfabren zur rascben Unterscbeidung zwiscben tuberkulosen und andersartigen Eiterungea," Deutsche med. Wochenschr., 1907, Nr. 44. 157. Miiller u. Jocbmann. Munch, med. Wochenschr., 1900, Nr. 29. 158. Nagelscbmidt. Deutsche med. Wochenschr., 1907, Nr. 40. Nicolai, A., Bibin. Inaugural Dissertation, Berlin, 1907. DIAGNOSIS OF TUBERCULOSIS 205 100. Nobecourfc et Mantoux. " Ophfcalmo- et cuti-reaction dans la tuber- culose experimentale du lapin," Soc. de Biol., 26 Oct., 1907. 161. Nobl. Wien. klin. Wochenschr . , 1908, Nr. 1, S. 27. 102. Oestreich, "Die Perkussiou der Lungenspitzen," Zeitschr. f. klin. Med., 1898, Bd. .35. 163. Oliner et Terras. " Cuti-roaction ilia tuberculine, Ophtalmo-reactioa." Presse mi'dic, 18 Sept., 1907. 164. Oppenheim. " Uber Hautveriluderungen Erwaclisener im Anschlusse an die Pirquetsche Reaktion," Wien. klin. Wochenschr., 1907, Nr. 32. 165. Painblan. Soc. de Mod. du Nord, Novembre, 1907. 166. Pappenheim. " Befund von Smegmabazillen in menschlichen Lun- gen." Berl. klin. Wochenschr., 12. IX., 1898, Nr. 37. 167. Penzoldt. "Uber den Einfluss der Korperbewegung auf die Tem- peratur Gesunder und Kranker," Milnch. med. Wochenschr., 1899, Nr. 15, und 1903, Nr. 1. 168. Petit. " Le diagnostic de la Tuberculo.se par rOphtalmo-reaction," 1907, Masson, Paris. 169. Pfaundler. Milnch. Ges. f. Kinderheilk., 14. VI., 1907, ref. Mon. f. Kinderheilk., Bd. 6, Nr. 3. 170. Pforinger und Bunz. "Uber das Mass der Bewegung bei der Behand- lung der Lungeutuberkulose," Milnch. med. Wochenschr., 1907, Nr. 2. 171. Pick. Darstellung der Antigene, in Kraus-Levaditi, " Hdb. der Technik u. Methodik der Immun-Porsch.," Juni, 1907, S. 364. 172. Pigger. " Kiinstlicher Pneumothorax u. opsonischer Index," Beitr. zur Klinik d. Tuberkulose, Wiirzburg, Stuber's Verlag, Bd. 8, H. 4. 173. V. Pirquet. " Diagnostische Verwertung der Allergic," Wien. klin. Wochenschr., 1907, Nr. 41, S. 1271. 174. "Der diagnostische Wert der kutanen Tuberkulinreaktion bei der Tuberkulose des Kindesalters auf Grund von 100 Sektionen," Wien. klin. Wochenschr., 1907, Nr. 28, S. 1369-1374, und Wien. vied. Wochenschr., 1907, Nr. 28. 175. 8. v., 1907, Berl. med. Gesellsch. : "Uber Tuberklinimpfung, 15. v., 1907. Berl. med. Gesellsch: "Demonstration zur Tuberkulindiagnose durch Hautimpfung," Berl. klin. Wochenschr., Nr. 22-23. 176. "Tuberkulindiagnose durch kutane Impfung," Berl. klin. Wochenschr., 20. Mai, 1907. 177. "1st die vakzinale Friihreaktion spezifisch ? " Wien. klin. Wochenschr., 1906, Nr. 47. 178. "Der diagnostische Wert der kutanen Tuberkulinreaktion bei der Tuberkulose des Kindesalters auf Grund von 100 Sektionen," Wien. klin. Wochenschr., 1907, Nr. 38. 179. " Die kutane Tuberkuliuprobe," Verh. der 24. Vers. d. Ges. f. Kinderh., Dresden, 1907, Vers, der Naturf. u. Arzte. 180. ProufE. Lettres a M. Calmette, 25 Juin, 4 Juillet, 1907. 181. Putreys et Stiennon. Soc. de Biol., 1907, No. 27. 182. Rabinowitsch. "Befuud von siiurefesten tuberkelbazillenahnlichen Bakterien bei Lungengangraa,"DeM^sc/^e med. WocJienschr., 19. IV., 1900, Nr. 16. 183. Berl. klin. Wochenschr., 1907, Nr. 2. 184. Railliet. Revue qin^rale de la Tiiberculose, .\out, 1907. 185. Rappaport. Milnch. med. Wochenschr., 1906, Nr. 5. 186. R^non. " Sur lophtalmo-reaction." Soc. des Hop., 6. 12. 1907. 206 THE OPHTHALMIC AND CUTANEOUS 187. Richet. Soc. de Biol, 1902, p. 170. 188. Rieder. " Zur Diagaose der chronischeu Luageutuberkulose durch das radiologischeVerfahrea," Fortschritte auf deiii Gebiete der Runtgenstrahlen, VII., S. 1. 189. Rosenau uud Anderson. Washington Government Printing Office, 1906. 190. Rotch und Floyd. " Der opsonische Index und der Tuberkulintest in der Diagnose der Behandluug der Friihtuberkulose," Journ. Am. Med. Assoc, Chicago, No. 8 ; ref. in Miinch. uied. Wochensclir., Nr. 48. 191. Sabrazes et Duperie. Gaz. hebd. des Sc. med. de Bordeaux, Juillet, 1907. 192. Sahli. " Traitement de la tuberculose par la tuberculine," Lemoigne, editeur, 1907. 193. Salvolini. " Tuberkulin-Augeureaktion," Rif. med., 1907, No. 42. 194. Schenk u. Seifert. "Die diagnostische Bedeutung der Ophthalmo- reaktion bei Tuberkulose," Miinch. med. Woclienschr., 1907, Nr. 46. 195. " tJber die diagn. Bedeutung der Koujunktivalreaktion," Z)e^^tec7^e med. Wochensclir., 1898, Nr. 2. 196. Schick. Vers. d. Naturf. Arzte, Kassel, 1903. 197. Schirmer. " Die sympathische Augeuerkraukung," Grafe-Saniisch, "Handb.,"Bd. 6. 198. Schleissner. " Uber d. allergische Reaktiouals Hilfsmittel der Diagnose bei Tuberkulose im Kindsalter," Verein d. Arzte in Prag, 30. X., 1907, ref. in Wien. klin. Wochenschr., 1907, Nr. 48. 199. Schmidt, Ad. " Zur Behandl. der Lungenphthise mit kiinstlichem Pneumothorax," Deutsche med Wochenschr., 1906, Nr. 13. 200. Schniitgeri. " Die BeschafEenheit der im Harn bei Morbus Brightii vorkommenden Leukozyten," Berl. klin. Wochenschr., 1907, Nr. 45. 201. Schrotter. " tJber Anzeigepflicht bei Tuberkulose," Wien. klin. Wochenschr. , 1907, Nr. 38. 202. Schubert. " Ophthalmoreaktiou bei Tuberkulose," Gesellsch. f. Natur- u. Heilkunde Dresden. Miinch. med. Wochenschr., 1907, Nr. 50, 203. Senator. " Farbenanalytische Uutersuchungen der Harnsedimeute bei Nephritis," Virchoio's Arch., Bd. 131, 3. Heft. 204. Sicard. " De I'emploi de la tuberculine pour le diagnostic de la tuber- calose," Soc. vied. Hup., 28 Juin, 1907 ; Sein. med., 1907, p. 323. 205. Sirakoff. Zeitschr. f. Tuberkulose, Bd. 11, H. 5. 206. Sirot. Congres de la Tuberculose, Aoi\t, 1898. 207. Slatineanu. Bull. List. Past., 30 Aout, 1907, et Rec. de Med. ve'tcr. d'Alfort, August, 1907. 208. Souques. " Sur I'ophtalmo-reaction," Soc. des Hup., 6. 12. 1907. 209. Soulier. Bulletin mi'dical, 14 Aout, 1907. 210. Stadelmann. " Uber kutane und konj. Tuberkulinreaktion." II. Tail. Klinische Beobachtungen. Verein f. inn. Med., 6 Jan. 1908. 211. Stephenson (Sydney). "The Calmette Serum Reaction in Ophthal- mology." Brit. Med. Jotirn., Oct., 1907, p. 2442. 212. Sticker. " Beitrag zur Diagnostik der tuberkulusen Lungenaffektionen und zur regionaren Beeinflussung derselben," Zentralbl. f. klin. Med., 1891, Nr. 3. 213. Stock. "Tuberkulose als iitiol. chron. Entziindungen," Gi'iife's Arch. Bd. 66, 1. 214. Striimpell. " Lehrb. der iunereu Medizin," Vogel, Leipzig. 215. Stiirtz, Metz, 4. Vers, der Tuberkulose-Arzte, 24. u. 25. Mai, 1907. 216. Treupel. Miinch. med. Wochenschr., 1908, Nr. 2. DIAGNOSIS OF TUBERCULOSIS 207 •217. Uriarte. Semana Mihlica, Nr. 33, 1904. '218. Valine. " Siir un noiiveau precede de diagnostic experimental de la tuberculosa," Acad, des Sciences, 3 et 17 .Juin, 1907. 219. Vierhoff. Deutsche med. IFoc/icJisc/ir., 1907, Nr. 15. •220. Waele. Soc. de Biol., Bd. 61, Nr. 28. 221. Wassermann u. Bruck. Deutsche vied. Wochenschr., 1906. 222. Milnch. med. Wochenschr., 1906, Nr. 49. 223. und Citron, Zeitschr. f. Hi/g., 1906, Bd. 52, S. 238. 224. Deutsche med. Wochenschr., 1906, S. 449. 225. Weber. " Primiire Impftuberkulose dor Mundschleimhaut,'' Med. Ges. in Giessen, ref. Deutsche med. Wochenschr:, 1907, Nr. 47, S. 1976. 226. Weichardt. Zentralbl. f. Bakt., Bd. iO 227. Weinberger. "Atlas dor Radiographie der Brustorgane," Wien, 1901. 228. Wertber. Ges. f. Natur- u. Heilk. Dresden, ref. Milnch. med. Wochenschr. , 1907. Table I. Ocular 4. '/i« Immersion. Ivymplioiyte sputum with tubercle bacilli. Prot. :J797. Ocular 5. '/n bom. Immersion. Lymphocyte sputum in suspected apices. Prot. 3771. Ocular 2. ' u horn, immersion. S))utum with lymphocytes and tubercle bacilli. #• ^ • • •ivy • m Ocular 5. '/i« immersion. Pleural exudation containing lymphocytes. Loffler. Lymphocyte sputa and their diagnostic significance (see text.) Wol ff- Eisne r, The Ophthahnic and Cutaneous Diagnoses of Tuberculosis. I'lhn Balp, Sons & Daiiulssim. Ltd., London. Tabu II. 01 & ^) '^'«^v Ocular 2. '/n Immersion. Tbc. Lymphocyte exudation. Tracid. t cells. Giant cells Giant ci :::a5Sfe.-.. OjO O / o 4 riant cells. Papule of lasting reaction produced by the cutaneous application of tuberculin; according to a preparation kindly lent by Herr Daels. Wolff-Eisner, The Ophthalmic and Cutaneous Diagnoses of Tuberculosis. John Bale, Sons & Danielsson, Ltd., London. L 006 775 328 5 L 006 775 328 5