'NEUMOTHORAX TRB^MBNT OF 'ULMONARY TUBERCULOSIS CORNELL UNIVERSITY. THE THE GIFT OF ROSWELL P. FLOWER FOR THE USE OF THE N, Y. STATE VETERINARY COLLEQE. 1897 RC 31 1.3.P4Rg^"'™"»V Library Cornell University Library The original of tliis book is in tlie Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924000243596 THE PNEUMOTHORAX TREATMENT OF PULMONARY TUBERCULOSIS Puhlished hy the Joint Committee of Henry Frovde and Hodder 22 ''2 23 23 CONTENTS CHAPTER III PNEUMOTHORAX APPARATUS Generai- Features The Manometer The Gas to Use Tubing .... Needles Preparation of Apparatus for Use CHAPTER IV TECHNIQUE OF INITIAL OPERATION Nursing Home or Private House Preliminaries ..... Selection of Sites for Puncture The Operation ..... Preparation of Patient Preparation of Instruments, and Apparatus Position of Patient .... Passing the Needle .... Manometric Response Entrance of Gas .... Amount of Gas at first Operation Manometer Readings Removing the Needle Effect on Chest Signs Hints for First Operation . Difficulties attending Initial Operation Absence of Manometric Response Incomplete ^lanometric Response Negative Pressure without Respiratory Flue tuations ..... Pleural Pockets .... Puncture of the Abdominal Cavity CONTENTS IX CHAPTER V EFFECT OF PLEURAL ADHESIONS: PARTIAL AND COMPLETE PNEUMOTHORAX I'AQIi Partial Pneumothorax . Ineffectual, with Example . Effective, with Example Complete Pneumothorax, with Example Diagnosis of Adherent Pleura Reduction of Adherent Pleura . CHAPTER VI REFILLS Intervals at Early Stage Operation of Refilling Manometer Readings Intrapleural Pressure Respiratory Fluctuations Re-expansion of Lung Condition of Mediastinum Putting in Gas Early Fillings, Quantity and Pressure of Gas Complete Collapse Its diagnosis from Radioscopy ,, ,, ,, Physical Signs ,, ,, ., Symptoms Intervals at Later Stages . Intrapleural Pressures Complete Pneumothorax Partial Pneumothorax Effect of Altitude on Gas Pressure CONTENTS CHAPTER VII PHYSICAL SIGNS IN ARTIFICIAL PNEU3I0- THORAX 82 CHAPTER VIII MENT AND ITS TI :rmi NATI ON Disappearance of Symptoms 88 Temperature and Pulse 89 Cough and Expectoration 90 Weight 92 Clinical Recovery 92 Anatomical Recovery . 94 Termination of Treatment 97 Partial Pneumothorax 97 Complete Pneumothorax 97 Re-expansion of the Lung . 98 CHAPTER IX PNEUMOTHORAX AS PALLIATIVE TREAT- MENT 101 CHAPTER X EFFECTS OF PNEUMOTHORAX ON THE THORACIC ORGANS The Mediastinum . Open Pneumothorax . Closed Pneumothorax The Diaphragm 104. 104 105 106 CONTENTS XI The Compkessed Lung The Functioning Lung The Lung Circulation Heart and Circulation PAGE 108 111 114 116 CHAPTER XI PLEURISY AS A COMPLICATION Prevalence Causation Dry Pleurisy Pleurisy with Effusion General Classification Purulent Effusions Chemistry of Effusions Cytology Bacteriology Mayer's Classification . (1) Exudates due to Tuberculous Pleuritis (2) Exudates of Acute Infectional Nature (3) Exudates due to Perforation of Visceral Pleura . (■1) Exudates of a Special Type : tlie com monest Form . Effects of Pleurisy Beneficial Effects Unfavourable Effects Sybiptoms Effects of Effusion on Manometer Readings Physical Signs Dry Pleurisy Effusions . 120 121 123 123 123 125 127 129 129 130 130 130 130 130 131 131 133 134 136 138 138 139 Xll CONTENTS Treatment Chronic Effusions Acute Serous Pleurisy Acute Purulent Effusions PAaa 140 141 142 148 CHAPTER XII GAS REPLACEMENT Instruments Needed The Operation .... 145 145 CHAPTER XIII THE ACCIDENTS OF PNEUMOTHORAX TREATMENT Pleural Shock . 151 Symptoms . 151 Conditions of Appearance . 152 Prophylaxis 154 Treatment 155 Gas Embolism 156 Mechanism of Gas Embolism 156 Intravenous Pressure 157 Source of Gas Embolism . 158 Alveolar Air 158 From the Manometer Tubin a; 158 Volume and Nature of Gas 159 Symptoms 160 Prophylaxis 160 Treatment 161 Surgical Emphysema 161 Superficial 161 Deep or Mediastinal . 162 CONTENTS xili rxc.K Perforation of THE Lung . 162 Symptoms . 163 Prognosis . . 168 Prophylaxis . ] 61. Treatment . 1 61. CHAPTER XIV GENERAL RESULTS OF TREATMENT 166 Bibliography 175 Intdex 181 LIST OF ILLUSTRATIONS PIG. PAOK 1. Pearson and Lillingston Pneumothoeax Appabatus 26 2. Weiteb's Tkocab and Cannula .... 34 3. Lillingston's Needle ...... 36 -t. Tempebatube Chaet showing Loss of Fever . . 58 .J. Pabavertebbal Triangles and Quadrangular SuBRESONANCE ...... S.') 6. autoinoculation following refills in partial Pneumothobax . . . . . .88 7. Atttoinoculation following Refills with Bilateral Disease ........ 89 8. Tempebatube Chabt showing Loss op Fever . . 00 9. Temperature Chart showing Delayed Deferves- cence Due to Incomplete Collapse . . 91 10. Types of Fever Associated with Purulent Effu- sions ........ 120 PLATE FACING PAGE I. Radiogram of Chest showing Partial Pneumo- thorax ........ 54 II. Radiogram of Chest showing Complete Pneumo- thorax . . . ... . . . .SO THE PNEUMOTHORAX TREATMENT OF PULMONARY TUBERCULOSIS INTRODUCTION No more hopeful ray of sunshine has ever come to illumine the dark kingdoms of disease than that intro- duced into the path of the consumptive through the discovery of artificial pneumothorax. Recommended on theoretical grounds nearly a century ago, so soon, indeed, as the elasticity of the lung was first clearly recognised, first practised only some sixty years later, but again forgotten, it is only during the present decade that it is reaching that place of supreme importance in phthisis-therapy which is undoubtedly its due. In 1821 James Carson of Liverpool, in a series of enlightened physiological essays, first urged its trial in the treatment of disease. He pointed out in the clearest and most convincing manner ho\\' certain were the advantages to be derived from pulmonary collapse both in phthisis and in other diseases, and proved its feasibility by animal experiment. His views and some quotations from his work will be set forth more fully in a succeeding section (pp. 6 and 7). Nothing came of these recommendations at the time, which was, perhaps, just as well in those pre-antiseptic days. True, a certain Dr. F. H. Ramadge is credited with having punctured the chest wall some where about the year 1832, but the writer can find no evidence of this reputed operation. Nevertheless in this year (1832) James Houghton was reporting improve- ment in a case of advanced phthisis as the result of a 2 PNEUMOTHORAX TEEJTMEXT spontaneous pneumothorax — the patient, a bricklayer, was found thereafter working twelve hours a day at building a chapel. A few years later, in 1837, Stokes was also pointing out the decided improvement which spontaneous pneumothorax might sometimes exert on the course of phthisis. "Of the temporary suspension of the symptoms of phthisis in cases of perforation," he said, " there have been now many examples. If the perforation has oecvirred in a ease where tubercle was not yet deposited in great quantit}', and above all, where it had not yet appeared iii the opposite lung, the accident in question really prodvices a temporary cure of con- sumption." It was not, however, till the 'eighties that the subject of pneumothorax again aroused special interest, to find then, at last, practical realisation. Toussaint, a pupil of Potain's, in the year 1880 once again set forth on clinical grounds the claims of pul- monary collapse, to be followed in 1882 by Forlanini, who, however, failed to put the matter to the practical test till the year 1888. Meanwhile Potain in 1884 actually treated a case of spontaneous hydro-pneumothorax by replacing the fluid with sterilised air on repeated occasions, " une methode non usitee jusqu'ici." He further treated two other patients by the same method, reporting the three cases in the year 1888. All recovered, though one suffered a fatal relapse at a later date. He used a manometer in these operations, and advised renewing the air in certain cases of spontaneous pneumothorax so as to leave the intrathoracic j^ressure somewhere between 7 mm. Hg. and atmospheric pressure. Later he affirmed the superiority of nitrogen as the most inert gas, and that least rapidly absorbed. At about the same time and independently, in 1885, Cayley treated a case of severe hsemoptysis in phthisis by open incision of the chest wall, and reported it to the Clinical Society of London. The haemorrhage was largely arrested, though adhesions prevented complete collapse, and the patient died five days later with miliar}' tuber- culosis. INTRODUCTION 3 From this time onwards the matter appeared again to drop out of notice, though Forlanini Avas treating some cases, and reported them in 1894 and 1895. In 1898 J. B. Murphy of Chicago was urging the treatment on grounds of surgical principle, and himself treated five cases, his operation being carried out with a trocar and cannula. His pupil Lemke a few years later (1901) reported the treatment of fifty- three phthisis cases, many of them with fairly early disease, by a similar method. In the interval Schell of Indiana, in 1899, had described a case in which an artificial pneumothorax was resorted to for the cure of hsemoptysis. These publications attracted the notice of Brauer, to whom this method of treatment has since been indebted for much profitable speculation and re- search. His method of operation, by open incision down to the parietal pleura and perforation of this with a blunt cannula, has since been practically abandoned, but the safety achieved by the introduction of the water- manometer still stands to his credit. His first cases were reported in 1905 and 1906 ; Adolph Schmidt also reported a case in the latter year. Meanwhile Forlanini had been working in silence, and in the year 1906 was able to report twenty-five cases of pulmonary tuber- culosis submitted to treatment since his earlier publica- tion. Since then pneumothorax treatment has found its way into all countries where the medical sciences can be said to flourish. In Italy Forlanini has attracted a distinguished following ; the method has returned to France with F, Dumarest (1908) and Dessirier (1908), with Balvay and Arcelin (1909), and with Kiiss (1910), and P. Courmont (1910); to America with Harris, Mary Lapham (1911) and Robinson and Floyd (1912) ; to England with Colebrook, Lillingston, Vere Pearson and Rhodes, all in 1911, though before that time, in 1905, Stuart Tidey was extolling its principles at the International Tuberculosis Congress in Paris. In Ger- many, Brauer was followed by Adolph Schmidt (1906) and Lexer (1906), and by Klemperer, Wellman and 4 PNEUMOTHORAX TREATMENT others. Meanwhile Deiiniark had been adding dis- tinguished share to pneumothorax treatment through Saugman (1907) and Begtrup-Hansen ; and Switzer- land through L. Spengler, v. Muralt and a host of others. Thus the treatment of phthisis by artificial pneumo- thorax, first suggested on theoretieal grounds, later sup- ported on clinieal grounds derived from the observation of cases of spontaneous pneumothorax, now at last holds a firm place on grounds of its own undoubted practical value. Of no method of treatment, perhaps, does less difference of opinion exist as to application, technique, and, above all, utility among those who have had fair opportunity for its study. " All the physicians who have applied artificial pneumothorax among a notable number of patients," remarks a French writer, " agree in assigning to it an eminent therapeutic value. We have not seen, up to now, an equal unanimity in the appreciation of any antitubereulous remedy, and this agreement alone is of a kind to give us encouragement." But in another important particular also pneumothorax treatment may be claimed to differ very markedly from all other remedies used in the treatment of tuberculosis — it is especially applicable to the treatment of advanced disease. "It is a noteworthy fact," writes a distin- guished physician, " that practically all the methods of treatment and remedies which have been recommended during the past three or four decades for the relief and cure of phthisis, have been stated to exercise their alleged curative effects only during the incipient stage of the disease, when diagnosis is doubtful in a large proportion of cases, and spontaneous cure is not un- known. It is therefore refreshing to find that the newest method of treatment of pulmonary tuberculosis is recommended for far advanced cases, especially those manifesting a tendency to pursue a rapidly fatal course, and showing signs of extensive excavations. When every other method for checking the progress of the process in the lung has been tried and found wanting ; when fever consumes the bod\-, or life is threatened with INTRODUCTION 5 copious and uncontrollable pulmonary haemorrhages, this method of treatment may be applied with striking success." This, indeed, is the role of artificial pneumo- thorax, to rescue those formerly beyond the bounds of human aid, to restore the lost to life, or at least to a measure of comfortable existence. The indications for its use and also the contraindications, the measure of its achievements and their limitations, and the most important of those points on which success or failure may turn, will be duly set forth in the following pages. CHAPTER I THE MODE OF ACTION OF ARTIFICIAL PNEUMOTHORAX Though a study of the effects of collapse of the lung in pulmonary tuberculosis must be expected to involve many obscure theoretical problems, yet to the clinician two clearly defined factors stand out for main considera- tion — on the one hand the local mechanical effects, and on the other the removal of toxaemia. MECHANICAL EFFECTS These and their possible application to the treatment of disease were noted so soon as the elastic recoil of the lungs was first a recognised fact ; James Carson, as already remarked, was urging the iJroduction of artificial pneumothorax as early as 1821. He pointed out how great a bar to healing processes were the elasticity, con- dition of extension, and constant movement of the lungs. In abscess of the lung he remarked " the sides of the abscess are prevented from falling into a salutary- contact, not by the matter which lodges between them, but by the powerful elasticity and retraction of the surrounding substance." As soon, he says, expect heahng in a divided tendo Achillis without mechanical aid. By collapse of the lung, on the other hand, " the diseased part Avould be placed in a quiescent state, receiving little or no disturbance from the movements of respivatiou wliich would be performed solely b^^ the other lung, and the divided surfaces would be brought 6 MODE OF ACTION OF PNEUMOTHORAX 7 into close contact by the same resilient power which before had kept them asmider. A wound or abscess in this Imig would be placed in circumstances at least as favourable to the healing process as the same affections on any other part of the body." Finally he recom- mended its application to the treatment of " consump- tion " — " in those cases in which the disease is placed in one of the lungs only, the remedy would appear to be simple, safe, and complete." Further on he remarks that " a haemorrhage from one of the lungs, the frequent prelude of consumption, if not immediately fatal, would be as certainly stopped by the collapse of that lung, as the flooding consequent upon parturition by the con- traction or rather the resilience of the womb." He was by no means blind to the danger, by sudden collapse, of flooding the tubes with blood or pus, and advised reducing the lung "to a state of collapse by degrees only." " It has long been my opinion," he says, " that if ever this disease (phthisis) is to be cured, and it is an event of which I am by no means disposed to despair, it must be accomplished by mechanical means, or in other words by a surgical operation." This sound statement of the manifest mechanical advantages to be expected from pulmonary collapse can hardly be bettered by any quotation from later work. But to it must be added certain less obvious factors, still in the main mechanical, which must also minister to the beneficial results, though in a degree which is at present uncertain and even contentious. These are, primarily, the effect on the growth of the tubercle bacillus of the diminution of oxygen in the collapsed lung, and of the rapid growth of fibrous tissue ; and, secondarily, the effect on this latter process of circulatory changes, whether hypersemia or its opposite, of lymph stasis, and of the imprisonment of toxins. Each of these has been expanded into the role of an explanatory hypothesis in the hands of one or more authorities, though it lias seemed to the writer more reasonable to regard tlicm as but notable factors in the 8 PNEUMOTHORAX TREATMENT total result. The local improvement foreshadowed by Carson has generally been well attamed in practice. Under artificial pneumothorax, as will be more fully described elsewhere, the lung is retracted to a firm, airless, fleshy mass, cavities and tubes are flattened out, their surfaces become adherent, and the diseased parts, and in some cases healthy parts also, become permeated with a dense growth of fibrous tissue. In such a lung, tuberculous processes are always in abeyance, and no fresh tubercles develop, a fact which post-mortem evi- dence has clearly established. So much for the local facts associated with pulmonary collapse — one side of the picture and certainly a very important one. SEROLOGICAL EFFECTS The other side of the picture shows itself in the removal of toxaemia as exemplified by loss of symptoms, fever, sweats, malaise, wasting, and therewith improvement of general health. It can hardly be doubted that these effects, analogous to those produced through reduction of autoinoculation by other means, are due to some mechanical bar to the absorption of toxins. In the collapsed lung fluid excretions are evacuated, and can no more accumulate, caseous areas are rendered drier, cavity walls unite, and the diseased surface is, at worst, enormously reduced in area. In addition a stasis of lymph is brought about by immobilisation, the lymph channels are distended as was shown by Bruns. Warnecke, and Ivistler, and absorption reduced to a minimum. The blood supply may be large or small according to the intrapleural pressure, and since toxic absorption docs not in practice vary with this pressure, tire removal of toxsemia, which is so important a feature of pneumo- thorax treatment, may well be ascribed, in the main, to lymph stasis. In what way, it may be asked, does the removal of toxaemia or avvtoinoculation act ? Not only does it release the nervous system and all the organs from a 3I0DE OF ACTION OF PNEUMOTHORAX 9 poisoning wliicli has seriously interfered with health, but it at once breaks through that " vieious circle" which has favoured the spread of disease in the other lung and elsewhere. For, as in overdosage with tuber- culin so in autotoxsemia, every large "general reaction " exhibited in fever, malaise, and other symptoms is accompanied by that stirring up of all the tuberculous foci known as a " focal reaction." In this way the active foci are flooded with blood, a further supply of toxin is carried into the circulation, and further " re- actions " rcsidt ; in addition these foci are helped to extend and grow, quiescent disease becomes active, and new centres arise. The removal of tubcrculo-toxaemia curtails those focal reactions in the better lung which were leading to the spread of disease, and thus helps to bring about in that organ the improvement ^xhich almost invariably follows. It may well be, also, that, as many believe, a favourable hyperemia of this lung may occur and assist the healing process (pp. 115 and 116). In association with the removal of excessive reactions the opsonic index rises as would be expected, and as has been demonstrated by Pigger, working under L. Spengler, by Carpi, and others. The removal of toxaemia is so important a feature in pneumothorax treatment that the writer is always less inclined to recommend the operation in cases where definite toxEcmia is absent, and where, in consequence, the mechanical effects can alone be brought into play. Where, however, large cavities can be compressed and widespread disease olaliterated, where, in short, the mere character and extent of disease render heahng well- nigh impossible in an extended lung, the mechanical advantages belonging to collapse are not to be withheld. But their influence by itself on the growth and develop- ment of tubercles, at least under experimental conditions, appears to be somewhat in doubt. Breccia, indeed, finds that pulmonary collapse in animals will prevent the spread of disease to this lung ; but Schvu- and Plaschke observed the development of tuberculosis in 10 PNEUMOTHORAX TREATMENT the compressed as well as in the functioning lung of animals submitted to tubercle inoculation, whether intravenous or intratracheal. On these grounds they argued that the removal of toxaemia A\as the all-important factor in pneumothorax therapy, and not the pulmonary collapse. Other serological influences may well be at Avork in cases of pneumothorax therapy, pei'haps pleural influ- ences analogous to those which often lead to general improvement as the result of a pleurisy, even a dry one, in cases of phthisis. These, if they occur in pneumo- thorax, are apparently enhanced in certain cases by the development of a pleural effusion, but of their nature we have, at present, no certain knowledge. CHAPTER II INDICATIONS AND CONTRAINDICATIONS The success of pneumothorax treatment is not a little bound up in the suitable selection of cases. Indicated for progressing, marked, or even advanced disease, it is, nevertheless, only applicable to a selected proportion of such material, although the majority of advanced cases must at some time or other have passed through a stage where a useful pneumothorax was at least a possi- bility. Among a mixed phthisis material containing patients at all stages of the disease, the incidence of cases immediately suitable for pneumothorax therapy appears to be a rather small one. Keller puts it at 7% with only 3'9% successfully achieved. Ferrio finds 5 or 6% fully suitable among 700 cases examined, but thinks that this number could be increased with experience. Fishberg notes 22 suitable cases among 210, or 10%, but regards this figure as high because he was at the time agitating to obtain such cases ; he believes the usual incidence to be below 5%. Bernard found 22 among 628 cases or 3*5% suitable, Courmont 31 of 352 or 9%. These percentages of suitable cases, however, based on a mixed material, necessarily take no heed of past or future — ignore, in other words, the fact that most progressive phthisis cases at some time in their course must reach a stage when pneumothorax treatment might be used with advantage, and that a case unsuitable to-day might have been suitable yesterday, or may become so to-morrow. This fact should be taken into account in gauging the pro})ortion of phthisis cases suitable for 11 12 PNEUMOTHORAX TREATMENT pneumothorax treatment, particularly since it is one which must gain increasing importance as the value of the method becomes more and more widely known. For the physician well acquainted Avith tlic method and familiar with its indications and contraindications will not allow his j^atients to pass beyond the point at which this, and with it every other form of treatment, becomes too late, but will seize the suitalde moment in each case, and thus ensure that tlic benefits of pneumothorax treatment are extended to a much larger proportion of phthisis material than could be tlie case at the present time. What may be called the "classical " ease for artificial pneumothorax is one of severe one-sided disease \\ith the other lung clear or nearly clear to physical examina- tion. These cases, less uncommon than might at first sight be expected, must form a large ijroportion of any well-selected pneumothorax material ; they appear to represent a distinct group capable of being separated off from more ordinary adult cases of pulmonary tuber- culosis. For ordinary apical phthisis of adults spreads but a little way in one lung before it presents physical signs at the opposite apex, and tJie conditions of ad- vanced one-sided disease can thus ne^er be reached. The unilateral case, so suitable both in theory and in practice for pneumothorax therap}', stands in a group apart, and merits special attention in a book devoted to this form of treatment. CLINICALLY UNILATERAL TUBERCULOSIS Since tJie writer's hospital pneumothorax material has been mainly composed of advanced one-sided cases, a reference to these will serAc conveniently to illustrate the peculiarities of the group under consideration. Among twenty of these eases, all adults, where the type of disease could be decided with some probability, seventeen appeared to stand in a group by themselves, presenting points of difference from ordinary- apical phthisis. All INDICATIONS JND CONTRAINDICATIONS 13 of these showed disease covering the -whole of one side, and in all the good lung was practically clear of physical signs. In six no signs of cavitation were present, but large moist crepitations extended from apex to base, in two on the right and in four on the left side. In eleven cavitation was evident to physical examination, in most of considerable extent, but in all it presented one striking peculiarity. In no case was it in proximity to the apex ; in all, the cavities Avere found comparatively low down in the lung. Among four with right-sided disease, in three cavitation was in the axillary region, and in the fourth at the base. In the remaining seven disease was on the left side and in six of these cavitation was just outside the heart about the middle of the lung, or in the axilla, or close above the heart ; in one it was basal. In practically all the cases notable evidence of fibrosis with displacement of heart and mediastinum was present, particidarly, as would be expected, in the cases with cavitation. The good lung, as has been mentioned, was clear to stethoscopic signs in all these cases, save for doubtful sounds which proved, in the light of subsequent events, to have been conducted from the diseased side. The use of the X-rays, however, soon dispelled the illusion that this better lung was in reality free of disease. In all cases there was visible a definite peri-bronchial tuberculosis, in most cases indeed quiet or arrested, spreading out from the hilum and often reaching a con- siderable way towards the surface. This occurrence of deep disease without surface signs is immediately reminis- cent of hilum tuberculosis of children, and further com- parison brings conviction that these clinically unilateral cases are none other than a parallel condition, hilum tuberculosis, in the adult. The main characteristic of this type of disease appears to be its tendency to a chronic course with little or no symptoms until a late stage is reached. Then at last toxaemia appears sufficient to overcome tolerance and the symptoms of an active autotoxic tuberculosis arise. By this time disease may have reached the surface on one side, and physical signs 14 rNFAijSiornouAX treatment apjicar almost simultaneously' over the \vhole or nearty the whole lung. On the other side, no physical signs of disease may be present, and the impression of a truly one-sided process is thus wrongly given since disease may, in reality, involve the whole of this lung except the immediate surface. If disease is active on this side also, a condition which is most easily discoverable to radio- graphic examination, the case is not suitable for pneumo- thorax treatment. But often disease on this side is of a quiet type or arrested, and the appearance of such a process on the plate, even if of wide extent, need not in practice deter from the attempt at pneumothorax treatment, since the results of this often prove them- selves as good as if the lung had been reallj' healthy. Unilateral tuberculosis as a distinct type of disease Avas first discussed by Straub and Often in 1912. They described twenty-seven cases mostly in children or young people, and they recognised, moreover, its peculiar relationship to pneumothorax treatment, and included in their paper an analysis of 99 cases treated by Brauer of Hamburg after this method. The main character- istics on which they laid stress were — the low level of the oldest process and hence of cavitation : the appear- ance of signs first over the front of the chest, the marked tendency to fibrosis and displacement of organs, and the more chronic course of left-sided disease proceeding in many cases to cavitation without complaint of either cough, sputum, fever or pain ; the characteristic spread of disease on the right side into the axilla, and on the left side towards the surface just above the heart. They analysed some general tuberculosis material to discover the incidence of this " unilateral " tvjje of disease. Among 135 cases of pulmonary tuberculosis, they found left-sided disease of this type, chosen as being most readily recognised, in 10 cases or 7'4%. Among Brauer's 99 pneumothorax cases, 48 had right- sided and 51 left-sided disease. Among the latter 17 or 38 -3% were certainly of this type, and a further 18 or 35-8% probably so. They produced figures also to IXniCATTONS .IND CONTIfJINDICATIONS 15 sho^v that the results of i)neumothorax treatment were strikingly better in this class of case than in cases of ordinary type. Their paper also contains an interesting comparison of the results of pneumothorax treatment with those of ordinary sanatorium treatment in these cases. Brauer's 17 left-sided tuber culous cases treated by Pneumothorax Straub & Otten's 27 cases under Sanatorium treat- ment Much improved. Improved. Xo change. Died. I 10 |=58'8% 5 = 20-4% 2 = llS"o ) 2 1=10-5% 7 = 30-9% 3 = 15-7% = 36-9% AVe thus find that among cases suitable for pneumo- thorax treatment, first place must be given to clinically unilateral tuberculosis, an outcome, though only an occasional one, of hilum tuberculosis in the adult. When the case is first seen severe disease is already manifest as a rule over a considerable part of one lung, though but a short history of symptoms may be obtain- able. The type and extent of disease hidden in the apparently healthy side must necessarily be a matter of doubt. Dyspnoea cannot be taken as clear evidence of widespread lung involvement, since it may be of toxic origin. The slightest degree of cyanosis, on the other hand, seems to the writer of ominous significance. When disease is active and near the surface on this side, the earliest signs usually show themselves in front, in axilla and near the base. The only definite evidence, however, of the condition of this lung, is obtainable by means of a radiogram. For the purposes of pneumo- thorax treatment shadows pointing to arrested or quiet disease, even though this is of wide extent, may generally be ignored. Active disease is a relative contraindication unless limited to the lung immediately around the hilum. 16 PXEUiMOTIiOFAX TBEATMEXT APICAL PHTHISIS Pulmonary tuberculosis of the ordinary apical type is suitable for pneumothorax treatment under much closer limitations. In two particulars it stands in marked contrast to the class of case, hilum tuberculosis, already described ; firstly, disease cannot be allowed to spread far without risk of the opposite lung becoming decidedly involved ; secondly, disease in the better lung is more likely to be active and to constitute a serious menace to successful treatment. As a matter of fact the first (if these is only an apparent difference, depending on the fact that apical disease gives manifest signs at a stage where central disease is still of necessity hidden. But the tendency to a more active type of disease on the better side in apical phthisis constitutes a real distinction and one that especially governs the use of artificial pneumothorax in this disease. Where, then, should the indications and contraindica- tions be drawn in a case of apical phthisis ? Forlanini warns us to think of pneumothorax treatment as soon as ulcerative processes appear in the kmgs, and this for the following reasons : — ■ (a) The probability of disease remaining one-sided becomes smaller the further the process advances. (6) The more advanced the disease the greater is the likeUhood of adhesions limiting or altogether preventing this treatment. (c) The more circumscribed the disease the more assurance is there that the kmg will expand and functionate perfectly when treatment is completed. {d) The more advanced the disease the greater is the risk of that fatal accident, perforation of the lung. These warnings against delay are apphcable to hilum tuberculosis no less than to phthisis, but in tl\at condi- tion it commonly happens that the extent of disease has only been realised when it is already far advanced. In apical phthisis the most ^'ital of these warnings is the first, the danger, namely, that active disease may INDICATIONS AND CONTRAINDICATIONS 17 appear in the better lung and jeopardise or altogether prevent treatment by artifieial pneumothorax. As a rule this will have happened before the other three warnings need serious eonsidcration, so that our first and main anxiety must be the state of the better lung. The Better Lung. — The favourable influence Avhich pneumothorax treatment exerts on the lesions in the functioning lung is not the least notable of its achieve- ments, but these lesions must be reasonably suitable both in quality and quantity for this benefit to accrue. Most writers are of opinion that active disease in the better lung is a distinct contraindication ; inactive or quiet disease is considered no bar. These qualifications are obviously open to various interpretations — the decision must always be directed to the conditions present in the individual case. As regards the activity of lesions, Forlanini finds that limited active lesions may improve ; Weiss remarks that they may sometimes be favourably influenced — he has at least seen them uninfluenced. L. Spengler, on the other hand, regards a fresh active process on the better side as an absolute contraindication. As regards position and extent Zink found that even arrested lesions which extended beyond the upper lobe generally got worse. Carpi remarks that it is the posi- tion of the lesions rather than their grade which is of import ; he says good results are obtainable only where disease is confined to the apex or the hilum of the hmg ; if at the base, especially at the left base in the region of the heart, they always tend to get worse. Forlanini, however, has described two cases with double lesions too advanced to treat, but which later developed spon- taneous pneumothorax ; this Avas kept up artificially, and resulted in great improvement in both cases, in one proceeding even to complete recovery. On this account he suggests that some cases Avith double lesions should be given the chance, and he and certain others have treated both lungs alternately with some measure of success. Certain operators, also, have treated cases of double 2 18 PNEUMOTHOEAA' TBEATMENT disease simultaneously with at least temporary improve- ment in the general health and comfort. Fagiuoli suc- ceeded in introducing gas into both pleural cavities in five among twelve cases of considerable double disease in which this was attempted ; 100 e.c. to 150 c.c. of nitrogen was put in at longish intervals, the i:)ressure being, of necessity, kept below the atmospheric. Under such measures he obtained notable improvement in three cases : temperature and pulse fell, the sputum diminished, the weight rose, and haemoglobin increased greatly. In spite of this practical success i\Iorelli con- demns this method on the theoretical ground that in a double pneumothorax it must be the elastic healthy lung tissue which will tend to collapse, leaving the diseased and less elastic parts unaffected. Parry Morgan,, how- ever, points out that elastic tension along a lung axis containing diseased tissue must necessarity be increased ; hence gas introduced into the pleural cavity will tend to collect over diseased areas, and by releasing this over- stretching, particularly in inspiration, Avill allow the diseased parts to rest. This seems to the wa'iter a likely occurrence where a minimal quantity of gas is admitted, as in Fagiuoli's eases, while Morelli's objection might hold if larger fillings were attempted. In any case it must be remarked that bilateral pneumothorax has not yet emerged from the experimental stage, and that for ordinary pneumothorax treatment the marked involvement of both lungs is for the most part a clear contraindication. Cases may impro^-c for a time ; Pisani recorded improvement in 33 of his 72 bilateral cases, and great improvement in the remaining 39, but in all it was but temporary. On the other hand, in some cases disease in the functioning lung may be stirred to increased activity- and the inevitable end actually hastened. The treatment of cases but doubtfully suited for artificial pneumothorax can but tend to discredit a method possessing quite definite indications within Avhosc limits brilliant and lasting results can be con- INDICATIONS AND CONTRA I NJ)I( AT IONS 19 :(idently anticipated. This is not to deny that these indieations may be legitimately extended ^^•ith increased ex23erience — indeed, the writer is tempted to anticipate a time when temporary pneumothorax may become a recognised method of treating the vicious circle of auto- inoculation in cases where fever cannot be reduced by " typhoid " rest. But that time is liardly yet, and the credit of the method is best served by conforming at present to its well-recogniscd limitations. Early Phthisis, — A disqualification, even more im- portant in the writer's opinion than that afforded by bilateral disease, exists in the instance of early phthisis. The application of pneumothorax treatment to these cases can be shown on many grounds to be contraindi- cated. In the first place it is unnecessary, since early phthisis is a disease readily curable by other methods ; in the second place it is inexpedient since it brings with it risks and the chance of serious complications out of all proportion to the benefits it can confer. Pleural shock and gas embolism are, perhaps, but infinitesimal risks where modern methods are properly applied, but they cannot be altogether eliminated — particidarly the former in cases of early disease. For de Carle Woodcock and Clark treating cases of this class have found pleural shock a not uncommon accident, and they especially note its greater frequency in early cases with a clear pleura than in patients with more advanced disease. This fact is in itself an important argument against the treatment of early pulmonary tubercle by artificial pneumothorax. Furthermore pleural effusion and the deformities which may follow it are complications which cannot in any way be avoided. Such effusions occur altogether in nearly 50 % of cases, and though the shorter treatment required for slight disease would reduce this percentage, yet effusions must occur in many, and lung expansion fail or be incomplete in a few. More- over, on re-expansion the pleural layers may become ad- herent, and the patient be thereby deprived of this treat- ment at a later stage when it may be truly indicated. 20 PNEUMOTHOTiAX TREATMENT la some cases perhaps the production of a pncuinothorax at an unusually early stage may be justified on economic grounds Avhere the cure must be shortened in the case of a patient Avho is obliged to return to his occupation. Having condemned, then, tlic extremes wherein pneumothorax treatment is resorted to too early or abandoned till too late, avc may return to the recom- mendation of Forlanini to think of pneumothorax treatment as soon as idcerativc i^roccsscs appear in the lungs. Cavitation of anything but small extent, an infiltrative process progressing in spite of a fair trial of sanatorium treatment, these, where the other lung will pass muster, are in general the indications in cases of ordinary phthisis. The acuter forms of tubercle need separate consideration. ACUTER FORMS OF TUBERCULOSIS Miliary tuberculosis does not, of course, come into consideration. The aeuter broncho -pneiunonie forms, whether for the most part apical or central, are commonly so far bilateral that the question of pneumothorax therapy hardly ever arises. Caseous pneumonic cases, on the other hand, are generally one-sided at first and have been submitted to pneumothorax treatment with varying success at the hands of different operators. Forlanini has not been fortunate in eases of this type, having foimd that disease always invaded the opposite lung. Brauer, and also Saugman, however, consider this type of disease suitable for pneumothorax treatment. The latter has had success in a ease of consolidation of the whole of one lung ; Dumarest and Keller also each claim a cure in a case of this class. Weiss treated five pneumonic cases ; in four disease appeared on the opposite side, in one case as late as six months after, and death resulted. In the fifth case, which was one of caseous pneumonia, on the top of chronic phthisis, recoA'cry occurred. The writer has treated one such case, published in INDICATIONS AND CONTRAINDICATIONS 21 the Lancet of July 15, 1916, with the best possible results up to the present. After eighteen months' pneumo- thorax treatment the patient is at work, in good general health ^vith gain of over a stone in weight, and with the functioning lung free of disease. The pneumothorax is, of course, maintained. This is a satisfactory outcome if time confirms it ; the writer has occasionally seen these pneumonic cases settle down to a quiet chronic process without pneumothorax treatment. Apart from the danger of active disease arising in the better lung in these cases, the pneumonic lung is, for mechanical reasons, but ill adapted to compression and collapse, or to pro- cesses of repair. This was particularly insisted on by Graetz, two of whose three autopsies exhibited pneu- monic changes. These areas he found to be much less completely collapsed, and much less readily encapsuled and obliterated by the growth of fibrous tissue than are scattered, isolated tubercles. HiEMOPTYSIS Collapse of the lung may be indicated in cases of severe recurrent haemoptysis, or as an emergency opera- tion for a bleeding otherAvise uncontrollable, and likely to prove fatal. Under the latter conditions a large quantity of gas, 1,000 or 1,500 c.c., may be put in at once, the needs of the moment outweighing the usual contraindications. Such large quantities are not, how- ever, necessarily required to arrest the hsemorrhage. Masenti, Borgogno, and Vergano found that no great amount of collapse was needed, nor even a positive intrapleural pressure, to achieve this object. A gas quantity varying between 250 c.c. and 1,100 c.c. sufhced to stop the bleeding in their five cases. If a small amount is found sufficient, further collapse may be deferred to a second filling, a complete pneumothorax being indicated Avherc the other lung is in a condition to stand the strain. In these eases, hoAvcver, it will often happen that the other lung is markedly diseased, 22 PNEUMOTHORAX TREATMENT and only partial measures can be undertaken. Pleural adhesions, moreover, may restrict or altogether prevent the operation ; in a case of the writer's treated for recurrent haemorrhage, only a partial pneumothorax could be attained, and this reduced but did not arrest the recurrent attacks. To effect a permanent ciu'e in such cases the responsible cavity must be completely flattened out and maintained so under a suitable positive gas jDressure. SPECIAL CONTRAINDICATIONS Disease of various organs, whether tuberculous or not, may contramdicate the production of an artificial pneumothorax. Lungs. — The condition of the functioning lung, qua tuberculosis, has already received consideration. In addition advanced emphysema and asthma are con- traindications. Dyspnoea may be toxic, and in this case will be relieved by the removal .of autoinoculation ; if dependent on widespread lung involvement or em- physema, when some degree of cyanosis will be detect- able, it is. of course, a clear contraindication. Circulatory Organs and Kidneys. — Pneumothorax therapy probably reduces the pulmonary vascular area, in most cases at least, and thus puts a strain on the right heart. In addition the loss or reductioii of intra- thoracic suction must tend to embarrass the venous circulation. The effect of these disabilities on circulatory organs and kidneys renders any serious disease of these organs a contraindication to pneumothorax treatment. On the other hand it must be noted that certain cases of nephritis appear to be of tubcrculo-toxic origin, and have been known to recover when autoinoculation was removed. Laryngeal Tuberculosis is no contraindication unless far advanced, and even then a doubtful one. Da Gradi has described three bad cases from Forlanini's clinic >vhicli improved and eventually Iiealcd under pneumc- INDICATIONS AND CONTRAINDICATIONS 23 thorax treatment. Slighter cases of laryngeal disease eommonl}'^ improve ; eleven such cases mentioned by Zink all recovered except one very severe one, which nevertheless decidedh' improved. Intestinal Tuberculosis is a contraindication unless very slight and consequently latent. Carpi records that six among his thirteen fatal cases died of intestinal tuberculosis, which became more active under pneumo- thorax treatment. Brauer and Spcngler, however, record two cases Avhere intestinal disease caused sj^mptoms, and where these entirely disappeared under pneumothorax treatment. Forlanini is not against the inclusion of slighter cases. Diabetes. — The presence of this disease should make the physician very wary of recommending pneumo- thorax treatment ; not only are latent foci in the better lung very likely to become active and progress, but the punctures may inflame and give trouble as in a case recorded by Weiss. CHAPTER III PNEUMOTHORAX APPARATUS GENERAL FEATURES The essentials of a pneumothorax apparatus arc sur- prisingly small and simple — no more, in truth, than could be collected for emergency in many a doctor's surgery, and put together in half an hour or less. A couple of wide-mouthed bottles of about 1 to 1^ litre capacity, a few feet of glass tubing of about 4 mm. internal diameter, and some rubber tubing of approxi- mately equal bore — these constitute the main require- ments. The corks of the bottles, which must be of rubber if for more than teinporary use, must be bored in tAvo places ; the gas bottle should be ruled so as to record volumes in 100 c.c. or less, and numbered from below upwards. The bottles are half filled with water, or better, an antiseptic solution (p. 37), and are put into connectit)n at their lowest point by tubing reaching to tlic bottom of each, whereby the fluid can be made to flow from one to the other by an appropriate alteration of their levels. The air above the fluid is in one, the " pressru-e bottle," ojoen to the outside air through a short piece of glass tubing, in the other, the " gas bottle," it is connected by a glass T piece with the manometer on the one hand, and with the pneumothorax needle on the other (fig. 1). A cotton wool filter of some kind must be introduced between the T piece and the needle to filter the air or gas. This can be a small glass chamber, corked at each end, or a length of stout glass tubing blown into a 24 PNEUMOTHORAX APPARATUS 25 bulb, and packed with sterile wool. The manometer consists of two uprights of glass tubing some 40 cm. long connected together below by a sliort lengtli of rubber tubing. Water is entered to a point about half ^\•ay up each length, and a scale ruled in half centimetres fixed between them and numbered 1, 2, 3, 4, etc., upwards and downwards from tlie level of the water. Readings are made only on the limb which is opeii to the outside air, and, since each half centimetre rise (ov fall) is accom- panied by a half centimetre fall (or rise) of the opposite column of fluid, the readings on the scale will record the difference between their levels, that is the pressure, in centimetres of ^^'ater. These are the essentials of a pneumothorax apparatus, and the nearer the instrument can be kept to these primitive lines the better it is likely to answer its purpose. Every departure from them must, for its justification, be shown to satisfy some important need. The appara- tus designed by Vere Pearson and Lillingston and shown on p. 26 is here recommended on the ground that it departs so little frona first essentials, and yet satisfies all practical requirements. The gas bottle has been made long and narrow, so that tlie slightest change of fluid level can be quickly noted, and small quantities of gas measured with fair accuracy. The manometer is conveniently fixed to a firm partition, and its upper ends are bulbous, so as to minimise the risk that the fluid be unexpectedly sucked or forced out of tlie ends. A small screw clamp surrounds the connecting rubber at the base, and thereby the respiratory fluctuations can be abolished at will, and a naean intrapleural pressure recorded. The tall graduated bottle is filled with nitrogen from a cylinder, or with air, the fluid, generally an antiseptic solution, being driven back into the i^ressure bottle, and retained there by clamps on the tubing. When the clamps are released the fluicl will flow from the higher level by syphon action, displace the nitrogen, and force it at any desired pressure through the pneumo- thorax needle, 2G PNEUMOTHORAX TREATMENT This apparatus is a good example of movable flask apparatus in general, and is similar in design to those used by Murphy, by Brauer, and by Spengler. These Fio. 1. — Peabson and LiixiNGSTON Pndtjmothoeax Appabatcts. The maDometer is better graduated in half-centimetres (for reasons given on p. 25) and numbered 1, 2, 3, 4, 5, etc., all the way up and down for facility of reading. operators, however, appear to have employed flasks of 2-litre capacity, which would seem needlessly cum- brous and heavy. Convenient but more elaborate and pxpensive apparatus on somewhat similar lines are those PNEUMOTHORAX APPARATUS 27 of de Carle Woodcock and Lister in this country, and of Kiiss of Paris. Woodcock's manometer consists of a narrow tube immersed in a larger reservoir of fluid, the upper end being continued into a T piece connected on one side with the needle, and on the other A\ith the gas bottle. Tlic bottles move up and down on upright metal rods, the nitrogen is manufactured by sucking the air through a pyrogallate solution, and in Lister's modiircation of this instrument an oxygen apparatus is also added. Lister has also designed a regulator with a dial whereby the different parts of the apparatus can be mechanically connected or cut oft with case and certainty. Kiiss's gas bottle is fixed at a somewhat elevated level, while the pressure bottle is made to move up and down on a rack and pinion ; the nitrogen is kept in a bag from which the gas bottle can be supplied whenever its contents have been replaced by fluid. A word must be said about apparatus Avith fixed bottles, since this is still used by such authorities as Forlanini, Saugman, and v. iMuralt. The pressure and gas bottles and the manometer are all fixed on a stand, and when prepared for use the fluid is driven back into the pressure bottle by connection with the nitrogen cylinder. When the needle is in the pleural cavity the fluid pressure is released, and forces the gas as slowly as desired into the chest. But this fluid pressure, since the apparatus is fixed, must be high at first, and must steadily fall as the fluid descends into the bottom of the gas bottle. Soon a jJoiiit is reached when the intra- pleural pressure balances the fluid pressure, and the flow ceases altogether. The pressure must then be raised by the use of a hand bellows, an ill-regulated and necessarily intermittent source of power, which Morriston Davies in this country has advantageously replaced by the use of an accessory pressure bottle. The disad- vantages of fixed bottles are manifest, and are especially apparent at a first operation, where it is advisable, in the interests of safety, to enter gas at first at a decided negative pressure. This is, of course, easily accom- 28 PNEUMOTHORAX TREATMENT plished Avith movable bottles })y raising the gas bottle so that tlie level of its fluid is tlic desired amount above that in the pressure bottle. Saugman with flxed bottles tries to overcome the dilTieulty l)y starting with the fluids at equal level, but is soon driven to use the hand bellows to secure the admission of gas. A position equally disadvantageous is experienced when a Replace- ment of Gas is indicated in cases of pleural effusion (]i. 1-17). With fixed apjaaratus the admission of gas into the pleural cavity in this ojjeration must almost necessarily be intermittent, since undivided attention must be given to its entrance under an uncontrolled fluid pressure, whereby undesirable or even dangerous intra- pleural pressures may inadvertently be reached. With movable bottles, on the other hand, the gas can be allowed to flow all the time the fluid is draining away, all that is required being an occasional raising of the pressure bottle to maintain the slight positive pressure generally desired. For these reasons, among others, the writer does not hesitate to raise his voice most strongly in favour of one or other form of movable bottle apparatus, and against all apparatus with fixed cylinders. Having dismissed the gas and pressm-e bottles or cylinders about which, beyond their mobility, there is not much to be said, a little more attention may be devoted to other parts of the apparatus. THE MANOMETER The water manometer should be some 35 to 40 cm. high, and made with bulbous upper extremities, as already noted, so that the fluid will not be so readily driven out by cough, nor drawn into the chest under an exceptionally great negative intrapleural pressure. The calibre of the manometer tubing is, in the writer's opinion, of some importance for the recording of those delicate fluctuations of pressure which keep us in touch with intrapleural changes. The wider is the discrepancy between calibre of neetllc and size of manometer tubing, PNEUMOTHORAX APPARATUS 29 the less responsive is tlie column of fluid. Thus the respiratory fluctuations on various occasions in the case of H. B. with a fine Brauns' needle (calibre '8 mm.) showed — Before Fillms. After li'illino;. 2 cm. H.,(). 2 „ „ 4 „ „ 2 cm. H.2O. 1 ,, but with a Saugman needle (cahbre 1"2 mm.)— Hefore Pilling. After Filling. 7 cm. H„0. 4 cm. HnO. 3 „ „ 3 ., a manometer of 4 mm. internal bore being used in all cases. Where the Saugman needle was retained but a mano- meter of large (10 mm.) bore substituted, the fluctuation was again diminished th\is : Before Pillin;,'. After Pilling. 4 cm. HoO. I 2 cm. H2O. The comparative inertia of the fluid in a manometer of so large bore is very obvious in practice, and the clean and amj^le fluctuation which is sucli reassuring evidence that the end of the needle is in the pleural cavity hardly shows itself with a needle of ordinary calibre. The fluctuation is somewhat reminiscent of that obtained with a half-blocked needle. Saugman uses manometer tubing of 3 mm. bore. The writer has found 4 mm. internal diameter gives very satisfactory results, and would advise against the use of any exceed- ing 5 mm. It must be borne in mind also that the chances of gas embolism arc increased by the use of tubing of wide bore, since tlie amount of gas released at the needle point at the moment of a negative fluctuation must thus be notably increased (p. 159). To reduce still further the risk of gas embolism Wiirtzen introduces at first operations a small water manometer close to the :M) PNFAJMOTnOlUX TBEATMKXT needle, so that rarefaction of gas in a length of tubing is avoided and a fluctuation of — 14 cm. only provides 1 c.c. of gas at the needle point. It has become customary in pneumothorax work to measure intrapleural pressure in centimetres of water, since the water manometer is that almost invariably used in practice. Equivalents in mm. of mercury ma)- be marked on the scale if desired, and some apparatus, as those used by Braucr, by Saugman, and by Hansen, jjossess a mercurial manometer in addition for the re- cording of very high or low pressures. Obviously 35 to 40 cm. is the limit of height for use in a convenient and portable apparatus, and positive pressures above that point will rarely, if ever, be required. It may happen, however, that a very large negative pressure may have to be expected in cases where fillings have been neglected or purposely discontinued, and in this event the use of a mercurial manometer will save the risk of the mano- meter fluid being sucked into the chest. Forlanini overcomes this difficult}^ by the provision of an air chamber, in connection with the open end of the mano- meter, of such known capacity that the pressures can be thereby doubled. During ordinary use it is kept open to the air, but when its mouth is closed by a tap, the manometer readings are suitably controlled by the presence of this column of rarefied or compressed air. It is an essential of proper treatment or even of safety that the manometer be never shut off from the needle, so long as this is in the pleural cavit)-. That this warning is not needless is illustrated by the remark of a recent Avriter that the pressure after a filling in one of his cases must have been high, since on opening communication with the manometer the manometer fluid was blown out of the tubing. Such an occurrence can only be condemned as malpractice. FXEUMOTHORAX JPPJRATUS 31 THE GAS TO USE The question of a suitable gas revolves about the properties of Oxygen, Nitrogen, and Air ; the suitability of fluids such as normal saline or oils for injection can be dismissed as, at the present time, irrelevant. Choice of a gas is open to discussion on two main grounds — the question of safety, that is, the risk of gas embolism, and the question of diffusibility. At a First Operation. — A good man>- operators, not- abl}'- Saugman, employ oxygen at the first filling with a view of avoiding the risk of gas embolism. Their argu- ment is that oxygen is more readily absorbed by blood than is nitrogen, and if it is venous blood that is in question, this reasoning cannot be gainsaid. But it must be borne in mind that serious embolism can but rarely occur through the wounding of veins containing venous blood, namely those in the thoracic wall (see p. 156). It is from the lung that the danger of embolism is real, and the pulmonary veins contain arterial blood, which has, as Brauer reminds us, no special power of absorbing oxygen (p. 159). There is then, in realitj^ no decided advantage of Oxygen over Air or Nitrogen at a first operation, inore especially since, if a pulmonary vein is wounded, embolism must be at least as easily pro- duced by alveolar air at atmospheric or at times a slight positive pressure, as from the apparatus gas at a dis- tinctly negative pressure (see p. 158). The Avriter has always given air or nitrogen at first operations, and relies on other precautions for the avoidance of gas embolism. At Refills. — For refills oxygen gas is unsuitable on account of its high diffusibility — the choice must lie between the highly inert gas nitrogen and air in which nitrogen is diluted with 21 vols, per cent, of oxygen. Nitrogen has come into extensive or almost universal use, but it is doubtful whether its slight advantage over air entirely justifies the trouble and expense of olrtaining 32 PNEUMOTHORAX TREATMENT it. It has been made clear by experiment that whatever gas is put into the pleural cavity its composition is in a short time entirely altered liy exchange with the gases in the blood and by the eflcct of the local respiration of the tissues. This was first recognised by Jolm Davy as long ago as 1823. Having found that the gas in a case of spontaneous pneumothorax consisted of a mixture of No 92% and CO3 8%, and in another case of N2 93% and CO2 7% he enquires— " What had become of the oxygen that had disappeared ; whence the carbonic acid gas with which the azote was mixed ? " To resolve these diffieulties he conducted on dogs numerous observations designed to test the absorption of various ' gases from the pleural cavity. Of recent j^ears William Ramsay, analysing the gas in a case of pyo-pneumothorax treated with oxygen replacement by Morriston Davies, fovmd the pleural content after the withdrawal of 1,250 c.c. of fluid and introduction of 750 c.c. of O2 to be— No 74-09%, 0^ 21-56%, CO2 4-35%. Four daj's later the oxygen had been practically absorbed and replaced, the content being— No 98-13%, O2 0-69%, COo 1-18%. Tobiescn of Copenhagen gave pneumo- thorax fillings of air, oxygen, nitrogen, and carbon dioxide, and bv the next reflU he always found the same mixtiu-e present, namely N2 90 °o, COo 6-T%, and O2 3-4 °o- Webb, Gilbert, James, and Havens found that within 24-48 hours in cases hllcd A\ith nitrogen. Oo 4% and COo 8% had appeared on the scene, and with air a pre- cisely similar result ocem-red, showing that some 15*^0 of oxygen had diffused into the blood at the same time that the COo had conie from the blodd. When oxygen is used for a filling, this remains in larger quantitv. as Oo 14%, COo 6% and No 80%, though olniously re- placed to an enormous extent by other gases. The authors argue that owing to tliis rapid diffusibility of gases, there can be no advantage of nitrogen over air in treatment by artificial pneumothorax. Though these various observations go some wa)' to support this conclusion, they cannot show to what extent in the PNEUMOTHORAX APPARATUS 33 respective cases the lost gases had been replaced, and hence how far the total volume was diminished after the 24-48 hours' interval. Notably it could not be said how much, if any, of the lost 15% of O, when air was used, was replaced by fresh nitrogen, nor whether the total loss was not in this case, on account of the high diffusibility of oxygen, decidedly greater than in the case where nitrogen alone is utihsed. The question is whether, in practice, the presence of some 21% of a more diffusible gas with the nitrogen shortens the efficiency of a refill to any recognisable extent. This question may be answered to some extent at least by the following record of refills from a case under t;he writer's care : — Mr. P. W. P 3.1.15 -10-1 620 CO. N. -4+5 22.1.15 - 10 to 660 c.c. N -5+5 11.2.16 -10-1 600 c.c. N -4+5 3.3.16 - 9- 1 560 c.c. N -4+5 24.3.16 - 8 too 460 CO. Air -4+5 14.4.16 - 7+ 1 400 c.c. N -4+5 5.5.16 - 8+ 1 490 c.c. N -3+6 26.5.16 - 8to0 520 c.c. Air -3+5 16.6.16 -8-1 560 c.c. N -3+5 Here we see, and the same can be shown for other cases, that the efficiency of air and nitrogen, as measured by their absorption in identical intervals of time, was hardly, if at all, distinguishable. Indeed for practical purposes air can be made to replace nitrogen for refills whenever convenience may dictate, and the Writer has almost abandoned the use of the latter gas during the past nine months. This does not, however, apply to complete gas replacement of fluid (see p. 143), where the presence of this amount of oxygen must exert a decided influence on absorption. TUBING The apparatus, including the tubing, must be sterihsed before putting it together for use ; afterwards the 3 34 PNEUMOTHORAX TREATMENT nside of the tubing below the filter, the only vital part, can be kept sterile most conveniently by closing the end with a glass rod as in Pearson and Lillingston's apparatus. An addition which should never be omitted is a small length of glass tubing inserted close to the needle ; this forms a window through which fluid may be seen should any enter the tubing in a case of pleural exudate (p. 150), and also blood under exceptional circumstances in the case of a wounded vessel. NEEDLES A large variety of needles have been used by different operators for the introduction of gas into the chest. For practical purposes they may be divided into those suitable for a first operation and those suitable for refills, since the indications in these two cases are some- what different. For Initial Operation. — The requirements of a needle for the first production of a pneumothorax are some- what complicated. It must be capable of piercing the skin and superficial structures with ease, but it should be blunt at the moment of its passing the parietal pleura Fio. 2. — ^Writer's Trocar and CANNtrLA roR first Operations. so as to avoid any ribk of wounding the lung. These requirements are best satisfied, in the writer's opinion, by the use of a trocar and cannula, such as that depicted in fig. 2. This instrument is provided with an obturator to prevent the admission of air ; the cannula of 1 '8 mm. ("072 inch) gauge has a side opening near the tip, is graduated in centimetres, and carries a sliding guide to PNEUMOTHORAX APPARATUS 35 regulate the distance it is to be thrust into the chest wall. With the skin held stretched the sharp trocar and cannula are pushed in up to the guide fixed at 1 cm. or less from the point (see p. 44), or till the shoulder of the cannula is well through the skin. The trocar is now drawn out into the obturator, the tap of the needle closed and obturator and trocar removed. The un- armed cannula, whose edge should, however, be sharp, is now carried forward through the parietal pleura. Kiiss pushes a cannula and sharp trocar down to the external intercostal muscle, and then replaces the latter by a blunt stilette for its further passage into the pleural cavity. The blunt stilette is grooved on one side so as to bring the side opening of the cannula into connection with the manometer. Murphy used a trocar and can- nula, but in addition notched the skin with a tenotome ; Morriston Davies in this country adopts a similar pro- cedure for the introduction of an obtuse-ended hollow needle. Other forms of blunt instrument in occasional use are those of Courmont, Adolph Schmidt, and Deneke. Many operators make use, both at initial operations and refills, of fine hollow needles with a point cut ob- liquely so as to form a cutting edge, whereby the danger of wounding the lung is notably reduced. Of these the needles of Forlanini and Brauns and Sahli are examples. The former uses for ordinary purposes a needle of 1"1 to 1'2 mm. external diameter, but in cases under a high positive pressure needles as small as '8 mm. to "4 mm. He claims, on experimental grounds, that injury to the lung never occurs with the use of these needles in his own hands. The writer has made extensive trial of a Brauns needle of '8 mm. (•032 inch) external diameter, but has found it so frequently become blocked, either with a particle of punched-out epidermis or with a drop of moisture, that he has finally abandoned the use of fine- calibred needles. Three centimetres is a sufficient length for an instrument used for first operations, but for refills a length of 5 centimetres will be needed on occasion. 36 PNEUMOTHORAX TREATMENT For Refills. — When once a gas collection has been formed, there is no longer danger of wounding the lung. A sharp pointed hollow needle is then indicated, whereby the chest wall can be pierced with ease. The needle should not be too small in bore (see above), and the Saugman needle of at least 1*2 mm. ('048 inch) external diameter answers the purpose sufficiently well. This may be provided with an obturator, as is done in the most modern pattern, or the end may be covered, to prevent the entrance of air while clearing, by a thin rubber cap, through which the stilette is passed. There is, however, the disadvantage with this needle that it may punch out a piece of epidermis in its passage through the skin, and carry this with it into the pleural cavity. This can be avoided by the use of a solid Fio. 3. — The Lxllingston Needle. Identical with the Saugman needle, but with the terminal opening closed with metal. pointed needle with a side opening, such as that of Kjer- Petersen used by Saugman himself at refills, or in this country the Lillingston needle, which is practically a Saugman with the end opening filled in and a side opening provided in its place (fig. 3). These needles, if properly dried before operation and if the local anaesthetic is allowed a few minutes for absorption, practically never get blocked at refills in a dry pleural cavity. Consequently the obturator and stilette are but seldom needed, and with the older pattern of needle the writer dispenses with the rubber cap on all but rare occasions. The stilette must, however, never be passed without protection from the outer air being provided by one or other of these means. The method of open incision down to the parietal pleura introduced and practised by Brauer can be dis- PNEUMOTHORAX APPARATUS 37 missed without further description, as it has now been almost universally abandoned. The " safety syringe " of Forlanini needs a word of reference ; it is made to fit on to the posterior free end of the needle and, by the use of suction, to exclude at a first operation the chance that the end of the needle has entered a vein. Saugman has rightly condemned it as liable to lead to blocking of the needle and, indeed, if the characteristic manometer fluctuation is always waited for as evidence of the position of the needle point, the safety syringe should at best be a superfluity. The only condition under which its use may be recommended is that where a large pleural effusion is to be replaced with gas, and the manometer fails to register, owing to the needle becoming blocked with fluid. Under these circumstances the suction of some fluid into the syringe supplies the required proof that the end of the needle is in the pleural cavity (see p. 150). PREPARATION OF THE APPARATUS FOR USE To prepare the apparatus for use, the tubing is steri- lised in boiling water, and afterwards carefully drained and dried ; the bottles are filled with an antiseptic solution, such as 1% or 2% Izal or Lysol, or 1 in 1,000 perchloride or biniodide of mercury. When this has been accomplished and the apparatus again put together, the bottles should be firmly corked with the gas bottle still full of fluid, but the other emptied of all but enough to cover the lower end of the longer glass tube, a clip being placed on the rubber tubing between them. To fill with gas, the bottles should be detached from the manometer, and the fluid in the gas bottle replaced by nitrogen in the following manner. The taps of the nitro- gen cylinder are first turned on so that gas escapes in a slow stream ; the clip is taken off the tubing between the bottles and the rubber tubing leading to the upper end of the gas bottle is then connected with the cylinder. The fluid is thus driven back into the pressure bottle 38 PNEUMOTHORAX TREATMENT and replaced by gas. At the end of this operation the gas must be brought to atmospheric pressure, either by opening the tubing to the air, or by bringing the fluids to the same level before fixing on the clip between the bottles. If air is to be used, it is sucked into the gas bottle by syphon action through a suitable alteration of the fluid levels. It has been recommended that the gas bottle should be filled twenty-four hours before use, to allow dust particles to settle : most conveniently it is refilled at the end of an operation so as to be in readiness for the next. The tubing for the gas bottle should be re-connected with the manometer tubing before it is put away so that their inner parts may be protected from dust contamination. The interior of the tubing connecting the cotton wool filter with the needle is kept sterile, as already men- tioned, by the closing of its extremity with a glass rod. It and the filter should, however, be resterihsed from time to time, and in particular after use in a case where fluid from the pleural cavity may have entered its end. It is well to have a spare tube ready for use in cases with pleural effusion. The needles, both pneumothorax and hypodermic, must be always kept clean and sharp, and a small hone, specially kept for this purpose, is an indispensable requirement. CHAPTER IV TECHNIQUE OF INITIAL OPERATION NURSING HOME OR PRIVATE HOUSE The operation for artificial pneumothorax can be quite easily performed in an ordinary bed and bedroom in the patient's own home. Nevertheless there is one important advantage which a Nursing Home or Hospital may have over a private house, and that is the presence of an X-ray installation. If this advantage can be obtained it is an important gain, for the X-rays during the early weeks of treatment are of very great assistance in supplementing the information derived from physical signs, particularly in deciding on the presence and posi- tion of adhesions, and in determining the moment when satisfactory collapse of the lung has been obtained. Nevertheless treatment can be carried through at these stages, and even altogether, without any use of the X-rays, and the writer has had to submit to this experience on many occasions. Forlanini has remarked : " The clinical examination alone suffices, in the absence of radioscopic apparatus, for the administration of the treatment." If treatment is carried out in the patient's home, the presence of a trained nurse is an almost indispensable convenience during the first few weeks — later she can be dispensed with, and finally the patient will be able, in most cases, to attend for refills in the doctor's consulting room. PRELIMINARIES The patient should be kept in bed for some days prior to the initial introduction of gas — this will probably be 39 40 PNEUMOTHORAX TREATMENT indicated on other grounds, since the pneumothorax case is generally a febrile case, " ambulant febrile " at least. During this time the temperature, pulse, and daily amount of sputum should be recorded on a four-hourly chart. It is convenient, also, during these preliminary days, to examine the chest carefully with a view to the selection of likely sites for puncture. These should be duly noted in order of their suitability so that no needless delay may occur on this account at the time of the operation. SELECTION OF SITES FOR PUNCTTJRE The possibility of producing an artificial pnemno- thorax depends on the discovery of free pleural space, and the comfort of the patient demands that this object should be attained with as few punctures as possible, and preferably at the first. In the selection of sites for puncture, certain facts must be borne in mind. Firstly, the base is a much more hopeful region than the apex, both because it is fa,rthest from the disease area, and because the free movement of the lung border tends to keep its surface free. Moreover the axilla is a more likely spot than the front or back of the chest, largely because the respiratory excursion is the greatest at this point. This respiratory excursion, the tidal movement of the lung border between deep inspiration and expira- tion, or deep inspiration and quiet breathing, should be marked and measured in front, behind, and in the axilla with a view to deciding on which surface pleural adhe- sions are least likely to be present. In this connection it must be remembered that the lung excursion differs normally at these three points, being largest in the axilla (10 to 12 cm. difference between forced inspiration and expiration), smallest behind (3 or 4 cm.), the front being of somewhat wider extent than this latter. If a free tidal movement is present, the production of a pneumothorax is practically assured — if, as often happens, there is little or no tidal fluctuation, the discovery of TECHNIQUE OF INITIAL OPERATION 41 free pleural space at some point is still possible since the limitation of movement may depend on adhesions above the base, or on diminished expansile power of the lung itself. The axilla, often the spot most free of pleural adhe- sions, has also the advantage that the intercostal spaces are wide and that the chest wall between the pectoralis and latissimus dorsi is but thinly clothed with muscle, and hence presents an easy and convenient site for puncture. The writer is accustomed to make a first puncture in this region, generally at the edge of the lung, as for instance in the sixth space in the anterior axillary line, unless there exist special reasons against it. Thus in some cases with marked basal disease it may be fairly certain that adhesions are present in this neighbourhood, and a better chance may offer itself at a point higher up the chest wall, often in the second or third spaces in or outside the mammary line. The point farthest from the older foci of disease and where signs of disease are absent or least marked is the point where free pleura is most likely to be found. Such points are generally near the base, and if a couple of punctures in the subaxillary region, one in front of this, and one at the back, all near the lung border, fail to find a pleural space, it is very unlikely that further attempts will be successful. Saug- man remarks that a useful pneumothorax is generally produced at the first puncture, rarely at the second to fourth, and never after this. Deneke has reminded us that the presence of pleural friction is evidence of the absence of adhesions. In a couple of cases with widespread adhesions the writer has been helped to success by the presence of a small amount of pleural fluid at the base. In such cases it will generally be found that change of position exerts a certain slight influence on the position of the fluid, and if the needle can be entered at a point where the dulness of fluid has been, by change of position, re- placed by resonance, a successful puncture is practically assured. The neighbourhood of cavities, particularly 42 PNEUMOTHORAX TREATMENT superficial cavities, is quite unsuited for puncture, partly because strong pleural adhesions are almost sure to be present, and partly because there is risk that, if the puncture fails, the needle may wound the cavity wall, and this wound be opened up into the pleural space if the operation succeeds elsewhere, and the adhesions at this point are subsequently broken down. The same danger applies to caseous areas and areas of marked consolidation, the wounding of which is hkely to favour a perforation of the lung, if an artificial pneumo- thorax is finally achieved. THE OPERATION Preparation of Patient. — Do not operate after a large meal. The procedure must also be so arranged as to avoid the period of the catamenia in women, since nervous hypersensitiveness at these times heightens the risk of pleural shock. A nervous patient may be treated with bromides as recommended by Kuthy, and some diffu- sible stimulant as alcohol or Spiritus Ammonise Aroma- ticus may be given just before the operation. About half an hour to one hour before the initial operation, an injection of Morphine (gr. ^^) with or without Atropine (gr. ihis), or Omnopon (gr. ^J), a satisfactory preparation for these cases, must be given. Preparation of Instruments and Apparatus.— A low table covered with a clean towel must be placed beside the head of the patient's bed. The pneumothorax apparatus is stood thereon wide open and with the manometer turned to face the operator ; the bottles are taken out and, if not previously filled (as is best so as to allow dust to settle), are now charged with N gas or air. The tubing is seen to be properly joined up, and the long tube to which the needle will be attached is coiled into a bowl of 1 in 20 carbolic, its end being closed by a glass rod. In addition the following instruments must be at hand : TECHNIQUE OF INITIAL OPERATION 43 Tincture of Iodine and some wool swabs and dressing forceps to apply it. Sterile towels. A sterilised hypodermic syringe with two fine and sharp sterilised needles. A bottle or ampoule of sterilised 2% Novocaine in normal saline, or in 1 in 10,000 Adrenahn solution. The Pneumothorax needle or needles lying in absolute alcohol. A bowl of hand lotion. Ether and other emergency remedies (see p. 155), and a spare hypodermic syringe and needle to apply them if required. Position o£ Patient. — The patient lies supine, but turned somewhat on to the sound side, so as to bring the proposed site of operation directly uppermost (see also p. 65). Saugman makes a great point of having the area of operation highest, so that a gas bubble, with its tendency to float upwards, may collect round the needle point before this can reach the lung. But it seems clear that the gas from the manometer tubing remains round the end of the needle for other reasons, since otherwise the respiratory fluctuation characteristi.c of a successful pleural puncture would not occur in other apparently less suitable positions, and this is not the case. The head is kept low to reduce the risks of gas embo- lism, and a small firm pillow is placed under the back and sound side, so as to stretch and open out the inter- costal spaces round the seat of operation. The arm is stretched above the head and may, if required, be supported by a nurse. The selected site or sites of operation are now found and marked with a skin pencil, and at the same time it is expedient and convenient to mark out the border of the heart and the lower limit of the lung, so as to keep the anatomical boundaries in clear view. Passing the Needle. — The operator now steriHses his hands and places a steriUsed towel round the site of 44 PNEUMOTHORAX TREATMENT operation, while an assistant or nurse swabs a wide area of the chest wall with Iodine lotion. He then fills the hypodermic syringe with novocaine, places it on the sterile towel, and takes the pneumothorax needle and. stilette out of its alcohol. These he proceeds to dry thoroughly by heat in the flame of a spirit lamp, heating particularly the needle end, and passing the heated stilette several times through its lumen, so as thoroughly to dry it. It is essential that the needle contain no atom of moisture when it is ready for use. The needle is now placed on the sterile towel, the operation spot selected, and the novocaine syringe taken in hand. A small super- ficial, intracutaneous bleb is first made with the syringe sloped. The needle is now moved to the upright position without withdrawing it, passed into the intercostal space near the upper border of the rib, and pushed forward very slowly in fractions of a centimetre, some novocaine being forced out at the end of each move forward. Enough must be left in the syringe to anaesthetise the parietal pleura, the whole or main object of this pro- cedure, and this will be found at a depth varying from 1 to 2 1 or even 3 cm., according to the site and the patient's nutrition and muscular development. In a thin patient the pleura may be only a bar« centimetre from the skin. If the ansesthetisation is done sufficiently slowly, and it should take 1^ minutes at least, no pleural pain may be felt — more often a sudden pain is experienced when the pleura is reached. Rarely, it appears, an intercostal nerve may be injured, and cause pain and tenderness of a more widespread and lasting character. The hypodermic needle is now withdrawn, and the pneumothorax trocar and cannula is fitted on to the end of the proper tubing, which is removed from the bowl of carbolic lotion for this purpose. The guide on the cannula if used is placed at about f cm. from the point, the trocar kept firmly in place by pressure of the iiollow of the hand on its upper end. With the first and second fingers of the left hand the skin of the anaesthe- tised point over the intercostal space is held tightly TECHNIQUE OF INITIAL OPERATION 45 stretched to each side, and the trocar and cannula are entered swiftly and firmly, but without any stabbing movement, up to the guide, noting carefully that the shoulder of the cannula is well through the skin before the trocar is withdrawn. The trocar is then pulled back into its stuffing box, the tap closed, and trocar and stuffing box removed. The cannula is now driven for- ward by careful, steady pressure, attention being closely fixed all the time on the manometer, and the final " snap " which announces perforation of the parietal pleura being carefully looked for. Before this is reached a painless resistance is often experienced at the inner muscular fascial layer ; the patient generally experiences some pain at the moment when the pleura is reached and pierced. Manometric Response. — Just outside the parietal pleura a feeble respiratory fluctuation about the atmospheric level may be observed. At the moment the end of the cannula enters the pleural space, the manometer fluid gives a sudden swing to a negative pressure of some 10 cm. of water, and continues round this point with clean respiratory fluctuations of some 4 to 8 cm. Such a clear indication of a high negative pressure with large respiratory fluctuations is complete proof that the end of the cannula is in the pleural cavity, and N gas can be entered forthwith. Entrance of Gas. — This is done at a negative pressure about one-half to two-thirds that in the pleural space, a measure which is accomplished by so placing the bottles before the operation that the column of fluid in the gas bottle stands some 7 or 8 cm. higher than that in the pressure bottle. The gas is thus sucked into the pleural cavity at a slight negative pressure, as soon as the cHps on the tubing are removed, a manoeuvre which probably reduces the very small risk of gas embolism to a minimum. As soon as a definite rise in the level of fluid in the gas bottle has demonstrated the fact that gas has entered, the fear of embolism may be finally set aside, and the bottle fluids placed at the same level, so that the gas flows in under ordinary atmospheric pressure. 46 PNEUMOTHORAX TREATMENT Amount of Gas at first Operation. — At a first opera- tion it is convenient to put in enough gas to maintain an adequate supply against the next puncture and " refill." Where the gas goes in easily 300 c.c. is a con- venient quantity, and this should be put in in all cases where no pain or distress occurs, and where the intra- pleural pressure is not brought above atmospheric pressure. Saugman recommends 200-250 c.c. and never above 300 c.c. at the first operation ; nevertheless the writer sometimes gives 400 c.c. where the succeeding interval has to be as long as four days and where this amount of gas disturbs the pressure but little, and he has never experienced any ill effects from it. With a free pleural cavity these quantities of gas suffice to raise the pressure but little — it remains markedly negative (see Case of W. N., p. 57). Where only a moderately free pleural chamber is present the pressure may rise to the atmospheric (Case of H. B., p. 55). Where, on the other hand, only a small pleural pocket is opened up, the pressure may rise above the atmospheric after only 50 or 100 c.c. or little more has entered. With so small a cavity a rise of pressure even greatly above that of the atmosphere will cause no disturbance, unless a certain amount of pain, and in such cases the con- venience of getting in at any rate 100 or 200 c.c. must be duly weighed against the patient's discomfort. The ease and safety of the next puncture depend largely on the certainty of finding an adequate gas pocket into which the needle may be entered without fear of wounding the lung. In the early days of pneumothorax therapy it was customary to enter a large volume of gas, 1,000 c.c. or more, at the first filhng, but this has been generally abandoned as at least inadvisable. In cases even where a large quantity appears to be readily borne at the time, it has commonly happened that symptoms of distress, dyspnoea, sweating, and the like have followed after an interval, and a smart febrile reaction on the following day. Apart from this, also, there is a very TECHNIQUE OF INITIAL OPERATION 47 real danger of infecting the other lung by the large quantity of sputum squeezed out of the collapsing cavities and tubes of the diseased lung — a quantity too large to be immediately and safely removed by cough- ing. Zink, who formerly employed 1,000-1,200 c.c. at first fillings, records no less than 280 c.c. of sputum coughed up by one of his patients. Robinson and Floyd have recommended a first filling of 600-1,000 c.c. with the object of encouraging the patient by the rapid improvement of symptoms which is likely to follow it. In cases where the pneumothorax operation is under- taken for haemoptysis, an immediate partial collapse of the lung is aimed at, and a large quantity of gas, 1,000 c.c. or more, must needs be run in without delay. Manometer Readings. — During the time that the gas is flowing in the manometer is kept in connection with both the pleural cavity and the gas bottles, and registers, in practice, the pressure in the latter. This pressure, the pressure at which gas is being entered, is kept at any desired figure by an adjustment, by means of blocks or other apparatus, of the heights of the columns of fluid in the two bottles. By cutting off the gas bottle by finger pressure the manometer can be made, at any moment, to record the intrapleural pressure, and when this rises to the level of the pressure in the bottles, and the flow of gas stops, the pressure bottle must be again raised if more gas is required to be put in. Removing the Needle. — ^When the desired amount of gas and pressure is reached, the clamps are replaced on the tubing of the bottle system, the manometer reading is recorded, and the pneumothorax needle withdrawn. In extracting the needle it is advisable to lift up a large fold of the chest wall with it, and after its withdrawal, by deep manipulation and rubbing of this to obliterate the needle track as far as possible, so as to prevent the occurrence of surgical emphysema. No dressing is needed for the small puncture wound. The patient will probably be quite comfortable after the operation, and will very likely be drowsy from the effects of the 48 PNEUMOTHORAX TREATMENT morphine. If there is any pain or discomfort a light poultice will, as a rule, readily relieve it. Effect on Chest Signs. — The effect on the physical signs of a 300 c.c. first injection is usually remarkably small, and is generally more noticeable some hours later than immediately after the operation. An area of in- creased resonance of varying extent, and some muffling of the breath sounds and rales, are very commonly demonstrable about the spot where the gas was injected, and this is particularly interesting when it is borne in mind that the old belief in " cohesion " of pleural sur- faces is said to be definitely disproved. Indeed Emer- son has shown by experiments on dogs that a small amount of gas introduced into the healthy pleural cavity always finds its way to the highest point in whatever position the body is placed. In the presence of disease of lung and pleura, such as occurs in artificial pneumothorax cases, this impartial wandering of gas probably takes place to a much smaller extent. Hints for First Operation. — ^Not to put in more than 300 c.c. of gas as a rule. Not to raise pressure by more than 10 cm. HjO. Not to attempt more than two punctures at first operation, as thereby the risk of pleural reflex is materi- ally lowered. In this connection Saugman recommends two punctures at a first sitting, and, if these fail, three or four attempts at a second sitting, if the patient is not over-nervous. He remarks that an efficient pneumothorax is generally found at the first puncture, rarely at the second to the fourth, and never after this, though in his earlier practice he was accustomed to puncture up to twenty or thirty times. DIFFICULTIES ATTENDING INITIAL OPERATION In the foregoing description it was presumed, for convenience, that the puncture discovered a free pleural space, as evidenced by a characteristic manometer reading, and nothing remained but the introduction of TECHNIQUE OF INITIAL OPERATION 49 gas. Unfortunately things do not always proceed so smoothly as this, and certain possible variations from the optimum must receive consideration. (1) Absence of Manometric Response. — The charac- teristic " snap " which generally announces the pas- sage of the needle through the parietal pleura may occur, but the expected manometer fluctuation fail to follow. This may indicate closely adherent pleura, or it may be due to a blocking of the needle with blood or other fluid, or with solid tissue. The latter possibility is readily excluded by passing the stilette through the needle, the proper use of the stuffing box not being neglected. If this fails adherent pleura may be pre- sumed, but in some cases there will remain a doubt as to whether the pleura has really been passed, and the cannula may then be cautiously pushed forward to see what further obstruction meets it. No force must be used or an adherent lung may be torn — the educated touch can generally appreciate the inelastic resistance of adherent lung, and the blunt cannula end will not penetrate this (unless a soft caseous patch is encoun- tered) if proper gentleness is used. Those who use, the writer thinks unwisely, a pointed needle for the first operation, will doubtless pass it onward into the lung. In this case two possibilities are open to them : (a) They may enter the lung parenchyma or a bronchial twig, when manometer fluctuations of 3 or 4 cm. II2O round the atmospheric pressure, or only a trifle below this, will occur. Saugman says that the fluctuations with the end of the needle in the lung may range as large as — 6 to +4 cm., and, rarely, if the portion of lung is shut off from a bronchus, a negative pressure as high as — 3 or — 4 cm. mean pressure may be observed. (b) The needle may perforate a vessel, when a small fluctuation occurs, but soon stops from clotting in the needle and the resistance of capillarity. In the case of a vein, blood may not enter the needle but gas may be sucked into the vessel (see Gas Embolism). 50 PNEUMOTHORAX TREATMENT (2) Incomplete Manometric Response. — When adhesions are present and are not very dense, the manometer may register a fall of pressure of a few centimetres accompanied by some amount of respiratory fluctua- tion. As a rule the fluctuations will be irregular in extent and vary with slight movements of the needle. Under these conditions no clear evidence of being in the pleural cavity exists, and the needle is withdrawn and entered at another spot. (3) Negative Pressure without Respiratory Fluctuations. — After apparent perforation of the parietal pleura, a large negative fluctuation (10 cm. or so) of the manometer fluid occurs, and this then remains fixed. So large a drop of pressure is proof that the end of the needle is in the pleural space and the lumen of the needle may have become blocked, but a similar phenomenon seems to occur in the presence of loose adhesions among which the manometer gas becomes lost and is not returned. Under these circumstances the writer has been accustomed to withdraw the needle and try again near the same spot or elsewhere. We have it on the authority of Saugman that with a negative pressure of 12 to 15 cm. H2O the certainty of being in the pleural cavity is absolute, and gas may be safely run in although the fluctuation is in abeyance. The writer would object to this that when adhesions are present it is difficult to keep the needle point in the pleural space, and though it may have been there at the moment of the manometer fluctuation, evidence that it is still there at the moment the gas should be entered is, in the absence of manometer fluctuations, altogether wanting. (4) Pleural Pockets. — On passing the parietal pleura the usual manometer fluctuation occurs, but 60 or 70 c.c. of gas suffice to raise the pressure above the atmospheric with but weak respiratory fluctuations, showing that a pleural pocket has been entered. The pressure should now be raised to + 5 cm. H2O or there- abouts, and if there is any outlet it will fall again TECHNIQUE OF INITIAL OPERATION 51 gradually and more gas can be entered. If the pres- sure does not fall and respiratory fluctuations continue, a further pressure up to 10 or 15 cm. H2O (Saugman) may be tried, but no more at a first operation. At subsequent fillings the pressure may be raised to 20 or 25 cm., and, in many cases, more in an attempt to obtain an efficient pneumothorax (see p. 79). If the pocket found seems to offer small hope of success, the needle may be withdrawn and search made for a larger pleural space to work on or, failing its discovery, two or more pockets may be enlarged simultaneously with a view to uniting them, after the manner recom- mended by Brauns. It has been asserted by v. Muralt that the intrapleural pressure, taken with the breath held, affords useful information of the capacity of the free pleural space entered, this showing a negative pressure of — 10 to — 4 according to the size of the pocket. This seems to the writer a very doubtful proposition, since among his cases the negative pressures in quite small pockets were asi high as those where a complete pneumothorax was subsequently attained. Even in a pocket too small to admit a measurable quantity of gas the writer has obtained a fluctuation of — 13 to — 6, and this seems to be confirmed by the experience of Saugman. The following initial pressures anaong cases of the writer's serve to exemplify this point : — Initial Pressure in Oases of Partial Pneumothorax Partial Pneumothorax Complete Pneumothorax. Oases. Efficient. Oases. InetEcient. -16-8cm. H2O. - 12 - 4 cm. H2O. -13-9cm. H^O. -12-8,, „ -15-7 „ „ -12-8,, - 14- 1 „ „ - 16- 9 „ - 13- 4 „ „ - 18-12,, (5) Puncture of the abdominal cavity through an unexpectedly high diaphragm, a rare and in most cases avoidable accident, shows a negative pressure with reversed respiratory fluctuations. In some cases, however, the fluctuations are said to follow the normal type at first, and only become paradoxical after some gas has been allowed to enter. CHAPTER V EFFECT OF PLEURAL ADHESIONS: PARTIAL AND COMPLETE PNEUMOTHORAX Pleural adhesions are the rock on which the immediate success of pneumothorax therapy is most commonly wrecked, for without a sufficiency of free pleural space an efficient pneumothorax cannot be produced. Adhe- sions entirely prevented the production of a pneumo- thorax in 11% of Saugman's cases, in 24% of Zink's, in 15% of Hamman and Sloan's, in 12 "5% of Keller's, in 7-5% of Weiss's, in 9-1% of Brauns', and in 20% of the writer's cases up to date. Among 757 American cases collected by Sachs, failure was recorded in 21 "5%. But even where a clear space can be found and gas be introduced, adhesions may prevent sufficient collapse of the lung to render the pneumothorax therapeutically useful. Such a result sufficed to swell the figure for adherent pleura to 27% of Zink's cases, 22% of Saug- man's, 37-5% of Keller's and 11-2% of Braun's, and 37% of the writer's. The marked discrepancies of these figures is due doubtless to the different class of cases selected for treatment by these several physicians — thus the high percentage of adherent pleura among the writer's material is certainly due to the fact that he has been content to confine this treatment for the most part to cases of advanced " unilateral " disease, for the farther advanced the disease the greater the likelihood of widespread adhesions. The common presence of adhesions, to a greater or less extent, among cases submitted to pneumothorax treat- ment, and the mechanical effects due to their presence, 52 EFFECT OF PLEURAL ADHESIONS 53 dictates a natural division of pneumothorax cases into three fairly well-marked clinical groups. PARTIAL PNEUMOTHORAX Ineffectual. — These are cases where a small pneumo- thorax is produced, often with difficulty, and all sub- sequent attempts to enlarge it to a useful extent fail. It is often difficult to prophesy at the outset how far a partial pneumothorax may benefit a patient. The writer has seen arrest and cure of disease produced by gas collections filling less than one-half of the pleural cavity. It can never be determined beforehand, either, whether a pocket of gas can or cannot, under suitable pressure, be expanded gradually into an efficient pneumo- thorax. Experiment is, in both cases, the only reli- able criterion. Where, however, to give a gross example, in a patient with extensive cavitation only a small and remote gas pocket can be achieved, and where a few fillings at 35 to 40 cm. H2O pressure fail to exert any influence on it, it soon becomes obvious that it is useless to proceed. Examples of partial pneumothorax which refused to yield to pressure and were abandoned as ineffectual are here given. H. B., aged 34 years. — Under my care on different occasions since 1905, with attacks of haemoptysis up to 2 pints on some occasions. Moist sounds over the whole left side, but no cavity signs and no certain cavity visible to X-rays. Sputum 2 or 3 oz. daily. General condition good. No less than seven attempts to find free pleura were made on three occasions and failed, but at the eighth, on 16.2.15, a free pocket was found at the base behind and 350 c.c. of N put in. J6.2.15 . - 16 - 9 cm. H^O. 350 c.c. + 4+ 12 cm. 1120, falling in a few minutes to -4+ 4 cm. H2O. 17.2.15 . -10-1 „ „ 160 c.c. -fl7+20 „ „ soon falling to + 15+ 18 cm. H2O. 19.2.15 . - 9-1 „ .. 100 c.c. +33+36 „ „ soon falling to + 29+ 32 cm. HjO. 54 PNEUMOTHORAX TREATMENT A pocket at the base rising about a quarter to one-third of the way up the back was thus formed, and gas was put in at a pressure just under + 40 every week, later every two, and finally every three weeks, the pressure on each occasion slowly but steadily falling from a leakage of gas inside the chest, presumably among the adhesions. The amount of gas was generally about 300 CO., and this could be increased to 400 c.c. or so by waiting for it to spread itself, thus : — 21.12.15 . -9-4 300 c.c. + 35+37 falling 11.1.16 . . -10-4 350 c.c. + 30+34 „ 1.2.16 . . - 8 - 3 325 c.c. + 33+37 „ 22.2.16 . . -9-2 400 c.c. + 34+ 38 „ The haemorrhages were somewhat controlled but not abolished, and treatment was persisted with for nearly eighteen months at the patient's desire, and with a view to testing the possibility in such a case of achieving success by persistent attack. The pneumothorax cavity, however, showed no increase in size, whether to X-rays (see Plate I, facing this page) or to physical examination ; it was but doubtfully effective, and has now been finally abandoned. O. M., aged 25 years. — Chronic disease of whole right side, with cavitation below the apex and moist sounds to base ; left side practically free of signs. T. B. + Heart pulled over to right. [ 9.5.16 . At second puncture free pleural spaec found at fifth space in mid axilla. - 18 - 12 cm. H3O. 300 c.c. + 6+ 8 cm. H2O. rapidly falling 350 c.c. + 15 + 18 cm. H.O. soon falling to + 8+ 10 cm. H-O. 225 c.c. +20+21 „ falling 200 c.c. +26+28 csi. H3O, falling 200 c.c. +34+ 36 cm. HjO, falling 200 c.c. +28+29 cm. H2O. falling Pleural pain was experienced near the end of each earlier filling, generally at the base, but on one occasion conducted to the shoulder region. Only a very small basal pneimiothorax was formed, and as it yielded little to pressure and could obviously exert no important influence on the extensive cavitation near the top of the lung, the treatment was abandoned. 12.5.16 . - 8 - 4 16.5.16 . - 8 - 4 19.5.16 . - 4 to 23.5.16 . - 4 to 30.5.16 . -3+1 Platk I. — Kadtogeam of Chest of H.B., whose Case is described ON" PAGE 53. It shows a small pneumothorax at the left base, disease in the lung above, and quiet or arrested disease in the right lung. Taken Aug. 24, 191G, by Or. OYerend. 64] EFFECT OF PLEURAL ADHESIONS 55 Effective. — ^As already remarked the value of a partial pneumothorax must be decided by practical results. Often an inconsiderable pocket of gas can be enlarged by repeated refills at a gradually rising pressure. Forlanini gives these refills daily until the desired result is achieved. Brauns, in cases where only pleural pockets can be found, puts in gas at several adjacent points, and strives by a rising pressure to unite them. The possibility of enlarging an inefficient pneumothorax till it becomes of therapeutic value may depend not only on the strength of adhesions to be separated, but also on the resistance of surrounding structures to the needful pressure. In particular the mediastinum must be care- fully watched since its weakness may lead to displace- ment or bulgings which may set a clear limit to the amount of pressure which it is safe to employ. (See Mediastinum, pp. 105 and 106.) This difficulty does not often arise where the gas collection is still of small size. Adhesions may either separate or stretch, and to stretching the diaphragmatic pleural bands are said to be most amenable, and to give rise but seldom to difficulty. An example of a small pneumothorax enlarging under pressure till it became fully effective will here be given. H. B., aged 34 years. — Five months' history of dyspnoea, cough with expectoration, and loss of weight. Nutrition poor, sputum one ounce daily. T.B. +. Moist crepitations over the whole left side and extensive excavation about the middle of the lung giving typical cavity signs in front, axilla, and near spine. Right side practically clear. 23.3.14 . At second puncture free pleura found below tho scapular angle. - 12 - 4 cm. H2O. 250 c.c. - 4+ 2 cm. H,0. 25.3.14 - 7 to „ „ 450 c.c. + 4+ 9 „ „ 27.3.14 0+2 „ „ 350 c.c. + 9+ 11 „ „ 31.3.14 +3+7 „ „ 350 c.c. + 12+ 14 „ „ 7.4.14 - 1+1 „ „ 600 c.c. + 14+ 15 „ „ 17.4.14 0+7 „ „ 600 c.c. + 16+20 „ „ A small pneumothorax was formed at the base extending up to the fifth rib in front and about one-third up the back. Pres- sures up to about 20 cm. H2O were continued at each filling and in about three months the gas chamber extended up to the 56 PNEUMOTHORAX TREATMENT third rib, cavity signs were abolished, and crepitations had become inaudible even at the apex. Expectoration had practically ceased, and weight was rising. The heart was displaced to the right nipple line, but the patient experienced no inconvenience, and returned to his work as a cabinet maker about the middle of May 1914. The fillings were now larger. 30.6.14 . -1+4 . 975 c.c. + 19+ 21 21.7.14 . -5+1 . . 1,100 c.c. + 19+ 21 7.8.14 . -2+2 . . 1,075 c.c. + 20+22 28.8.14 . 0+ 6 . 900 CO. + 15+ 17 He kept in good health, the functioning lung remained clear of signs, as did also the adherent upper part of the collapsed lung. In the spring of 1915 he had a small pleural effusion which lasted about two months, and was then absorbed ; it did not in any way put him back, but diminished gas absorption to a convenient extent, so that he was able to retain his gas over a longer interval. He continues his work and returns every month for a refill as : — 4.1.16 . -10-6 . 525 c.c. + 17+19 8.2.16 . - 7-2 . 400 c.c. + 18+ 21 10.3.16 . -7-3 . 300 c.c. + 18+ 21 11.4.16 . -9-5 450 c.c. + 17+ 20 9.5.16 . -8-4 . 450 c.c. + 18+ 20 COMPLETE PNEUMOTHORAX This term must be used in not too limited a sense, since it is convenient to include under it those cases, all too common, where adhesions exist, often at the apex, without detracting from the clinical completeness of the result. Where, however, there is cavitation at the apex and the whole apical cupola is adherent, this may pre- vent or seriously impede the emptying and obliteration of cavities, and the case will fall into a different category. A complete pneumothorax is one where the lung as a whole is able to attain an efficient collapse, and a few adhesions, providing they are not too far from its root, may not sensibly interfere with this result. The earlier in their course cases are submitted to treatment the more often will a complete pneumothorax be obtainable ; nevertheless, with advanced one-sided disease it not infrequently happens that a fairly complete pneumo- EFFECT OF PLEURAL ADHESIONS 57 thorax is attained, even though this may occur at a fairly late stage of the treatment. An example of com- plete pneumothorax in a case of advanced disease will be here given with outline notes covering a year's treatment. W. N., aged 24 years. — A history of cough and wasting over many years. In-patient at City of London Hospital for Diseases of the Chest six years ago, and then had tubercle bacilli in the sputum. Rather thin, hoarse with tuberculous ulceration of the larynx, some dyspncea. Pulse 108 : Temp. 99°-I02-2° : Resp. 24. Chest emphysematous. Moist crepitations all over right lung to base': good tidal movement at base. Left lung practically clear. Heart, dulness wide, from L.N.L. to 2J inches to right of mid-sternal line. A radiogram showed disease of both lungs, on the right side large blotches of shadow throughout the lung, but most marked in the upper two-thirds ; on the left side smaller shadows extending out into two-thirds of the lung mostly towards its middle, and with a tendency to coalesce into larger masses near the root. 14.5.15 . Free pleura found at first puncture in fourth space close to nipple. — 13-4cm. H2O. 300 c.c. - 10-|- 2 cm. H2O 18.5.15 . - 12-t- 1 „ „ 500 c.c. - 8-f4 „ „ 21.5.15 . -lOtoO „ „ 550 c.c. -54-7 „ „ By this time breath sounds feeble all over right side and crepita- tions abohshed. Heart f inch outside L.N.L. Temperature here fell to normal and has remained so since (see Fig. 4, p. 58). 25.5.15 . -lO-t-2 . . 800 c.c. - 4-h 7 Right side breath sounds nearly abolished ; percussion note high pitched. V.R. and V.V. unaltered. Left side a few crepitations in lower half. Heart 1 inch outside L.N.L. 27.5.15 — X-ray screening. — Diaphragm moves freely ; inspira- tory suction of mediastinum to pneumothorax side. Lung dense and of largish size, lies against mediastinum and shows no respiratory movements. Unimportant adhesions at apex and base. 28.5.15 . — 6-^l . . 500 CO. -3-1-6 1.6.15 . -2-H2 • . 275 c.c. - 1 -f- 5 8.6.15 . -44-2 . . 625 c.c. 4-1-1-8 Left lung now free of crepitations. S8 Plate II. — Radiogram of Chest of W.N.. whose Case is described ON PAGE 57. It shows the collapsed right lung lying against the mediastinum, a strong adhesion at its base ; also areas of disease in the functioning lung. Taken on Oct. 28, 1915, by Dr. Overend. 159 EFFECT OF PLEURAL ADHESIONS 59 16.6.15 . -4+2 22.6.15 . -1+4 29.6.15 . -2+6 9.7.15 . -3+5 600 c.c. + 2+8 450 CO. + 4+7 350 c.c. + 1+10 400 c.c. + 1+9 Heart 1 inch outside L.N.L. and 2 inches to right of mid- sternal line. A few crepitations outside heart. Percussion note somewhat high pitched both sides of chest. 17.7.15 . 0+4 . . 275 c.c. +2+10 27.7.15 . Feels a "splash" when he moves. A little fluid found. 0+5 . . 240 c.c. +2+ 9 3.8.15 . Fluid now 2 inches up. 10.8.15 . Less fluid. -2+4 . . 300 c.c. +2+ 9 24.8.15 . Now only 1 inch of fluid. - 6 to . . 570 c.c. +3+10 3.9.15 . Fluid gone. It was ushered in with no symptoms and no fever, but temp, rose to 99°-99-5° in evening between 3.8.15 and 13.8.15. -4+1 . . 425 c.c. +2+ 9 14.9.15 . -4+3 . . 450 c.c. +6+10 24.9.15 . -3+5 . . 300 c.c. +6+10 12.10.15 . -1+6 . . 150 c.c. +1+8 Radiogram on 28.10.15 shows^tho shadows in the left lung are smaller and denser than at beginning of treatment five months ago, and tend to range themselves into a more reticulated outline (see Plate II, facing this page). 9.11.15 .-6+4 . . 250 c.c. 0+ 7 By now has gained 12 lb. in weight. 7.12.15 . - 8to0 . . 350 c.c. -1+9 5.1.16 .-10+1 . . 400 c.c. +1+10 1.2.16 . Has now gained 15 lb. in weight since beginning of treatment. Left lung quite clear to physical examination. -9-1 . . 420 c.c. +4+11 Left Hospital on 2.2.16 and returned for refills. 29.2.16 . Measurement R. chest, 17| in. L. „ 161 in. - 9 - 1 . . 300 c.c. +1+7 28.3.16 .- 12to0 . . 360 c.c. +1+7 2.5.16.-11-1 . . 450 0.0. +2+10 30.5.16.- 8+2 . . 250 c.c. -1+7 This gives an outline of gas fillings and pressures during the first year of treatment in an avetage case of complete pneumothorax. 60 PNEUMOTHORAX TREATMENT The patient left the Hospital early in February 1916, and has since lost some weight, but seems fairly well, though some sputum remains, doubtless from central areas in the functioning lung. This lung remains clear of stethoscopic signs, and has a tidal excursion of J inch on the posterior aspect. On account of his emphysema and some consequent cardiac embarrassment he is always decidedly dyspnoeic and complains of an increase of this " tightness " after his refills. DIAGNOSIS OF ADHERENT PLEURA In recommending the adoption of pneumothorax treatment in any individual case, it is right that some attempt should be made beforehand to gauge the likelihood of finding a free pleural space. Though this is notoriously a matter of difficulty in some cases, such points as are likely to be helpful in making the decision will here be outlined. In the first place, the earlier and the more localised the disease the less likely are widespread adhesions present. For this reason, among others, useless delay must be avoided as soon as it is clear in any case that other methods have failed to arrest the spread of disease, and that no clear contraindications to pneumo- thorax therapy exist. The decision must be prompt and must be put to early trial. On the other hand, disease involving the whole of one lung by no means predicates adherent pleura of such extent that a useful pneumothorax cannot be obtained. The writer has produced a practically complete pneumothorax in many cases where crepitations extended from apex to base, and also in cases of predominantly basal disease. A clear history of pleurisy (not merely dependent on the patient's statement) certainly gives some likelihood of adherent pleura, but it must be noted that pleurisy by no means always leads to adhesion of the pleural surfaces. Adherent pleura may announce itself by dragging, or interrupted breath sounds, or by a local shortening of inspiration. Particularly suspicious are superficial creaking sounds over a lung extensively EFFECT OF PLEURAL ADHESIONS 61 diseased. Feebleness of breath sounds at the base is of ominious significance — particularly is this the case if marked displacement of organs and evidence of fibrosis are present. In these cases it must be noted that even if a pneumothorax is capable of production, the capacity of a chronically diseased and fibrosed lung to collapse must be somewhat limited. Nevertheless the writer has often in practice obtained quite useful collapse in lungs of this description. Large superficial cavities are nearly always covered by extensive and dense pleural adhesions, and in many, perhaps most, these adhesions are incapable of separation, and the cavity thus incapable of complete extinction under pneumothorax therapy. On the other hand the writer has seen cases where the thickening involved the vis- ceral pleura only, and but few or no adhesions were present to bind the lung to the chest wall. In the writer's experience troublesome adhesions have been found more commonly on the left than on the right side of the chest. The most reliable test of free pleura exists in the tidal movement of the lower edge of the lung during inspiration and expiration (see p. 40) ; if a clean ample excursion is noted it is practically certain that a useful pneumothorax can be achieved. On the other hand, if movement is restricted or abolished, it does not necessarily follow that adhesions are present at the base — such a limitation may be due to adhesions higher up the side or to marked pulmonary changes leading to diminished expansile power of the lung itself. Finally the use of the X-rays may help to decide the presence or absence of adhesions — the finely stippled shadows of pleural thickening may be noted, or a basal pleurisy may lead to distortion and restricted movement of the diaphragm. THE REDUCTION OF ADHERENT PLEURA In cases where no free pleural space can be found, an attempt to create one is recommended by some 62 PNEUMOTHORAX TREATMENT operators. Forlanini's method is to enter his needle just short of the pleural space and then to move it slowly forward, at each movement forcing out some J c.c. of gas by pressure of the fingers on the manometer tubing. If the gas thus pressed out enters the tissue outside the pleura, the manometer shows a negative fluctuation equal to the volume of gas forced in ; if the gas bubble is very near the pleura there may be small oscillations. Where the gas gets between the pleural layers and separates them more marked negative pres- sure and wider respiratory fluctuations occur. When these show themselves an attempt is made to enter more gas under pressure — if the needle is outside the pleura still the rising pressure gradually falls again as the gas diffuses through the tissues. If the needle misses the pleural cavity and enters the lung, small oscillations about the zero point occur. Saugman's plan is to pass the needle into the lung and then cau- tiously withdraw it till oscillations cease. He then uses his suction syringe to exclude a woimded vein, and if no blood comes he forces in Nitrogen under a pressure of 20 to 30 cm. H2O. If the pleural surfaces become separated a negative pressure will occur after some cubic centimetres of gas have entered. More may then be put in if pain allows. These operations cannot be considered entirely free of danger even in the hands of experts, and must generally fail to achieve their object — namely, the pro- duction of an efficient pneumothorax. They should only be attempted, if at all, in cases where pneumo- thorax offers a last chance but a hopeful one, and where the patient understands the immediate risk and elects to face it. If these operations fail or are not attempted, and collapse of the lung is urgently indicated, there re- mains as an alternative the trial of some form of thoracoplastic operation. Of these the one which appears to offer most chance of success is the " Pfeiler- resektion " devised by Wilms. Portions of the posterior EFFECT OF PLEURAL ADHESIONS 63 ends of the upper seven to nine ribs are removed at a first operation, and subsequently, after an interval, portions of the costal cartilages of the first five or six ribs through an anterior incision. By this means the ribs are " mobilised," and losing their support drop downwards and inwards, leading to a very complete collapse of the side of the chest and with it the con- tained lung, especially in its upper parts. CHAPTER VI REFILLS INTERVALS AT EARLY STAGE Having successfully initiated a pneumothorax the question of enlarging and maintaining this needs con- sideration. It has been calculated that the healthy pleura will absorb some 100 c.c. or more of nitrogen daily, though in conditions of disease absorption is somewhat slower. Balboni after putting in 400 c.c. at an initial operation discovered gas still present at the end of a month. In practice it is found that absorption is rapid at first and tends gradually to diminish, so that the refills begun at short intervals will be needed at longer and longer spacing even up to two months or more in some cases. It is convenient to give the first refill the day after the initial operation, or, at latest, on the second day. An extra day or two can then be added to each succeeding interval until col- lapse is as complete as can be obtained ; if marked autoinoculation follows any of these fillings it may be wiser under certain conditions (see p. 89) to postpone the next refill for a day or two. Later the interval will be judged for each case according to the rapidity of absorption and the ability of the lung to remain collapsed under a falling pressure. OPERATION OF REFILLING THE PNEUMOTHORAX This operation differs in but few respects from the initial operation already described. The main points of distinction are as follows : 64 REFILLS 65 (1) No general anodyne is necessary in the case of refills unless in a highly nervous and apprehensive patient for the first one or two. (2) There is no longer need to avoid the use of a sharp-pointed pneumothorax needle, since there now exists a substantial pocket of gas to protect the lung against any possibiUty of injury. Suitable needles for use at refills have been already described on p. 36. They must be sterilised in absolute alcohol and dried by heat in similar manner to that used for the trocar and cannula at the initial operation. Care must be taken that the tap of the needle, whereby it would communicate with the outer air, is closed before the operation is commenced. Puncture Spot. — The spot for puncturing at the first refill is best chosen near the original puncture so as to be well within the gas pocket should the gas be localised by adhesions. In addition it is important at the first refill that the operation spot be directly uppermost so that, in a free pleural cavity, the maximum of gas may collect here and protect the lung from puncture. The position of the gas can generally be pretty accur- ately defined by gentle percussion if any doubt exists as to its location. At later refills, when the pneumo- thorax has bacome complete or at any rate large, the puncture may be made at some other part of the chest if more convenient. A change of position may be dic- tated by progressive thickening of the pleura at the spot where repeated punctures have been made. In puncturing in a new neighbourhood due caution must be directed to the possible presence of stretched adhe- sions, which in some cases may contain lung tissue. These provide the possibilities both of gas embolism and of infection of the pleural cavity if a piece of lung is pierced, accidents which otherwise should be well out of reach in the course of an ordinary refill. Novo- caine should be used, as at the initial operation, since the risk of pleural shock cannot be entirely put aside. 66 PNEUMOTHORAX TREATMENT MANOMETER READINGS Intrapleural Pressure. — The needle is entered slowly and steadily with the eyes closely fixed upon the mano- meter to catch the moment when the needle end enters the pleural cavity; when the pneumothorax is large the needle may be pushed well in so as to support its own weight. The manometer will not necessarily, as at the first operation, register a large negative pres- sure, but the reading will depend on many factors, and in particular, on the final gas pressure at the preceding refill, and the time which has since elapsed. At the first or second refill a considerable negative pressure is still to be expected in the case of a complete or nearly complete pneumothorax. Thus, for example : W. M. 2.2.14 . Initial Operation —16— 8 cm. 325 c.c. — 5 to cm. 3.2.14 . . . . -12-4 „ 700 c.c. - 2+ 5 „ 6.2.14 . . . . - 5+1 „ 950 c.c. + 3 + 10 „ B. J. 9.3.14 . Initial Operation — 12 — 10.3.14 . . . . _ 8 - 12.3.14 . . . . _ 7 _ Where only a partial pneumothorax has been produced the fall of pressure is much less striking. Thus : H. B. 23.3.14 . Initial Operation — 12 — 4 cm. 250 c.c. — 4+ 2cm. ?5.3.14 . . . . - 7to0 „ 450 c.c. + 4+ 9 „ 27.3.14. . . . 0to+2„ 350 c.c. + 9+11,, At later refills the fluctuations are commonly round about the atmospheric line at the time the refill is usually required ; in some cases more below than above, but commonly more above than below, and where with a well- collapsed lung absorption is rendered slow by the presence of thickened pleura, the pressure 8 cm. 300 c.c. — 1 + 1 cm, 5 „ 700 c.c. + 8+ 10 „ 3 „ 600 c.c. +11+13 „ REFILLS 67 may remain positive even at the end of very long intervals. In these results the condition of the lung undoubtedly exercises a decided influence. Where the lung is well compressed, absorption is slow and a posi- tive pressure is readily maintained ; where, on the other hand, the lung is imperfectly collapsed the pres- sures quickly fall to or below the atmospheric, and larger quantities of gas are needed to maintain the pneumothorax. In theory, of course, the latter class of case should be eliminated by the production of a more efficient pressure, but in practice the class exists just because this cannot be done, the usual bar being a discrepancy between the resistance of the lung to compression and of the mediastinum to displacement. A few examples of these different conditions of pressure are here given. H. W. Pressure inclining to negative : 21.7.15 . 13th Filling -4+ 2 cm. 650 c.c. +6+ 10 cm. 3.8.15 . . . -6+ .S „ 825 c.c. +4+11 „ 17.8.15 . . . -5+ 3 „ 650 c.c. +2+11 „ 31.8.15 . . . -7+ 4 „ 740 c.c. +3+11 etc. W. N. — Pressure inclining to positive : 22.6.15 . 9th Filling - 1 + 4 cm. 450 c c + 4 + 7 cm. 29.6.15 . . . - 2 + 6 „ 350 c.c. +1+10 „ 9.7.15 . . . -3+ 5 „ 400 CO. +1+ 9 „ 17.7.15 . . . Oto+ 4 „ 275 c.c. + 2 + 10 etc. M. P. — Pressure wholly positive : 10.11.14. . . +2+ 8 cm. 175 c.c. +16+ 21cm. 22.12.14. . . + 4+ 10 „ 150 c.c. +16+21 „ 5.1.15 . . . +2+ 8 „ 175 c.c. + 16+21 „ 23.3.15 . . . + 1 + 7 „ 200 c.c. + 17+21 etc. This last was a case of partial pneumothorax of old standing with a thickened pleura, and absorption was reduced to a minimum. The most disturbing condition to encounter is one with small fluctuations about the atmospheric — a manometer reading very similar to that which would be 68 PNEUMOTHORAX TREATMENT obtained with the needle thrust into a bronchus or the lung parenchyma. Thus : E. H. B. 17.11.14. . -4+4 250 c.c. +11+17 and in the same patient, a case of partial pneumo- thorax, at a later stage — 29.6.15. . -3+3 175 c.c. + 21 f 25 These cases are disturbing because the possibility of the needle passing into or through adhesions has always to be kept in mind. It is, indeed, a fact that most of the recorded cases of fatal gas embolism have occurred during refills and not at the initial operation. Scrupulous care must therefore be exercised at re- fills, no less than at the first operation, to make certain that the needle is in the pleural space before gas is entered. For this, the wide, easy, respiratory fluc- tuations of the manometer fluid are usually a sufficient guarantee, and this evidence may be safely supple- mented, in the writer's opinion, by the manometric effect of a gentle short cough. Xo deep breath must be taken, and the cough must be gentle indeed or the fluid may be driven out of the manometer tubing. This accident has often to be guarded against during refills by getting the patient to give notice of the need to cough and by firmly compressing the tubing between the needle and apparatus till the coughing is finished. If the characteristic clear swing of the manometer fluid fails at a refill, no gas must be entered until this swing has been estabUshed or re-estabUshed. Its absence often implies a blocking of the needle, but this must not be presumed without proof, and unless the needle can be cleared, it must be withdrawn, re- sterilised and dried, and entered elsewhere. Fortu- nately needles of over 1 mm. calibre, such as the LiUingston or Saugman needles recommended, very REFILLS 69 rarely get blocked if used with proper care, but the small needle below 1 mm. of the type used by Forlanini, Brauns and others readily becomes obstructed or par- tially obstructed, generally by a piece of punched-out skin, so that a clean manometer swing fails to appear. When this occurs the temptation to presume a blocked needle is so great that the writer has himself entirely discarded the use of needles of this calibre as likely to lure the operator into unjustifiable risks. Finally, at a refill, it is always open to the operator to prove the intrapleural location of the needle point in cases of doubt by the withdrawal of some gas from the pleural cavity into the gas bottle. This is achieved by lowering the pressure bottle and must be followed, after the conclusion of the refill, by emptying of the gas bottle and sterilisation of the tubing and filter between this and the needle. The . writer has never yet resorted to this expedient. Before leaving the subject of the manometer at refills, a few additional words may be added as to the significance and interpretation of various mane metric readings. Respiratory Fluctuations. — These in their amplitude are dependent on many factors, such as the calibre of the manometer tubing and needle (p. 29), the type of the patient's breathing, as from nervousness, dyspnoea, mouth or nose breathing, and the like. To obtain a uniform manometer reading the patient should be instructed to breathe through the mouth ; chest breath- ing also supplies a notably larger fluctuation than does abdominal breathing, but the patient cannot with advantage be instructed on this point. In addition it will often be found that respiration naturally runs in cycles about a somewhat different mean pressure according to whether inspiratory or expiratory stimuli are in the ascendancy. The respiratory fluctuations also change in the individual case according to the stage of the pneumothorax and the pressure at which it is kept. Thus when gas is first put in the fluctua- 70 PNEUMOTHORAX TREATMENT tions are those of the normal intrapleural pressure and are large. At a later stage, when the lung is collapsed, the fluctuations are usually but not always found to become smaller, and several factors seem to the writer responsible for this change. First and foremost is the diminished respiratory movement of the expanded hemi- thorax released from the elastic suction of the lung. But in addition, factors dependent on the intra-thoracic pressure come into play. Thus with a pressure near the atmospheric the inspiratory change must be to some extent balanced by a movement of the mediastinum towards the pneumothorax side (see p. 105), while the expiratory phase is reduced by its return towards the side of the functioning lung. At higher pressures this mediastinal movement ceases, but then the diaphragm, on whose action the respiratory change of pressure in large part depends, ceases any longer to functionate (see p. 107). Certain points of practical information are said to be deducible from various types of manometer reading, thus : Be-expansion of Lung. — If the lung tends to expand in the interval between the fillings there will be a gradual rise of pressure at the refill, whereas if the lung remains collapsed the pressure is said to rise abruptly at the end of the refill (Hamman and Sloan). This has not been the writer's experience. Condition of Mediastiniun. — ^An elastic flexible media- stinum is shown by a slow rise of pressure, often with wide respiratory fluctuations, whereas a rigid media- stinum causes the pressure to rise rapidly at the end of the refill and the respiratory fluctuations are small (Hamman and Sloan). These points, it will be seen, are measures of the elasticity of the walls of the gas chamber. The pre- sence of a fluid effusion reduces this, perhaps, to its lowest point, and in this condition the addition or sub- traction of a few centimetres of gas may make startling differences in the intrapleural pressure (see p. 137). REFILLS 71 PUTTING IN GAS Having successfully entered the needle the mano- meter reading is recorded, and connection with the gas bottle established, the level of the fluid being so adjusted that a pressure 5 or 6 cm. above the intra- pleural, that is, in most cases a decided positive pressure (above atmospheric) is supplied. The column of fluid in the gas bottle and the patient are carefully watched, as in the first operation, and as soon as some gas has gone in, the pressure bottle may be raised so as to establish a more decided pressure. Gas is passed in, not, as at the first operation, up to a definite quantity ; the criterion now and henceforth is one of intrapleural pressure, not quantity of gas. EARLY FILLINGS. QUANTITY AND PRESSURE OF GAS In cases where the lung collapses readily, it is wisest to keep below atmospheric pressure till about the third filling of gas (see Case of W. N., p. 57). After this the pressure is gradually raised till collapse is complete. Com- plete collapse in such cases is generally obtained, if this method is followed, by the fourth or fifth filling, and a sufficient pressure must then be maintained to prevent further expansion. By following this procedure it will be found that some 500-800 c.c. are put in at each filling after the first, perhaps 500 c.c. at the second filling, and 500-800 at the third, fourth, and fifth fillings. It has been calculated that the adult pleural cavity holds some 2,000-3,000 c.c. of gas, but an un- known allowance must be made in each case for re- absorption during the intervals between the refills. COMPLETE COLLAPSE What evidence, it will be asked, is needed to decide when complete collapse of the lung has been obtained ? The matter can be to a large extent decided by physical signs and symptoms, but the great value of the X-rays in this decision is undeniable. 72 PNEUMOTHORAX TREATMENT Radioscopy. — As the volume of gas put in becomes larger the lung is seen to retreat towards the media- stinum, its shadow increasing in density as its size diminishes. Eventually, where a complete collapse is possible, it comes to lie close against the mediastinum in a large gas cavity whose dimensions overstep those of the normal pleural cavity in every direction. For the pleural cupola becomes lifted above the normal lung apex, the mediastinum is bowed over to the healthy side, and the diaphragm assumes a less convex, flattened, sunken, or even in rare cases an altogether inverted position. Seldom, however, is the lung en- tirely unadherent ; more commonly adhesions at apex or base or at points between cause the lung to assume an irregular outline, and to give a shadow of various thickness, size, shape, and localisation on compression. The more effective the compression the deeper and smaller the shadow — only in hepatised lung is a deep shadowing associated with a large but immobilised organ. The test of effective collapse is the failure of the contracted lung to expand with inspiration — mere movements communicated to the lung by respiratory fluctuations of the mediastinum must not be mistaken for changes of volume. In the production of complete collapse stages may be passed through where one lobe of a lung is immobilised but the other still expansile — obviously the elastic, healthy lung tissue is more readily compressed than inelastic diseased parts. On the other hand, once full collapse has been attained, the inelastic diseased part is, as a rule, less easily re- expanded, and a lowering of pressure, whether tem- porary or continuous, may in some cases be permissible. Saugman describes a case where the lower lobe, the seat of advanced disease, maintained its collapse while the healthy upper lobe expanded by three or four times its volume on deep inspiration. At a higher pressure the whole lung became immobilised. Radioscopy should be applied soon after a filling, and at intervals so that the period during which the lung will remain collapsed, REFILLS 73 and the pressure at which it begins to re-expand, may be discovered and thus the intervals best suited for refills decided upon. This may be revised from time to time as the lung, and particularly its most diseased portions, becomes less liable to re-expansion. Physical Signs. — ^A general impression of collapse of the lung is obtained where a large or complete pneumo- thorax has been formed, and the lung has disappeared from view, save, perhaps, to careful percussion in the interscapular region. AH moist sounds should have vanished, but breath sounds do not entirely disappear owing to conduction of these from the air tubes or the opposite lung. At the same time movement of the pneumothorax side is diminished, and its girth some- what increased with a widening of the intercostal spaces. When this condition is reached at a pressure somewhat above the atmospheric (say -J- 5 mean pres- sure) there is a presumption that the lung is completely collapsed, but physical signs alone cannot prove the point, and further evidence must be sought from the symptomatology. Symptoms. — The process of collapsing the lung is accompanied by a squeezing out of secretions — this leads to an increase of cough and sputum after each refill, and when the collapse is complete the sputum disappears or sinks to an even supply. This, in absence of radioscopy, is the best test of the production of collapse, and, if the intervals between the refills are kept short, is a fairly good one. If it is doubtful whether at a given pressure (say 5 cm. H2O mean) the lung is fully collapsed, a cautious rise to a some- what higher pressure (say 7 to 10 cm. H«0 mean) a day or two later may decide the matter through the effect or absence of effect of this on cough, sputum, and temperature. In gauging the best intervals between refills the same rough test may be apphed. If the interval is so long (and the pressure fallen so low) that the lung re-expands to any marked extent, the subse- quent refill will be followed by an increase of cough 74 PNEUMOTHORAX TREATMENT and sputum, and possibly by symptoms (fever, head- ache, etc.) pointing to an auto inoculation. If a still longer interval is left between refills some of the symp- toms present before treatment was begun may reappear and gross evidence of re-expansion in fever, cough, wasting, etc., be present. Such a disorganisation of treatment must, of course, never be allowed to occur, and this brings us in more detail to the question of the intervals between the refills. INTERVALS It has already been pointed out hoA\', as a routine measure, it is convenient gradually to increase the intervals by a day more each time at the beginning. But when full collapse of the lung has been attained, more rational means for deciding the suitable interval should be adopted. Our purpose is to prevent any re-expansion of the lung, and gas should be put in at such intervals as to prevent this. Signs of re-expan- sion may be gauged by the means decided above ; in addition the rapidity of absorption varies notably with the duration of treatment. Whereas at first the pleural cavity may be expected to absorb some 80—100 c.c. of gas per diem, after the pneumothorax has existed for some months the daily absorption falls to 25-50 c.c. Patients vary somewhat also in the quantity of gas they will tolerate at a refill {i.e. the amount they can lose between) without interference with the favourable course of the cure. Forlanini is especially insistent that immobilisation once achieved should thereafter be kept complete. He gives 100—200 c.c. daily until full immobilisation ; later every other day, and gradually at longer intervals till after two to four months this gradually suffices to maintain the pressure when given at six to nine days' interval. Even at later stages he keeps the intervals so short that 200-300 c.c. suffices, though in some cases he increases this to 300-500 c.c. These frequent injections and consequent small refills are a REFILLS 75 council of perfection which the endurance of the patient and the leisure of the operator may not always permit. At the same time there is no doubt that this is the path of safety, that short intervals and small injections should be aimed at, and that in cases where patients do not improve as they ought with a somewhat coarser procedure, this method should at once be resorted to. It is a very common mistake, in the writer's experience, to leave intervals so long that injections of 800 or 1,000 c.c. are needed to restore the pressure, and this must lead not only to general intrathoracic disturbance from marked changes of pressure, but also in nearly all cases to a notable expansion of the collapsed lung. Where large quantities are needed at short or moderate intervals, this is probably due to an incomplete col- lapse of the lung, and is an indication for a somewhat higher pressure. 500 or 600 c.c. should probably be the outside quantity of gas at refills, and if this is exceeded the intervals must be correspondingly shortened. Saugman mentions cases where even refills of 400-500 c.c. were too large and led to loss of weight, palpitation, dyspnoea, general illness and rapid pulse after the injections, and where improvement immediately fol- lowed a change to 200-300 c.c. at shorter intervals. This intolerance, he thinks, is especially likely to occur where the other lung is of doubtful soundness, or where the digestive organs suffer disturbance owing to de- scent of the diaphragm. Some patients experience dis- comfort or even distress for mechanical reasons, if a long interval is left between injections, especially, in the writer's experience, in cases of long standing and with thickened pleura. One of Saugman's cases pre- ferred to continue the treatment indefinitely on account of the discomfort of leaving off. The rapidity of absorption of gas from the pleural cavity diminishes with time as already noted ; it also varies apparently with the amount of lung disease, and certainly with the activity of the patient. It has been commonly stated that the more advanced the. disease 76 PNEUMOTHORAX TREATMENT the greater the absorption — possibly there is a certain amount of dyspnoea in these cases. The activity or quiescence of the patient exercises a marked influence on the rapidity of absorption of the gas, particularly, perhaps, in cases of partial pneumothorax. In a patient at work, or otherwise " at large," it is often possible to gauge the amount of his activity by the experiences of the next refill. This supplies an indica- tion for restriction in cases where absorption is incon- veniently rapid. But the influence which bears most closely on the question of gas absorption is the presence of pleurisy, whether past or present. At the time of a pleural attack absorption is often nearly in abey- ance — indeed the intrapleural pressure may rise apart from the appearance of demonstrable fluid (see Pleurisy). The pleural changes also which follow an attack of inflammation generally lead to a permanent fall of absorption so that in some cases refills may be left to intervals of two months or even longer. One of the writer's patients who sustained a pleurisy some years ago and is left with a thickened pleura loses now only some 20 to 25 c.c. of N per week. Thus : 5.1.15 . . +2+8 175 CO. +16+21 23.3.15 . . +1+7 200 c.c. +17+21 a loss on this occasion of only 200 c.c. in eleven weeks. As a rule the intervals, gauged by the indications given above, lengthen themselves into a wctk after the first one or two months, reach two weeks after three or four months, and thereafter expand themselves till three or four weeks can be left between. This is the average interval at which refills need to be done after cases are well advanced in treatment — only after pleuritic attacks can the intervals be carried further than this as a rule. Intervals in Partial Pneumothorax. — Here, on ac- count of the smaller size of the pneumothorax and the effect of the still mobile lung on absorption, fillings REFILLS 77 will generally require to be more frequent than in cases of complete pneumothorax. For a long time fillings at one or two weeks' interval may be needed to main- tain the maximvim collapse, but after long periods, and particularly if pleurisy supervenes, these intervals rnay often be lengthened to three or four weeks without disadvantage. *to"- INTRAPLEURAL PRESSURES Complete Pneumothorax. — In a complete or nearly complete pneumothorax a quite small positive pres- sure, -f 3 to + 5 cm. H3O, or at most + 10 cm. mean ■pressure, is generally sufficient to maintain complete collapse. Saugman recommends pressures of + 5 + 10 + 15 cm. according to the individual needs, but considers a mean pressure of + 10 + 15 very high for a complete pneumothorax with normal pleura. He advises that during the first four to six fillings the pres- sure be kept low, from + i to -f 2 or 4 cm. mean" pres- sure, and this in some cases will suffice to maintain collapse. But in many cases a higher pressure will be required for this purpose, and very different pres- sures are needed in different patients even where the pathological conditions appear to be closely similar. The difference appears to depend, even in cases of complete pneumothorax, on conditions of pleural thickening, and it fortunately happens that the need for a high pressure is generally accompanied by the presence of a resistant mediastinum. It cannot be too strongly insisted that the optimum pressure is the lowest pressure which will suffice to maintain the lung fully collapsed. There is a curious temptation, which must be recognised and resisted, to raise the intra- pleural pressure in artificial pneumothorax above the strictly needful level. This depends largely on the increased tolerance of the patient which will permit a gradual rise to pressures which are needless, undesirable, or even dangerous. For it must be recognised that a 78 PNEUMOTHORAX TREATMENT pressure which causes no immediate discomfort to the patient may yet be a source of ultimate danger by displacement and embarrassment of other organs, and particularly of the opposite lung. The danger of attaining a too high pressure exists particularly, the writer thinks, during the first few months of treatment when the pressure-optimum is being sought for — it must always be remembered that it is an optimum and not a maximimi that we are out to find, and that at the lowest pressure possible. When this has been discovered, it should be strictly adhered to at subse- quent refills. It has been demonstrated that high positive pressures as a rule fall rapidly during the first few hours after a filling ; nevertheless the rise to a too high pressure even on a single occasion may, the writer believes, cause a permanent stretching and displace- ment of the mediastinum, which ^ill remain an em- barrassment to the opposite lung even after a return to lower pressures. The effect of such displacements on the mediastintun, diaphragra, and neighbouring organs will be more fuUy discussed under another heading (see Chap. X.). Partial Pneumothorax. — In cases of partial pnemno- thorax much higher pressures can be tolerated and are needed than those suitable for complete pneumo- thorax. Here the indication is to enlarge the space and to render the pneumothorax complete, or at least efficient. To achieve this result refills should be given at short intervals, and the pressure gradually raised with due attention to the effects on sm-rounding struc- tiu-es, particularly the mediastinum. In a small gas pocket decided positive pressures may be reached without harm at the first refill, and these may be sub- sequently increased. Often a gradual stretching or separation of adhesions shows itself in a gradual fall of pressure and the occurrence of marked plem^al pain. This may be situated at the base of the chest, or, if diaphragmatic, be referred to the shoulder through the reflex association of the phrenic with the fourth or fifth REFILLS 79 cervical nerves (Mackenzie). Occasionally a rapid stripping of adherent pleural surfaces announces itself by a sudden fall of pressure accompanied by pleural pain during the course of a filling. E. W., s, patient with advanced one-sided disease and marked basal cavitation. Pnevunothorax initiated by puncture in second space, and 400 c.c. of gas introduced. Refill three days later, when it was found that the breath sounds, before weakened, had returned over the region of puncture. The needle was entered close to the first puncture, a manometer reading of — 12 — 7 obtained, and the gas bottle opened. But gas did not flow in, and when pressure was raised the intrapleural pressure rose correspondingly. It was concluded that a small pocket had been entered. The needle was withdrawn and reinserted in the track of the initial puncture of three days earlier. Here a reading of — 5 — 1 was found, and on raising the pressure again no gas entered. It was then concluded that the pleural surfaces were in contact and adhesive, and the pressure raised till + 14 was reached. At this point the patient remarked on a " prick- ing " sensation, not amounting to pain, experienced in the chest at the site of operation. At the same time gas began to flow in readily, while in spite of this, the pressure steadily fell till after 600 c.c. had entered it stood at + 1 + 6. Pressures of 20 cm. H2O or thereabouts will often suffice to achieve the desired enlargement of the pneumothorax cavity (see case on p. 55), but with a resistant mediastinum and diaphragm mean pressures of 30 to 40 cm. may be used in some cases, and thereby strong adhesions gradually stretched or separated. Such pressures must not, of course, be reached where they lead to marked displacement of mediastinum or diaphragm, or ballooning of pleura (pp. 104 to 108), but often these parts are stiffened by pleural thickening and offer a good resistance to pressure. In cases where the pleural pocket is in close proximity to the heart high pressures may lead to dyspnoea, pain, cyanosis, and a small, slow pulse on occasion (Brauer, Weiss). The following case of the writer's illustrates this condition, though no distress was observable save some pallor and a sensation of " tightness " at the first filling : 80 PNEUMOTHORAX TREATMENT A. A. (?., aged 39, with a 3J- years' history, showed moist sounds over the whole right side with a cavity at the apex, but left side practically clear. On 31.3.14 three attempts to find a free pleural space failed, but on 20.4.14 at the second puncture, free pleura was found in the nipple line, and a small gas pocket produced between this point and the heart. After 150 c.c. of gas had entered marked cardiac pulsations of 3 or 4 cm. amplitude appeared, and obscured the respiratory fluctuation. 200 c.c. were put in and brought the pressure to + 8 + 12, when the patient became somewhat pale and complained of a sensation of " tightness." The pulse remained 76, the heart was displaced IJ inches. On three subsequent occasions gas was put in in quantities between 110 and 300 c.c, and the pressure raised to + 38+ 41 with marked cardiac pulsations without any further discomfort; the gas pocket, however, failed to extend and the attempt was aban- doned. It cannot be denied that a certain amount of danger exists that under high pressures lung-containing adhe- sions may be torn and embolism result ; Saugman also records a case where the patient died after a refill at a pressure of -f 40 cm. with symptoms suggestive of heart failure. The desirability of running such risks — and they may be justified by the patient's otherwise hope- less condition — must be duly considered for each in- dividual case. Sometimes it becomes quickly apparent with a small unyielding pocket that even the highest pressures will achieve no useful result. This has generally been so among the writer's cases where no more than 300-350 c.c. can be put in at each filling (see cases on pp. 53 and 54). It is remarkable, on the other hand, how greatly in certain cases a small pneumothorax can be enlarged with patience even at such pressures as 20 cm. or thereabouts (see case on p. 55). Effect of Altitude on Gas Pressure. — Since tuberculous patients are particularly partial, with some reason, towards treatment in the high Alps, it is needful to remember the effect of altitude on the gas in a pneumo- thorax. It has been calculated that a rise of 1,000- 1,500 metres (about 3,250-5,000 feet) is equivalent, in a full pneumothorax of 3 or 4 litres, to adding 400- REFILLS 81 800 c.c. of gas. This may suffice to cause a dangerous or even fatal intrapleural pressure, and pneumothorax patients must be warned of the dangers of such changes of level. Before going to the mountains their pressure must be suitably adjusted and the patient warned to respect dyspnoea as an indication to break the journey, and abstain for a time from higher elevations. CHAPTER Vil PHYSICAL SJONS IN ARTIFICIAL PNEUMO- THORAX 1'iiE physical signs over an artificial pneumothorax often fall far short of that elassic^al picture drawn for the student from cases of spontuneous pneumothorax. After the first filling with sojne 300 e.c. of gas an area of res(;)iance and weak breath sounds may be dis- covered to careful examination, and, in addition, an erieroachment on the eoinjdemental pleural spaces, especially the incisura cardiaca on the left side. These signs increase at subsequent refills, rales and crepita- tions diminish and disappear, and when the pneumo- thorax is complete the following signs are present. The affected side increases somewhat in size, and the intercostal spaces widen even when before treatment these were narrowed. An increase of size of I to 1 inch above that of the opposite hemi- thorax rnay be indicated to the tape-measure. Forlanini deserjbcs the appear- ance of a network of veins over the thorax and upper limb on the affectfd side, but the writer has not observfd this among his eases. The percussion note varies according to the pressure ; if this is low, the side of the chest is hypcrrcsonant, but wJien the pressure is raised the note becomes high- pitched and even tympanitic. The percussion note over the healthy lung also varies according to the amount of irudiustinal dis])laccment. Where this is not much a normal resonant note is maintained, but where a weak mediastinum is carried far over and the healthy lung 82 PHYSICAL SIGNS IN PNEUMOTHORAX 88 relaxed, the note over it will be high-pitched or even, a somewhat ominous sign, tympanitic. It will thus be seen that the comparative resonance of the two sides varies according to the conditions of each, and is often different before and after a refill, and also somewhat for front and back of the chest. The outline of the pneumothorax can be obtained with fair accuracy by light percussion — auscultatory percussion advocated by Morelli and also by Saugman has not proved of much assistance in the writer's hands. The breath sounds over a pneumothorax are rarely or never wholly lost, even when the lung is completely collapsed, being conducted in this case from the air tubes or opposite lung. Over the good lung the breath sounds commonly become harsh or puerile, but in some cases they are but little altered from the normal. The distant breath sounds over the pneumothorax side may be of normal quality, but after a time they tend to become amphoric. This change may appear as early as the second filling, but is often delayed for a couple of months or longer. The metalhc clang or bell sound, also, is a late sign in these cases, is seldom present in the degree common in spontaneous pneumothorax, and is not infrequently absent altogether. It is always of later appearance than the metallic breath sounds, and is found at first only after refills. According to Saug- man it is earliest seen in wasted cases with considerable lung infiltration. A metallic echo of the voice sounds, metallic whisper, and metallic rales or tinkling may also be present. Rales and crepitations over the diseased lung disappear, as would be expected, when this lung becomes collapsed. Where the pneumothorax is incomplete they may be retained in greater or less degree. It is a remarkable fact, also, that the production of a pneumothorax will often abolish crepitations from the opposite lung, notably from the apex and from the mid-dorsal region, suggesting that some of these signs were in all probability con- ducted from the diseased side. The conduction of lung 84 PNEUMOTHORAX TREATMENT signs, both rales, particularly if consonating, and also bronchial breathing, is a well-known phenomenon in children's chests, but has perhaps received too little recognition in the case of adults. The writer thinks that this explanation applies more to signs over the mid- dorsal region, even out to the axilla in some cases, than to the apex. The disappearance of signs in this latter region he is more inclined to ascribe to the effects of relaxation of the lung due to mediastinal displacement, since they often bear no relationship to signs at the opposite apex. In cases where considerable mediastinal displacement has occurred, crepitations may appear from time to time at the base of the opposite lung, and must be regarded as a danger signal. Such signs are probably, in some eases, atelectatic, but in others almost certainly due to oedema, and further reference is made to them imder another heading (p. 112). Vocal resonance is often remarkably little affected by the production of an artificial pneumothorax ; it niay remain as strong or nearly so as on the healthy side, but it is conimonly somewhat diminished, may be nasal, and, in some cases, as already remarked, is metallic or accom- panied by a metallic echo, The vocal vibration, also, is often quite strong on the pneumothorax side, though a rise of intrathoracic pressure may diminish it. If fluid appears the vocal vibration is lost over this, but quite good over the gas above it. In artificial pneumothorax the collapsed lung can often be mapped out by careful percussion in the inter- scapular region between the first and the seventh rib as it lies along the spine, and it might be supposed that the physical signs, particularly as regards conduction of breath or voice sounds from air tubes or opposite lung, might be different in this part of the chest to elsewhere. This, however, is not the case, and no blowing breath sounds, bronchophony or other signs of condensed lung tissue are present as a rule. Where, however, the lung apex is adherent, harsh or blowing breathing, and per- haps crepitations may be audible here and in the inter- PHYSICAL SIGNS IN PNEUMOTHORAX 85 scapular region, and sometimes also across the spine in cases where mediastinal displacement is a notable feature. These signs may only appear with that pul- monary exacerbation or congestion which so often Fig. 5. — Showing thk Paeavebtebbal Tbiangles, a^td, on the SIDE OPPOSITE THE PnEUMOTHOKAX, A FURTHER QUADRANGULAR Area of Impairment hitherto undesoribed. accompanies the onset of pleurisy, and may be absent or less striking during the remainder of the treatment. It is of some interest that the basal paravertebral triangle of percussion impairment, associated in disease with the name of Grocco, but also present in the healthy, is still to be found over the pneumothorax side, showing that, in normal chests at least, its explanation must be 86 PNEUMOTHORAX TREATMENT found in thoracic muscle and not in pulmonary con- gestion, as some would have us believe. In addition there is present, on the healthy side, a quadrilateral area of impairment not hitherto described, and not easy of explanation (fig. 5). This has its upper limit at the eighth dorsal spine, the common level of the paravertebral triangle in the healthy, and it overlaps this latter by an inch or two at the base. Its width increases after a refill suggesting some relationship to mediastinal displacement. In cases of pneumothorax with effusion this quadrilateral area of subresonance is greatly increased in size, and reaches up to the fluid level in similar manner to Grocco's triangle in a simple pleural effusion. CHAPTER VIII GENERAL REGIME : THE COURSE OF TREAT- MENT AND ITS TERMINATION During the early course of pneumothorax treatment rest in bed is required — this must be maintained at any rate till the pneumothorax is complete (ten days or so), but in many cases is best continued longer. Where a patient's general condition is good, and fever has quite gone, a fortnight's bed may suffice, but even here the displacement of organs makes it wise to proceed slowly from bed to armchair, and so gradually onward. Where prolonged fever and illness have undermined the patient's health a month or more in bed may be needed, and sub- sequent progress must be even slower and most carefully graduated. A time soon comes, however, in a favourable case, when the patient can be up and about, and then appears one of the most gratifying of the immediate changes achieved by pneumothorax therapy. For whereas before treatment the patient's movements were severely limited by the advent or increase of fever and symptoms on the slightest excess, he can now proceed to lead an ordinary life without any signs of autoinocula- tion. This enables him quickly to regain his lost physical and mental health, and to bring his muscular system once more into condition. Briefly, he achieves in a few weeks or months the same general improvement which urider a spontaneous cure could only be accomplished after years. The whole effect of this is startling and most beneficent, providing the patient is prevented from carrying his energies too far at first. Finally he 87 88 PNEUMOTHORAX TREATMENT can return to a fairly ordinary existence, and will often undertake his original occupation with impunity. The more immediate results of pneumothorax therapy thus divide themselves conveniently into three stages, which may be separately considered under the headings of 106- lOB- I04' i'.'.'-'X..' i .;.:;, ::::tr rtl: ;'; ;j " ;;; ~T" /' , ^ '6+1 r : 7— : 1 . . '7.. }M ^.::,.- ■ . |l02' 1 <»: NORMAL 98° 96- ::::.".:.". .; -ij ^ — ■ - / - A ^A - . . , :,-:: ::::■:::::: ....!::, :p:\ \/ ' A ;A h :::::;:ir .::::i:::::: zz.:z. /' s/ V \/ / I . L J" yi / -: '■■■■■ i: V \l ^ t \. often experienced a slowing, in one case from 80-88 down to 50-60 due, he thinks, to vagus irritation. The fall of the pulse resulting from the removal of toxaemia is liable to confuse or to counteract any changes depending on circulatory disturbances. In addition to displacement the heart may be called upon to compensate for a reduction of passage-way through the pulmonary channels, a matter on which some diversity of evidence and opinion at present exists. Bruns in numerous experiments on dogs, goats, and rabbits found hypertrophy of the right heart always followed pneumothorax of three months' standing. Postmortem experience in man gives somewhat con- flicting evidence. Right-sided hypertrophy, or dilata- tion, or both have been found by Carlstrom, Waller, 118 PNEUMOTHORAX TREATMENT Hymans and a few others, but five autopsies by Brauer and Spengler, tiiree by Graetz, and others by Saugman and Hansen and by Kistler revealed no cardiac changes. These discrepancies probably permit a similar explana- tion to that already offered for the various views of pul- monary blood content advanced by different authorities (pp. 114to 116). It may well be that in some cases the total pulmonary circulation is in no way embarrassed, while in others a notable reduction of the pulmonary blood channels may be present, and place an increased strain on the right heart, all according to the degree of pressure on the collapsed lung, and the degree of expansion, determined by mediastinal displacement, of the function- ing organ. This view seems to receive support from discrepant blood-pressure observations obtained by different observers — Bursten finds a rise of general blood pressure for a short time after a refill, whereas Guichard records a fall of systolic blood pressure of about 20 mm. Hg. with a subsequent rise to normal after twent>'-four hours. Kessel and Taschman found no constant blood- pressure changes before, during, or after the gas fillings. Be it noted, however, that the total circulation may be reduced, and yet the functioning Ivmg be decidedly hypersemic, so that there is nothing contradictory in the discovery by Brvms of hyperffimic lung, but at the same time obstructed circulation, as a result of his experiments on animals. So far as cardiac enlargement is concerned the writer has long been struck with the ^\idth of the deep cardiac dulness in his pneumothorax cases, the transverse measurement being often 5-| inches in place of the 4| commonly found in health. This he has been inclined to attribute to a dislocation of the heart forwards against the chest wall rather than to true enlargement, since a forward displacement of the heart was noted by Salle in cases of pleural effusion, and the same might well be'expected in pneumothorax. The matter at present remains in some doubt, and further work is needed for its satisfactory elucidation. One of the writer's cases EFFECTS ON THE THORACIC^ ORGANS 119 in whom a good deal of emphysema was present, and in whom the cardiac dulness measured some 7 inches across throughout treatment, died rather suddenly of heart failure after pneumothorax had been kept up for more than a year with otherwise satisfactory results. CHAPTER XI PLEURISY AS A COMPLICATION Pleurisy is so common and so important a complication of artificial pneumothorax that no apology is offered for the amount of space accorded to its discussion in this small volume. PREVALENCE It is to be expected in some 50% of cases followed through from start to finish. It is not always an early complication, and it may thus be missed in cases treated for a short time or drifting into other hands. Among ninety-nine effusions recorded by Saugman (in ninety-six patients of whom three were treated on both sides at different times) the fluid appeared at the following intervals of time : In 21% within the first 3 months. M 33% „ ,, ,, 6 „ 41% Q ,,44% „ 1 year. ,,50% „ 3| years. It is thus seen that the chance of effusion increases greatly with the length of treatment, von Muralt was also able to show that the character of the case greatly influenced the occurrence of effusion, this occurring more frequently and also earlier among advanced and acute cases. Weinstein of Davos experienced effusions among 70% of advanced cases, among 33% of moderately 120 PLEURISY AS A COMPLICATION 121 advanced, and only 10% of early and slight cases. These factors, no doubt, are largely responsible for the discrepancy between the figures of various observers. Thus the writer experienced no case of effusion within the first eight months of commencing to use this form of treatment — subsequent events have caused him to abandon the hope that this immunity bore any relation- ship to his methods of operating. Brauer and Spengler had 44% of cases within fifteen months, von Muralt 50 % of fresh effusions among 64 cases of long duration, Begtrup-Hansen 50% among 66 cases, Ziegler of Heide- haus about 50% among 60 cases, Weinstein 33% among 71 cases, Tecon and Sillig but 29 "63% among 108 cases, Brauns of Hanover mentions but 5 effusions among 98 pneumothorax patients, 80 of whom were cases of phthisis. Mayer punctured and studied 18 exudates among 46 cases, and saw also 4 more too small and fleeting for inclusion among his cases, a total of nearly 48%. It must be admitted that small and fleeting effusions may sometimes be missed, and that particu- larly where X-ray examination is not to hand ; this perhaps may furnish additional explanation of personal discrepancies. In partial pneumothorax pleurisy is much less prone to develop than in those which are complete or nearly so. CAUSATION Apart from a few cases of infection with pyogenic organisms, the effusions of artificial pneumothorax are sterile to culture and, for the most part, are found to contain tubercle bacilli either to microscopic examina- tion or to animal inoculation. Saugman was able to demonstrate the tubercle bacillus in all his large and chronic exudates with only two exceptions. When the mechanical conditions present in an artificial pneumo- thorax are borne in mind, it is hardly surprising that tuberculous pleurisy should be of common occurrence. For whereas with an expanded lung the activities of 122 PNEUMOTHORAX TREATMENT surface foci are readily limited by the formation of adhesions, when the diseased lung is collapsed and the pleural layers separated, any small focus of tubercle at the surface, its barriers removed, may run riot and lead to the production of a serous or sero-purulent effusion. This explanation probably covers the main part of the ground, but other influences are at work also which modify and thus tend to specialise the exudates arising in these cases. Thus the effect of separating these normally opposed moist surfaces must be borne in mind, and, in addition, the influence on them of the constant presence of nitrogen gas, an influence which Kaufmann found to lead to histological changes of the serosa in experiments on dogs. The changed gas absorption from the pleural surface during the course of artificial pneumothorax cases is evidence of this influence, and both Mayer and von JMuralt tend to regard it as of some importance in the occurrence of pleurisy. The irritating effect of these influences on the pleural sur- faces, even after a minimal interval, is well illustrated by the ease of E.W. described on p. 79, though it must be admitted that the action of the local anaesthetic, novocaine, cannot be excluded. Here, unadherent sur- faces were rendered adhesive after less than three days of separation by a layer of nitrogen gas, and on coming together needed a pressure of + 14 cm. HoO to strip them ajDart again. In addition it has been observed by Forlanini, and von Muralt is in agreement also, that often a severe exposure or chilling, or some over-exertion, has preceded the onset of these pleurisies, and has constituted at least an important predisposing cause. Indeed, von Muralt even speaks of pleurisy as an epidemic occurrence among these patients during certain cold months of the year, particularly I\Iarch and April. The writer has found a certain connection between catarrh of the upper air passages and the onset or exacerbation of pleurisy among his cases. He has also been much struck with the relationship between the more severe of these pleural attacks and evidence of PLEURISY AS A COMPLICATION 123 congestion or of renewed activity in the pulmonary tuberculous foci, and this has been remarked on by Mayer. Pleurisy may also arise from the breaking down of adhesions, and possibly from puncture of adhesions containing a core of lung tissue. Infection of the pleura from without should never occur under proper aseptic precautions ; even the carrying in by the needle of a punched-out portion of skin seldom or never seems responsible for pleurisy — nearly always the piece of skin remains within the lumen of the needle. DRY PLEURISY The occurrence of dry pleurisy is an undoubted fact ; it may or may not lead on to effusion. It may be ushered in by pain in the side, vomiting and fever, as in a case of the writer's ; areas of impairment may be found and of shadowing to X-rays, but yet no fluid follow. Where, in these cases, the pneumothorax is only partial, pleural friction may appear and confirm the diagnosis ; or friction may be heard at the diaphragm- atic reflection in complete pneumothorax. Among forty cases of pleurisy recorded by L. Spengler, twenty- six remained dry. In some cases pleurisy reveals itself by the discovery of high pressure on puncture without any fluid appearing, a condition regarded by Zink as due to pleural irritation without exudation (see Emerson's observations on p. 137). It is the writer's experience that a local pleurisy with thickening may occur if puncture is repeated too often at the same spot. PLEURISY WITH EFFUSION Pleural effusions in pneumothorax are conveniently divisible into three main classes : CLASS I Mild serous effusions which appear slowly, have little tendency to increase, and mostly disappear after a few 124 PNEUMOTHORAX TREATMENT weeks or at most months. Among thirty-five exudates described by von Muralt fourteen, or 40%, belonged to this category. CLASS II Effusions which begin like Class I, but show a tendency to increase slowly with but few symptoms till a large quantity is present. Of von Muralt's 35 cases, 16, or 45 "5%, belonged to this group. The fluid may last for years unabsorbed, and if it is removed it generally returns again. In a case of the writer's the pleural cavity had to be emptied four times before the formation of fluid began to abate. At the beginning the effusion is serous and lymphocytic, and tubercle bacilli are difficult to demonstrate. It may remain serous as in the case quoted above, or it may become purulent, either quickly within a few days, or slowly after many months. For its further course under these circum- stances, the reader is referred to the heading Purulent Effusions on the following page, and especially to Jacot's Groups 1 and 2. CLASS III Pleurisies of acute course with high fever and general illness. Only 5, or 14'3%, of von Muralt's 35 cases were of this nature. Such pleurisies may follow infection elsewhere such as tonsillitis or " influenza," and the exudate tends to be richly cellular and may contain pyogenic organisms. Spengler observed five such cases in wliich diplococci were present in the exudate, but which, nevertheless, did just as well as cases of sterile effusion. Jacot's Group 3 of Purulent Effusions (p. 127) would also come under this heading, as would, indeed, direct infections of the pleural sac through perforation of the lung (p. 162), or by wound- ing of the lung with the pneumothorax needle, or from the outside during puncture, an accident which must be, in properly conducted operations, extremely rare. Many writers, without remarking on their acute or PLEURISY AS A COMPLICATION 125 chronic nature, divide their effusions into serous and purulent. Thus Begtrup-Hansen remarks that of his 33 cases, 10 were purulent. Tecon and Sillig among 32 cases record 81 "15% serous, 6 "5% serous becoming purulent, and 12 "4% purulent from the beginning. Since tuberculous effusions may be purulent or become so, this division gives no clear indication either of etiology or of the acute or chronic nature of the case ; it only presents the matter from another point of view. Very useful and interesting, however, is Jacot's classification of purulent effusions which supplies an almost in- dispensable supplement to the preliminary division into three main groups. Purulent EfEusions. — Jacot was able to collect twenty- five cases for classification, and these he found to fall conveniently into four Groups : (1) Pyopneumothorax of benign nature, always tuber- culous in origin, beginning as a serous pleurisy (Class II, above) in a pneumothorax patient of good general con- dition. After the initial fever the temperature falls to normal and remains there, the patient keeps well, but the fluid gradually passes through a lymphocytic stage to turbidity, and finally after three or more months into frank pus. There is no tendency for this to absorb and generally it needs removal at intervals which fortunately become longer and longer. In a case of this nature under the writer's care, pus formation has only begun to slacken after over a year's treatment, during which time no less than 6 1 litres altogether, or nearly 1-| gallons of pus, swarming with tubercle bacilli have been removed and replaced by gas. The patient meanwhile has felt well, led a fairly active fife, and steadily gains in weight. Katz mentions two such cases clearing up after repeated tappings. Nine of Jacot's cases, or 36%, were of this benign character. Their treatment should be by fairly frequent tappings and replacement by air or gas at pressures sufficient to prevent re-expansion of the lung. It may happen occasionally that the fluid becomes thinner in its later stages, but commonly there is a 126 PNEUMOTHORAX TREATMENT tendency to the formation of membranes and shreds of fibrin. If the lung is suitably collapsed these will lead to its permanent encapsulation ; if the pneumothorax is only partial, or some expansion of the lung has been permitted, fibrinous bands may anchor the lung to the chest wall, split up the cavity into a number of isolated compartments, and finally lead to re-expansion of parts of the lung, causing, in some cases, an exacerbation of disease within it (see pp. 133 and 142). ^0 ^^-iM (4-) mmmJM Fig. 10. — A Diagram after Jaoot to iixusteatb the Types of Feveb associated with his FoirR Groups. (2) Pyopneumothorax becoming grave secondarily. This occurs in pneumothorax cases which are febrile, and where the general condition is bad. The early stages are similar to those of Group 1, the fever of onset wanes and the fluid remains serous for two to seven months (Class II, p. 124). When, however, the purulent stage begins the temperature again rises and the case terminates in a hectic fever to which the patient finally succumbs after some months in spite of treatment. The picture appears to differ from that of perforation of the lung in the less sudden and violent onset of fever and the absence of pain in the side. Four, or 16%, of PLEURISY AS A COMPLICATION 127 Jacot's cases were of this nature. He has not found that a pleurotomy improves matters and only resorts to it where pus formation is very rapid, mainly to relieve pressure. He prefers, but without any expectation of recovery, lavage with lysol, Lugol's iodine solution, or collargol. (3) Pyopneumothorax grave at first onset ; generally in cases of partial pneumothorax, still febrile and where the general condition of the patient is unsatisfactory. There is here no discoverable serous phase, but the onset with pain in the side and fever is precisely similar to that of a serous effusion save that the formation of fluid is more rapid. Treatment is by tapping and replacement with gas, and under these measures the fever drops gradually, often to a lower level than formerly, and the patient may do well in spite of the violence of the onset. Seven, or 28%, of Jacot's cases were of this nature, and of these five were cured within one and a half to ten months. In one case where he was tempted to perform a pleurotomy a fistula was still present one and a half years later. Jacot insists on the tuberculous nature of these acute cases which resemble the " septic " pleurisies in all particulars save in the absence of pyogenic microbes. He ascribes them to the breaking of a subpleural tuberculous focus into the pleural cavity. (4) Jacot's fourth class is nothing less than pyo- pneumothorax due to perforation of the lung, a subject discussed in this volume under a separate heading (p. 162). Five, or 20%, of his cases came into this group, and all died of the empyema within two to ten months. CHEraSTRY OF PNEUMOTHORAX EFFUSIONS Work done on this line, mostly that of Mayer, shows that the effusions of pneumothorax come well within the designation of exudates. Thus the specific gravity falls above 1018, a figure commonly accepted as the lowest limit for inflammatory effusions, and the albumin 128 PNEUMOTHORAX TREATMENT content is generally about 4% or over. Runeberg, as a result of his investigations of pleural fluids, found the following albumin content among different classes of effvision : Inflammatory Exudates . . 4% — 6% Congestive Transudates . . 1% — 3% Hydrsemia ..... below 1% These figures must be taken to apply to freshly formed effusions, since transudates of long standing show a rise of albumin content when they result from chronic stasis or are under pressure. The same also applies to the exudates of pneumothorax cases. Mayer has shown, also, that there is more albumin in those going about or at work than in those confined to bed. Among his eighteen cases the albumin content was mostly near the minimal point for exudates, 4% or somewhat less ; it was found to increase with the gas pressure, to rise where there was no tendency to absorption, but to sink if absorption was imminent. It was highest in long- standing cases. All the cases gave Rivalta's reaction (precipitation with acetic acid), a test which hardly ever occurs in transudates but never fails in exudates. Two examples of clear effusions from among the writer's cases examined by Dr. Roodhouse Gloyne may here be quoted : H.W. — Clear amber fluid with minute white deposit. Protein 3 5%. Cells chiefly endothelial, also a few lymphocytes and erythrocytes. Tubercle bacilli absent. Sterile on culture ; no antibodies. This was an effusion which rose 6 inches up the chest but was gradually absorbed by the end of six months. Mrs. F. — Transparent amber fluid with small amount of flocculent white deposit. Reaction alkaline. Specific gravity 1024. Rivalta's test positive. Coagulable protein (dried and weighed) 5-375%. Tubercle bacilh absent. Sterile on culture. Antibodies present in small amount. This effusion had to be tapped and replaced by gas on repeated occasions. PLEURISY AS A COMPLICATION 129 CYTOLOGY Very few cellular elements are present save in cases of mixed infection where the albumm content is generally high and the cells polymorphonuclear. These cases apart, a lymphocytosis is characteristic and the longer tne duration the more the cells. These " lymphocytes " are, according to Lippmann and Plesh, but offshoots of the serous endothelium and not of hsematogenous derivation, and with this opinion Mayer agrees. In addition endothelial cells or plates are often found, evidences, according to Koniger, of a traumatic irrita- tion. In the fluid from Mrs. F. referred to in the preceding paragraph, the cell count showed poly- morphonuclears 10 '6%, small round cells 80 '7%, and endothelials 8 "7%. Eosinophiles are present in a large proportion of cases and are generally considered to mdicate a favourable prognosis — 10% or 12% was a not uncommon figure among Mayer's cases. It is in these cases especially that the presence of antibodies in the exudate is said to be demonstrable. Polymorphonuclear leucocytes may be present in large numbers in quite chronic tuberculous effusions, as in a case of the writer's where a, sterile turbid fluid with a heavy deposit showed about 95% of these cells and numerous tubercle bacilli. In these cases, imlike acute infections, the pus cells tend to early degeneration and much debris is usually present. For figures of the proportions of purulent exudates likely to be met with the reader is referred back to p. 125. BACTERIOLOGY Apart from pyogenic infection — which may provide to view the pneumoeoccus, streptococcus, staphylococcus or other organisms — the question of bacteriology hangs round the presence or absence of the tubercle bacillus. As already remarked, Saugman found this organism in 9 130 PNEUMOTHORAX TREATMENT all his larger exudates save onlv two. von Muralt's experience was somewhat similar with tubercle bacilli demonstrable, either microscopically or by animal experiment, in all examined, save in five acute cases. Begtrup-Hansen found tubercle bacilli in eighteen or nineteen cases examined. Mayer, on the other hand, only found them in about 45% of his cases — in ten animal experiment was negative, contrary to the results usually obtained for " idiopathic " pleurisy. Partly on the strength of this finding and also on account of other factors, Mayer has made a classification based on the character of the exudate, and this necessarily differs somewhat from the purely clinical classifications already put forward. MAYER'S CLASSIFICATION (1) Exudates due to tuberculous pleuritis. — Very highly albuminous with few niorphological elements and often mixed at first but later becoming lymphocytic, es- ])ecially if tubercle bacilli are sparely present. No endothelial cells, eosinophiles few or absent. Resulting intra-pleural pressure not characteristic. Often exerts an unfavourable influence on the course of the case and associated with a spread of disease in the lung. (2) Exudates of acute infectional nature. — ^After tonsillitis, bronchitis or other infections. Shows high albumin content, mixed cytology passing to leucocytic, or leucocytic from the beginning. Leucocytes much degenerated. Fleeting eosinophilia. Tubercle bacilli mostly absent. Intrathoracic pressure tends to rise, absorption is slow and case tends to do badly. (3) Exudates due to perforation of visceral pleura (see p. 162). — The fluid is rich in leucocytes and these not degenerate but staining well. Endothelial plates, a few eosinophiles. Tubercle baciUi always present and often a mixed infection. (4) Exudates of a special type : the commonest form. — Albumin abundant, 3% to 4%; Rivalta reaction PLEURISY AS A COMPLICATION 131 + ; cytology spare and lymphocytic ; no endothelial cells. A high eosinophilia characteristic, up to 20%, and this accompanied by a blood eosinophiha. Specific complement-fixing antibodies, absent in the other types, are present in this, and exert a favourable influence. Tubercle bacilli are difficult to find, or absent. No increase of intrathoracic pressure. These exudates are of favourable outlook, especially where the eosinophilia is high or increases. They absorb quickly though they may reaccumulate withovit harmful results. Mayer ascribes them to the effect of nitrogen gas on a diseased and functionally disturbed pleura. It will be seen that Class I of the clinical grouping is made up of cases of this type. Class II probably covers the remainder of Type 4 of Mayer's classification, and also includes Type 1. Class III consists of Types 2 and 3 described by Mayer. The classification of Mayer, though based on all too few cases, is suggestive and stimulating, and well worthy of further verification on a larger material. EFFECTS OF PLEURISY IN PNEUMOTHORAX On this point it is particularly difficult to generalise since a pleurisy, according to circumstances, may exert either a harmful or a beneficial infiuence, or perhaps none at all. Even in the individual case it may be difficult or impossible to prognose the effects of an effusion before these have already announced them- selves. ' The points which work for good or evil, in these cases, may be separately considered. Beneficial Effects. — The value of an effusion in the course of many cases of phthisis may be regarded as common knowledge, its mode of action being generally accepted as immobilisation of the lung. But it has been shown that these mechanical effects do not cover the whole ground. Koniger's careful study of the question led him to remark that the beneficial effects might be present with minimal or fleeting effusions, and even i32 PNEUMOTHORAX TREATMENT after an attack of dry pleurisy. On these grounds it has been held that, immobilisation apart, some serological influence must be exerted, and this view is supported by the improvement said to follow pleurisy in a pro- portion of cases of artificial pneumothorax where all mechanical effects are of necessity excluded. This view seems to fall into line with the discovery of specific antibodies in the fluid in many cases, as reported by Mayer and also by von Muralt for the eti'usions of pneumothorax. An immunising effect may proceed from these similar to that aimed at byDebove's "Auto- sero-therapy," but of slow and long-continuing effect. A general improvement accompanying or following the advent of an effusion in artificial pneumothorax belongs to certain cases only, and according to Mayer (seep. 131) is the special attribute of a specific type of case. The writer has not been convinced that any lasting influence, at any rate a favourable one, was exerted by pleural effusions among his cases. The only possible exception to this was a patient with a tuberculous empyema which persistently recurred after removal, the patient mean- while putting on weight and appearing to improve in general health. A convenience, which may prove an advantage, following the occurrence of effusion in pneumothorax cases, is the marked reduction in the absorption of gas from the pleural surface. This allows of longer intervals between the refills, and removes the risk of lung expansion where the intervals, as often happens, tend to be longer than they really should be. It is quite probable, in the writer's opinion, that some of the improvement noted after an attack of pleurisy is due as much to this factor as to the serological in- fluences already outlined. In certain cases of partial pneumothorax, a mechanical advantage may accrue from the development of an effusion. In the first place adhesions may, it is said, be broken down by the pressure of the fluid and a more efficient cavity result. In the second place a stiffening of the mediastinum may allow of higher and more PLEURISY AS A COMPLICATION 133 effective pressures being employed than were permissible before the pleurisy occurred. Unfavourable Effects. — The serious effects of a pyogenic infection of the pleura will not be considered here — they will be found under the heading of " Per- foration of the Lung." But, these apart, the pleural effusions occurring in the course of artificial pneumo- thorax are by no means always of beneficent influence, and the patient's general condition may very seriously deteriorate. In effusions accompanied by fever and general symptoms, a temporary set-back is the rule, and this is often completely made up again, but in some this is not the case. This occurs particularly where, as not infrequently happens, a tuberculous pleurisy is accom- panied by a spread of disease in one or other lung. Such a spread may show itself by a haemoptysis, by an increase of sputum or accession of fever, or by the appearance of fresh physical signs. Not the least serious of the effects of a pleurisy may be that on the final expansion of the collapsed lung. Thickening of the visceral pleura may prevent or greatly limit this expansion ; if expansion is attained, on the other hand, the likelihood of adherent pleura is probably greatly increased. An occasional but most serious out- come of pleurisy is the gradual obliteration of the pneumothorax cavity under the irresistible pressure of contracting new fibrous tissue in the organised pleural layers. The effects of this pleural thickening, a true pleuritis deformans, do not come into play at once, but gradually a marked contraction of the side and drawing in of mediastinum and diaphragm come about as the fibrous tissue condenses. No gas pressure serves to delay the effect and a curious result commonly ex- perienced is a drawing out and partial expansion of the collapsed lung towards the sinking chest wall. Such a lung expansion, unpreventable by the injection of gas, may lead to opening up of cavities and caseous areas, and a fresh hghting-up of disease. The risk of such serious after-effects of pleurisy is a strong argument 134 PNEUMOTHORAX TREATMENT against the employment of pneumothorax therapy in cases still curable by ordinary sanatorium or other treatment, though it must be admitted that the occur- rence of pleurisy is less common in cases of slight and early disease. SYMPTOMS Acute cases apart, pleurisy in pneumothorax may show but few symptoms to mark its onset. It may discover itself by the appearance of an unaccustomed intrapleural pressure, or the patient may announce the presence of a " splash," or this may be discovered by the physician. W. N. announced a " splash " on moving on July 27th, and fluid was present to physical examination. This liad been pre- ceded by no symptoms, and no fever appeared till August 3rd. There was no rise of intrapleural pressure and all signs of fluid had disappeared a month later. This is a fairly average example of a pleurisy of Class I. In some cases an unaccustomed intrapleural pressure from diminished gas absorption appears to be pre- monitory of the appearance of fluid, and this latter not discoverable till a later examination. H.W. produced T.B. in his sputum for the first time on August 4th. A slight rise of temperature occurred between September 6th and 9th, and he lost some weight. On Septem- ber 28th, his initial pressure +2+11 showed that but little gas had been absorbed during his usual interval, and only 50 c.c. of N were required in place of the usual 500 c.c. Nevertheless no signs of fluid were present, this being first discovered some weeks later. Pain in the side and some amount of fever occur in the largest proportion of cases. The pain is generally in the side or back at about the level of the diaphragm, and is probably produced by a rubbing of the surfaces in the pleural reflection at this point. It may, on the other hand, be referred to the region of the shoulder (see p. 78). It may precede the appearance of fluid by many weeks. PLEURISY AS A COMPLICATION 135 Mrs. F. complained of pain in the left side on July 10th accom- panied by slight morning fever, headache, and general aches over the body. The pain was accentuated by deep breathing and by movements of the arm, and there was tenderness on deep pressure. No increased intrapleural pressure or diminution of gas absorption occurred till August 7th, when a small amount of fluid was for the first time present. Pain is not often severe and is generally relieved by aspirin and the application of warmth to the chest wall. Fever also tends to precede the appearance of fluid, and unless it is associated with pain in the side, its significance may be at first doubtful. On the other hand, fever may not appear till after the fluid is present, and may be very slight and fleeting. W. N., already quoted, showed a splash on July 27th, but no fever till August 3rd, when the temperature rose at night to 99 or 99-5 during ten days' duration, the fluid meanwhile increasing a little. Fever may be considerable in some cases, as 101° or 102° or even 103°, and yet only an inch or two of exudate follow. The persistence of fever for many weeks does not afford any indication that the fluid is becoming, or is likely to become, purulent. In some cases the temperature may drop and remain normal, and the patient feel well and comfortable, and yet the effusion steadily increase from week to week. Mr. C, after a violent onset with vomiting and high fever, produced only 1 inch of exudate which remained at this level during six weeks or more ; the patient by then felt recovered from his attack, but the fluid without further fever or other symptoms steadily rose to 7J inches in the ensuing six weeks. There seems to be a marked tendency of the pleurisy, once an exudate is present, to light up again on but slight provocation. The writer has found a nasal or bronchial catarrh, or the shaking of a journey to be apparent causes of exacerbation in some of his cases. Exposure to cold is made much of by some authorities, notably Forlanini and von Muralt, and it seems possible 136 PNEUMOTHORAX TREATMENT that the pleural surfaces may be more susceptible to such an influence where they are separated, as in pneumo- thorax, by a layer of gas. On this account it is wise to adopt particular precautions against chill in the case of pneumothorax patients living under sanatorium con- ditions. Nausea and dyspeptic symptoms may usher in the onset of effusion particularly, according to some, when this occurs on the left side. A right-sided pneumo- thorax case of the writer's vomited at the onset of a pleurisy and continued doing so two or three times daily for the following three or four days. The occurrence of pleurisy in pneumothorax is often suggested to the physician by the appearance of the patient. The face becomes pale and generally assumes that sallow yellow tint so highly characteristic of pleurisies of whatever nature. During the early stages, if fever and symptoms are present, the patient visibly loses weight and strength. When these have subsided the general health will again improve, though the fluid remains or even increases a bit. The writer has been impressed by the frequency with which pleurisy is associated with a pulmonary exacerbation. In such cases improvement will often quickly set in, but in bad cases it may mark the beginning of a serious or even fatal spread of disease. This was so in two of the writer's patients, both cases of advanced disease. In other cases such incidents as haemoptysis, an increase of sputum, or the sudden appearance in it of tubercle bacilli, or the appearance of fresh signs, have preceded or accompanied the onset of pleurisj^, and proved the more widespread nature of the attack. In many of these cases recovery does not appear to have been prejudiced by the temporary lung activity. EFFECTS OF EFFUSION ON MANOMETER READINGS These changes, due for the most part to a diminished absorption of gas, a lowering of elasticity, and stiffening 100 c.c. + 4+7 35c.c. + 4+8 30 CO. + 5+9 PLEURISY AS A COMPLICATION 137 of the mediastinum, are best brought out by a few case references. J. C. Before onset of pleurisy — 5—1 640 c.c. +5+6 Pleurisy began 26.7.15 6.8.15 . . -0+4 18.9.15 . . + 1 + 4 15.12.15 . ' . -4+1 These readings appear to show an increasing loss of elasticity so that enormous changes of pressure were produced by minimal amounts of gas. On these occa- sions but one or two inches of fluid, tuberculous pus, was demonstrable, but that part of the diminished elasticity was dependent on the actual fluid layer appears to be shown by a gain immediately after the fluid had been replaced by gas thus : 2.6.16 . . - 4+ 2 150 c.c. +5+ 8 rs. F. Before onset of pleurisy 23.7.15 . + 1+ 6 400 c.c. +7+10 With 1 inch of fluid 6.9.15 . + 1+ 6 140 CO. +7+11 25.9.15 . +11+15 gas removed + 6+10 Here again the presence of fluid reduced the elasticity, and in addition a decided rise of pressure occurred without any increase of fluid, suggesting an actual con- traction of the space brought about by irritation of the pleural surface. In this connection reference must be made to the interesting experiments of C. P. Emerson on the ac- commodative powers of the pleural cavity. He found that the chest, by elevation of the ribs and descent of the diaphragm, can accommodate very various quantities of fluid without any change of pressure. If fluid is con- tinuously injected into the pleural space, the pressure, of course, must rise, but it tends to do so in stages or jerks, owing to attempts on the part of the chest to accommodate itself to the increase and so keep down the pressure. This he ascribes to a reflex mechanism since 138 PNEUMOTHORAX TREATMENT his experiments were performed on dogs anaesthetised with ether, and some distm-bance of this mechanism may well afford explanation of certain of the curious pressure anomalies of hydro-pneumothorax, such as that just referred to, and even of the changes ascribed above to loss of elasticity. H. W. Before onset of pleiirisy 31.8.15 . -7-4 740 c.c. +3+11 14.9.15 . -5+5 520 o.c. . +7+14 After onset of pleurisy 28.9.15 . +2+11 50c.c. +2+13 This demonstrates diminished absorption of inflamed pleural surfaces. No fluid was demonstrable till some weeks later. Weiss has remarked that the respiratory fluctuations are diminished in the presence of pleural effusion, but this has not been the writer's experience. Thus for example : J. C. Respiratory fluctnations. Before filling. After filling. 4 cm. 2 cm. 4 cm. 1 cm. 4 em. 2 cm. 4 cm. 1 cm. 4 cm. 2 cm. 4 cm. 3 cm. 5 cm. 5 cm. 6 cm. 4 cm. Before pleurisy With fluid. 2 inches up . „ „ n „ „ j> >> t> ,, ,, PHYSICAL SIGNS Dry pleurisy may give but little evidence of its presence, since no rub can appear unless at the diaphragm- atic reflection. In one of the writer's cases fever and chest pain were followed by the appearance of an area of light impairment over part of the chest wall, sug- gestive of the deposition of pleural lymph though no fluid appeared. The writer has noticed areas of similar slight impairment round the spot where repeated punctures have been made, in cases where the increasing depth of puncture needed suggested the development PLEURISY AS A COMPLICATION 139 of pleural lymph. Such pleural membranes may be pushed before the needle with the creation of a false pocket — a painful and somewhat puzzling mischance. The X-rays will afford some help in the detection of pleurisy without effusion. Effusions. — The presence of a small quantity of fluid is easier to diagnose with pneumothorax than where this is absent. The " succussion splash " will probably detect even a few drachms of fluid if the stethoscope is applied close to the base and the proper impetus given. For this purpose the chest should be held between the two flattened hands, the stethoscope end being grasped between the thumb and index finger of one of them, and thus applied to the chest wall. Anything more than a trace of fluid will give an area of movable dulness at the base, and in gauging its amount it must be noted that the pleural cavity at the back of the chest (and the resonance in pneumothorax) extends fully 1 inch lower than the lower margin of the sound lung Thus with an inch of fluid, resonance extends to the same level on the two sides behind. With effusions larger than a trace, certain points determined by the presence of gas are noteworthy. The movable nature of the fluid has been already re- ferred to. In addition the parabolic curve so character- istic of the upper level in ordinary pleurisies is lost, at any rate in the larger effusions. In small effusions the writer has noted a modified parabolic curve, but in large effusions the upper limit is horizontal. Corresponding to this is the loss, to radiological examination, of Gar- land's triangle produced by lung retraction allowing the fluid to creep up to the highest point in the axillary region — the upper line is horizontal to X-rays as to physical examination. Dulness to percussion is similar to that of effusions generally, though the writer thinks it is less resistant and certainly the upper limit is harder to fix with accuracy. In place of Grocco's triangle at the opposite base the writer has generally found an area of quadrilateral outline ; this may also be observed in 140 PNEUMOTHORAX TREATMENT pneumothorax apart from the presence of effusion (see fig. 5, p. 85). In some cases, it is said, a mediastinal hernia may convey the impression of a double-sided effusion — a warning, indeed, against harmful intra- pleural pressures. To auscultation the feeble breath sounds of pneumo- thorax may alone be noted. In many cases, however, the fluid appears to conduct sound better than the gas above it, and loudish bronchial breathing and friction may be audible over the fluid, the friction, presumably, being produced at the extreme limit of the pleural reflection. In some cases signs of this nature may point to re-ex- panding lung. Where the lung cupola is adherent, bron- chial breathing and crepitations may appear at the apex above the effusion, and point to lung congestion or con- solidation, a not uncommon accompaniment of the larger effusions. These signs are readily conducted to the opposite side, especially in the interscapular region, and may lead to a very confusing clinical picture. Vocal resonance may be nasal and somewhat diminished with these pleurisies while vocal vibration, often quite good over the pneumothorax above, is lost over the fluid. Mayer has described tympanitic areas over the fluid caused by the presence of tongues of adherent lung ; these, if found, must be carefully avoided on puncture. To X-Rays the fluid takes a horizontal level as already mentioned ; the displacement of organs may be more or less according to the intrapleural pressure. In some cases fluid niay be held up by basal adhesions or may be at first interlobar, conditions for whose discovery the X-rays are better adapted than are physical signs. Saugman and Hansen regard the appearance of para- doxical diaphragmatic movements, Kienbock's phenome- non, as an early sign of effusion. TREATMENT The treatment of effusions in pneumothorax cases should be made as conservative as circumstances will PLEURISY AS A COMPLICATION 141 allow — it must necessarily differ according to the nature of the individual case. At the first appearance of symptoms suggestive of pleurisy the patient should be sent to bed, warmth applied to the side, and salicylates given. According to Saugman these latter exert a specific effect. Chronic EfEusions. — Small serous effusions should be left alone, the pressure being taken at intervals and kept at about tne accustomed figure by the occasional addition or removal of gas. The same measures apply to larger chronic effusions, but in these particular care must be taken lest the lung expand under the fluid and become adherent to the chest wall. Patients are often quite comfortable with a considerable bulk of fluid in the pleural cavity and some gas must always be kept above the fluid so that the lung remains under an effective positive pressure. This pressure may have to be raised if the lung gives evidence of re-expansion. If the fluid causes respiratory distress, particularly on the right side by hampering the thin-walled right heart, or if it threatens to fill the whole pleural cavity, it must be withdrawn and replaced by gas (see Gas Replacement, p. 145). This operation may have to be repeated several times, as in two of the writer's cases in which the pleural cavity has had to be emptied on four and five occasions respectively up to date. When the fluid has ceased to increase attention must be directed, especially by radiological examination, to keeping the lung col- lapsed by the needful refills at the same or an increased pressure, the patient meanwhile in many cases being able to go about as usual. Mayer recommends the trial of Debove's " auto-serotherapy," the subcutaneous in- jection of 1 c.c. of the exudate immediately on its re- moval, in the treatment of serous effusions. In cases where specific antibodies are present a reaction is said to occur in six to eight hours after the injection with fever and other symptoms similar to those experienced in a tuberculin reaction. Diuresis follows, and at the end of two or three days the pleural fluid has gone. 142 PNEUMOTHORAX TREATMENT von Muralt hesitates to approve of this treatment owing to the fact that virulent tubercle bacilH may be present in the pleural fluid. The larger effusions may remain unabsorbed for months or even years, and in many cases appear to do no harm even though the fluid is turbid or quite thick and contains numerous tubercle bacilli. If pleurisy occurs in a small partial pneumothorax, there is danger lest the cavity be obliterated and the lung drawn out by the action of organising lymph and spreading adhesions. For the avoidance of this the pressure must be still further raised (it has generally been high in these cases), and in some the fluid removed if not too thick to pass through a needle. But organisa- tion of lymph and fibroid contraction of the pleural space is a danger which may also threaten a complete pneumo- thorax under the influence of a chronic pleuritis (see p. 133). The fluid becomes thicker, the lymph masses organise and the collapsed lung becomes embedded in a mass of fibrous tissue. This has generally occurred, in von Muralt's experience, in cases of advanced Itmg disease and provides in such cases a not altogether un- satisfactory termination. In cases, however, where the other lung might need collapse-treatment at a later stage, such an occurrence must be warded off by the replacement of fluid by gas at an early stage, and Spengler and Sauerbruch recommend this measure for all effusions which show no tendency to absorption. Thereby, also, the risk of the lung expanding under the fluid and becoming adherent to the chest wall is avoided. Examples of forcible re-expansion of an unhealed collapsed lung through fibroid contraction of a pneumo- thorax cavity, with resulting exacerbation of pulmonary disease, are fortunately rare ; should this occur some form of plastic operation offers the sole remaining chance of obtaining further collapse of the lung. Acute Serous Pleurisy. — In effusions accompanied by fever, intolerance of the accustomed intrathoracic pressure may be evident, and the fever tend to drop only when this is reduced to atmospheric level. The PLEURISY AS A COMPLICATION 143 lowering of a high pressure may be aecomplished by merely releasing gas through the tap of the pneumo- thorax needle, but if the pressure is to be brought as low as the atmosjDheric this should be done by lowering of the pressure bottle so that gas is sucked back into the nitrogen chamber. It is important to remember that the tubing and filters must be re-sterihsed if this manoeuvre has to be carried out with the ordinary pneumothorax apparatus. If, in acute and febrile cases, thus lowering the pressure does not remove the symp- toms, the whole effusion should be drawn off and replaced with gas (see Gas Replacement, p. 145) without waiting for a reduction of activity. This commonly leads to a drop of fever and general improvement. Forlanini re- places with air instead of nitrogen in these cases, the choice of gas depending on its desired rate of absorption. The fluid may return once after removal or even twice, but very rarely, it is found, for a third time in these cases. Where effusion is accompanied by fever and other symptoms, it must be ascertained that these are not due to an underlying lung exacerbation before treat- ment directed to the pleural condition is undertaken. Acute Purulent Effusions. — These, whether tubercu- lous or the result of a mixed infection, must be treated conservatively no less than serous pleurisies, since if the pleural cavity is freely opened to the air the patient will be left with a suppurating cavity impossible to fill and close (see Perforation of Lung). At the outset the fluid should be removed and replaced by air or N. This, repeated if necessary, should suffice in the case of a tuberculous effusion (see Jacot's Group (3), p. 127). If no improvement follows this measure in the presence of a mixed infection, the fluid should be again drawn off and the cavity washed out through the pneumothorax needle with some litres of J% to 1% lysoform solution, con- tinued until the fluid returns clear through the aspirating needle. This manoeuvre, first recommended by Forla- nini, is highly commended by Spengler and Sauerbruch, who have found that most cases proceed to healing under 144 PNEUMOTHORAX TREATMENT this treatment. Dumarest uses warm lysol solution up to 5% passed through in small quantities. The injection of such antiseptics as quinine, coUargol, and iodoform preparations appears to achieve but little good in these cases, though von Muralt has seen improvement in small effusions so treated with electrargol. If a thoracotomy becomes eventually necessary this should be preceded, according to L. Spengler, by constant draining so as to reduce the cavity by lymph formation to the smallest possible dimensions before opening. In cases resulting from lung perforation the chances of saving the patient's life are sadly remote since the perforation in the collapsed lung can seldom be made to heal. Spengler has, how- ever, cured some cases by repeated plastic operations to reduce the size of the thorax until the wound was eventually closed (see Perforation of Lung). CHAPTER XII GAS REPLACEMENT This operation was first performed by R. W. Parker, in this country, in cases of empyema, and described in a paper read before tlie Medical and Chirurgical Society as early as April 1882. It was utilised, independently, by Potain two years later in the treatment of effusions complicating spontaneous pneumothorax. At a still later date Vaquez, Achard and others appUed it to serous effusions, and in recent years it has again been recom- mended by Kiiss and Wenckebach in the treatment of certain cases of empyema. Since gas replacement is occasionally needed in pneumothorax cases and has been referred to in foregoing paragraphs, some descrip- tion of the procedure seems called for here. Instruments needed. — (1) Those already noted as required for a gas filling (p. 43). (2) Po tain's aspirator and trocars. The apparatus should be fixed up before the operation and water aspirated through it so as to establish its efficiency. This is especially important because the pump usually supplied with this instrument is fitted to act, through different openings, either as exhaust pump or force pump. The use of the force pump by accident is likely to be followed by disastrous consequences. (3) Collodion dressings and adhesive plaster. The Operation. — The patient should lie flat while certain landmarks are pencilled, notably the border of the displaced heart and the usual lower level of the lung, or the lowest limit of gas resonance, i.e. the limit of the 10 145 146 PNEUMOTHORAX TREATMENT pleural reflection ; in all chest operations it is well to establish this graphic warning of the presence of sur- rounding organs. The patient then turns towards the sound side with the arm raised and the aspirating trocar is entered about the posterior axillary line or somewhat behind this at a level about that of the lower lung border or somewhat above this. In cases of effusion the diaphragm, owing to the weight of fluid and the gas pressure above it, is displaced considerably downwards so that the trocar might be entered much lower and still strike the pleural cavity. There is, however, no ad- vantage in this, and also a possible disadvantage, since, if the pressure is unduly lowered, the diaphragm may rise and obstruct the trocar, an accident whose explana- tion might not be at once understood. When the trocar is in place, the patient is turned back so that the point of aspiration occupies a dependent position. The trocar is now withdrawn and a little fluid sucked into the aspirating bottle to determine that the cannula is in place. The pneumothorax needle is now entered in front at a point where good resonance guarantees the presence of gas. This is best accomplished at the base of the chest, the patient's head being lowered so that the gas accumu- lates in this region — the chest wall is thicker to perforate near the apex and adhesions more likely to be entered. As soon as the position of the needle in the pleural cavity is established by the usual manometer fluctua- tion, the pressure is recorded and the removal of fluid begins. The aim should be to maintain the patient's usual positive pressure as near as may be throughout this operation. If the initial pressure is high, some fluid may be released before the gas bottle is opened, but afterwards it is, in the writer's experience, best to keep the gas flowing in at an even pressure all the while the fluid is running out. This is accomplished by gradually raising the pressure bottle as the gas flows, so as to maintain, as recorded by tlie manometer, the same positive pressure, perhaps some 5 or 6 cm. H2O mean ^T 147 pressure. This, the ideal method, can only be carried out where the movable bottle type of pneumothorax apparatus is used. Those who work with fixed bottles recommend, as does Forlanini, the alternate flow of fluid and gas, 100 c.c. to 200 or 300 c.c. at a time, a somewhat tedious procedure where 2,000 or more c.c. have to be displaced. A case of the writer's which he displaced partly by this alternate method showed the following readings : Gas put in. 150 c.c. 200 c.c. 200 c.c. 200 c.c. 470 c.c. Fluid TLemoyed. 200 c.c. 200 c.c. 200 CO. 300 c.c. 230 c.c. Manometer reading. + 2+6 - itoO + 4+8 - 4to0 + 4+8 - 4to0 + 6+9 - 3 toO + 2+6 -8-2 + 4+8 Here it will be seen that considerable alterations of pressure were produced, and though these might have been reduced by the replacement of only 100 c.c. each time, yet the continuous displacement is admittedly more perfect and seems to the writer a further argument in favour of a movable-bottle pneumothorax apparatus. The removal of fluid in any quantity without its re- placement by gas is, of course, not to be considered in any pneumothorax case. There is no longer a displaced lung to take its place, as in an ordinary pleural effusion, the fully collapsed lung (perhaps fortunately) will not readily expand, and aspiration of fluid soon develops a high negative pressure for which a harmful displacement of the mediastinum and rise of the diaphragm will attempt to compensate. This is, of course, a very undesirable result likely to lead to pain and distress, and if the lung expands, possibly to more serious symptoms. Thus Weiss, who aspirates to a negative pressure causing cough, dyspnoea, chest pain, and inspiratory retraction in 148 PNEUMOTHORAX TREATMENT the supraclavicular region, before he replaces with gas, finds that the operation is nearly always followed by a rigor and fever above 39° C. (102 "2° F.) of several days' duration, a result which can hardly be considered surprising. When the fluid has all been removed gas and froth begin to bubble out through the trocar, and this should now be removed. In doing this the puncture must be carefully sealed with gauze and collodion, and over this a pad may be strapped. Leakage into this needle track is very apt to cause an inflammatory swelling or even an abscess and sinus. The gas pressure is now adjusted to a suitable figure, and this is usually positive but somewhat below the usual since it is more likely to be augmented by the secretion of further fluid than lowered by absorp- tion of gas. Forlanini meets this difficulty by replacing with air (see remarks on p. 33). If the operation has been skilfully performed the pneumothorax side should be resonant even down to the limits of the pleural reflection 1 inch below the lower margin of the opposite lung in the dorsal region. It is interest- ing that the amount of gas required seems always some- what less than that of the fluid displaced, even allowing for differences of initial and final pressure. This seems to the writer due to the greater displacement of diaphragm and lower levels of the mediastinunr by the fluid on account of its great weight, tliese structures recoiling somewhat under the lower pressure of the elastic gas. In a case of the writer's 1,630 c.c. of fluid was replaced by 1,420 c.c. of gas, and in another, on three occasions, 2,500 c.c. of fluid by 1,830 c.c. of gas, 1,400 c.c. of fluid by 1,350 c.c, and 1,450 c.c. by 1,250 c.c, to bring the intrapleural pressure up to the same figure. The eftect on the patient of a gas replacement properly , performed is quite insignificant, being little more than thatcaused by an ordinary nitrogen refill ; the operation is very much less serious than an aspiration alone on account of the avoidance of changes of intrathoracic GAS REPLACEMENT 149 pressure and displacement of organs. Indeed, in an ambulant pneumothorax case the operation can ,be quite safely carried out in the consulting room, and the writer has done this on many occasions, the patient afterwards returning home in a taxicab under the care of his medical attendant. This could, of course, never be done with safety in the case of an ordinary pleural effusion. It has happened in some cases that a fistula develops in connection with the puncture wound. Weiss and Brauer have each had a case of this, and the latter remarks that, if secondary infection is avoided, the fistula closes again after some months. The writer has experienced subcutaneous inflammatory swellings the size of a tangerine orange at the puncture sites in a case with chronic purulent effusion containing abundant tubercle bacilli. These were reminiscent of excessive local reactions to tuberculin and remained over long periods, and even proceeded to definite fluctuation. One of these cold abscesses after a period of three months suddenly acquired a pneumococcus infection and became acute, but fortunately no further trouble occurred after this was opened, and no sinus formed. In cases of effusion in pneumothorax it may happen, as on two occasions to the waiter, that the whole pleural cavity has become full of fluid before the case presents itself for aspiration. This must be avoided when pos- sible, since the operation is more troublesome where no adequate gas bubble can be found for the pneumothorax needle to enter. The conditions are then similar to those of pleural effusion apart from pneumothorax where a replacement with gas has been decided upon. In such cases a needle of wide bore is indicated, such as the writer's trocar and cannula depicted on p. 34, since otlierwise pressure fluctuations may fail to be com- municated to -the manometer through the fluid layers. Where this is the case anxiety will be/felt lest the needle has entered an adhesion, an accident particularly warned against by Mayer in these cases. In the absence of a manometer fluctuation evidence that the end of the needle 150 PNEUMOTHORAX TREATMENT is in the pleural cavity may usually be obtained by the appearance of fluid in the manometer tubing ; a small length of glass tubing near the needle should form part of the usual apparatus, so that this may be visible. Failing its appearance a syringe, after the pattern of the " safety " syringe of Forlanini, may be applied to the end of the needle, and fluid aspirated into it. The position of the patient may be moved to facilitate this test if it does not come off at the first trial. As soon as fluid appears the intrapleural location of the needle end is established and gas may be entered while fluid is withdrawn just as already described. The gas tubing shouiu ijfi kept raised so that fluid cannot run far up it and the continuous passage of gas will usually suffice to dry the needle, so that proper manometer readings are available before the end of the operation. CHAPTER XIII THE ACCIDENTS OF PNEUMOTHORAX TREAT- MENT PLEURAL SHOCK Pleural shock presents a series of symptoms which are generally indistinguishable from those caused by gas embolism (see p. 156). On this account difference of opinion as to the cause of symptoms arising in individual cases is bound to occur, and cases of embolism have probably been from time to time described under the heading of pleural reflex or pleural shock. But in spite of this there seems no doubt that pleural shock is a reality, and the fact that it tends to repeat itself time and again in the same patient suffices to exclude em- bolism in a proportion of cases. Symptoms. — The symptoms have appeared as a rule with great suddenness at the moment when the needle is entering or leaving the pleural space. More rarely they come on after an interval of fifteen to twenty minutes, or even after some hours. In a typical case the patient develops a sudden pallor, loses consciousness and soon becomes cyanosed. The pulse and respiration become irregular, the pupil large, and clonic and tonic spasms of the limbs ensue, or these occur only on the side of the pleural reflex or in isolated muscles. Some patients cry out at the onset. Forlanini divides the symptoms into three groups : Psychic Symptoms. — Giddiness and loss of conscious- ness. 151 152 PNEUMOTHORAX TREATMENT Motor Symptoms. — Tonic and occasionally clonic spasms, flaccid paralysis with contractions, hemiplegia or monoplegia. Never involuntary micturition, seldom vomiting. Circulatory and Respiratory Symptoms. — Rapid, weak, irregular pulse, pallor of skin, cyanotic patches over face, neck and thorax. It is impossible to say amid this medley of symptoms what, if any, have been drawn from cases which were in reality embolic. De Cerenville thinks that a flaccid paralysis points to embolism and not to pleural refit x, of which, also, a much more acute disturbance is char- acteristic. His experience of shock symptoms, the result of irrigation of the pleural cavity, include the appearance of unequal pupils, convulsions, coma, and sudden death. Occasionally he has seen vaso-motor disturbances — general skin flushing, retinal haemorrhage. Mere contact of a probe with sensitive granulations sufficed to bring about symptoms in some cases. Symptoms of shock last a few minutes up to half an hour or even for hours (Forlanini), and no memory of the attack remains after recovery of consciousness. Conditions of its Appearance. — It seems a peculiarity of pleural shock that it is apt to reappear at subsequent fillings, and to get worse and even become fatal if treat- ment is persisted in (Forlanini). Forlanini remarks that if it occurs twice it is best to give up the treatment. In 134 cases with over 10,000 fillings he met it in 12 cases, in 5 of whom it was repeated and treatment abandoned. Other authorities have had much less experience of shock, and the main description of this condition comes from the writings of Forlanini. Saugman sounds the warning that wc must only speak of shock when embolism can be excluded, and Brauer is even more emphatic on this point. Saugman during seven and a half years of pneumothorax treat- ment had 22 accidents altogether, of which 6 were mere faints. From December 1906 to October 1910, when his first fatal accident occurred, he treated 98' patients THE ACCIDENTS OF TREATMENT 153 with 2,200 punctures and experienced 14 accidents. In those days he was careless of puncturing the lung. From October 1910 to September 1912, when his second fatal accident occurred, he treated 63 fresh patients with 1,200 punctures and only 2 accidents. During this period he exercised great care to avoid wounding the lung. Since September 1912 he had treated (up to date of communication in 1914) a further 49 fresh cases with 2,000 punctures and no accidents. It is difficult to say how many, if any, of these accidents could be ascribed to pleural reflex. He thinks shock greater where blunt instruments are used — he is also careful to avoid more than two punctures at a first operation. If shock occurs he desists from further operation unless an open incision after Brauer's method. Sachs records 26 cases of pleural shock, none of them fatal, among 1,058 pneumothorax cases from the American literature, a percentage incidence of 2 "5. Lillingston saw slight shock on only one occasion among 32 patients. Brauns ex- perienced pleural reflex with fainting twice in one nervous patient. The writer with many hundreds of fillings up to date has had no experience of shock, though he uses a blunt instrument at all first operations. De Carle Woodcock and J. A. M. Clark have noted that shock symptoms are much more commonly seen in cases of early disease where the pleura is healthy, than in cases of more advanced tuberculosis — a strong argument, in the writer's view, against the treatment of early phthisis by artificial pneumothorax (see p. 19). It is reasonable to expect, if pleural shock is a real entity, that many cases must have been recorded during various surgical manipulations of the pleura in the past. This is the fact, and Zesas has been able to collect from the literature records of 54 cases where serious pleural shock occurred. Twenty-one of these were of fatal issue, but 33 recovered ; all were in cases of empyema. Thirty- two occurred during irrigation of, and injections into, the pleural cavity ; 14 during empyema operations ; and 8 during explorations. In 45 the clonic and tonic 154 PNEUMOTHORAX TREATMENT convulsions came on at once, but in 8 after some minutes or even hours. In 3 patients they recurred on several occasions. In some mere touching of the pleural surface with a probe or drainage tube sufficed to elicit symptoms. In these the evidence for pleural shock seems overwhelming. In some others, where the manoeuvre which led to symptoms had been previously practised, often many times, without result, the ex- planation seems more doubtful. In such, a difference of temperature of the lotion or injection-fluid might afford an explanation, and this was recorded by Paget and by Cayley in a couple of cases. But the possibility of embolism, whether of blood clot or gas, cannot in these cases be excluded. Indeed its occurrence seems pro- bable in that paresis or paralysis of one or more limbs, or complete hemiplegia with disturbance of speech, is described as the after-effect in a proportion of cases. In animals it has been shown that mere mechanical irritation of the pleura will not cause shock, but it is readily produced by the introduction of chemical irritants such as tincture of iodine, phenol, acetic acid, and alcohol. Cordier showed by experiment that shock occurred mainly through the vagus nerve — but he also found that ligature of the carotids prevented its oc- currence, and that when the ligature was cut and the cerebral circulation re-established after two or three minutes, convulsions and death ensued. Cordier dis- covered, in addition, that shock could be avoided with certainty if an anaesthetic was given at the time of the operation, or morphine beforehand. Prophylaxis.— The treatment of shock is especially prophylactic and its most important point is based on the discovery of Cordier referred to above. In all cases an injection of morphine or omnopon (p. 42) should be given before the initial operation, and in some nervous and apprehensive patients it may be repeated before the first one or two refills. In addition the careful use of a local anaesthetic, of which novocaine appears to be the safest (pp. 43 and 44), giving time for it to act before each THE ACCIDENTS OF TREATMENT 155 forward movement of the needle, should reduce the risk of pleural shock almost to nil. Indeed in recent years, since precautions of this sort have become universal, one hears but little or nothing about pleural reflex. Other precautions consist in the avoidance of the menstrual and premenstrual period in women, and confining the first operation to two punctures according to the advice of Saugman. There is no doubt that the patient's nervous stability is visibly lessened by the strain of two unsuccessful punctures at a first operation. The writer has been accustomed to give a small dose of alcohol or sal volatile at the first operation in some nervous patients. Saugman also advises that an assistant keeps a finger on the pulse during the initial operation, and that this be suspended if any irregularity or marked diminution of volume occurs. Treatment. — If in spite of these precautions " col- lapse " symptoms should appear, their treatment must be prompt indeed if success is to be achieved ; the necessary drugs and apparatus must be to hand in case of such symptoms arising. Since embolism cannot be excluded, a line of treatment appropriate, so far as possible, to either or both conditions must be adopted. The needle should be immediately withdrawn, the patient's head dropped as low as possible, and ether or ether and camphor injected. Saugman recommends faradisation of the heart region and of the phrenic nerve, rhythmic traction of the tongue, and finally the intravenous injection of digalen. Forlanini gives, in addition to restoratives, whiffs of ether or chloroform and performs artificial respiration. This last measure should, however, be confined to cases, such as those of repeated attacks, where gas embolism can be excluded with* some certainty. 156 PNEUMOTHORAX TREATMENT GAS EMBOLISM Cases of gas embolism have been, as already sug- gested, often attributed in the past to pleural shock. As a matter of fact it has never been possible to dis- entangle the symptomatology of these two conditions, and there remain but few points upon which a differential diagnosis may be based during life. The exclusion of conditions which might have led to embolism is notori- ously difficult, since the gas may come from the mano- meter tubing or, according to Brauer and others, from the alveolar air, when the needle has entered the lung. Two points stand out above the mass of general symptoms as capable of giving the clue in an individual case — on the one hand the reappearance of symptoms at more than one filling is strong evidence of their reflex nature and points to pleural shock ; on the other the discovery of gas bubbles in the retinal vessels, as es- tablished by Brauer and his co-workers Stargardt and Wever, is definite proof of gas embolism. Unfortunately these diagnostic points are not always available. Even in fatal cases with full postmortem examination the evidence of gas embolism may be doubtful or incom- plete. Gas is normally present in the veins of the pia mater, but, this apart, the discovery of gas in the blood stream at any point may be held sufficient, in the opinion of Brauer, to establish a diagnosis of gas em- bolism. Mechanism of Gas Embolism. — It seems fairly certain that the introduction of gas into the greater circulation, as by puncture of a vein in the chest wall, in the quan- tity and with the rapidity probable at a pneumothorax operation, is hardly likely to give rise to gas embolism. Nevertheless this accident happened to a case under the care of Zink. Such gas is carried through the right side of the heart into the pulmonary circu- lation and its passage to the brain is blocked. It is, then, the lesser or pulmonary circulation with which we must concern ourselves, since gas from a THE ACCIDENTS OP TREATMENT IS-J" pulmonary vein enters the left heart and may proceed straight to the cerebral circulation. In the absence of adhesions the lesser circulation is sharply cut off, and it is only by entering the lung that a pneumothorax needle can reach it. On the other hand, when pleural adhesions are present a wide communication between the two circulations may exist through the formation of new blood vessels, such vessels often becoming almost angiomatous in character. But even when this con- dition exists it appears that the blood usually flows towards the greater circulation, and substances injected into the pulmonary veins appear in the intercostal and diaphragmatic vessels. To cause cerebral embolism it seems as if it must be necessary to wound the lung deeply, and this probably accounts in part for the rarity of this accident in recent years when the lung's integrity has been more scrupulously respected than in the past. Intravenous Pressure. — The entry of gas into a wounded vein depends on the negative intravenous pressure, a force which is maintained both by the suction power of the thorax and, to a less extent, by that of the heart. The negative pressure in the intrathoracic veins Ues but little above the general intrathoracic pressure and shows the same fluctuations during inspiration and expiration. Thus Piper in experiments on cats and dogs with a cannula passed down the jugular vein into the thorax found such pressures as —12 cm. HgO during expiration, —14 "8 cm. during ordinary inspiration, and —19 cm. on deep inspiration. These showed a mean difference, when cardiac fluctuations of pressure were taken into account, of about 5 cm. HsO above the general intrathoracic negative pressure taken at the same operation, both during inspiration and expiration. But the minimum difference amounted to little over 3 cm. H2O, and if these figures are at all apphcable to man it is obvious that a very powerful suction power exists in the pulmonary veins, particularly during inspiration and the heart's diastole. Such a suction power might amount in man perhaps to —7 or 8 cm. H2O 158 PNEUMOTHORAX TREATMENT or even more, and in such an accident as blocking of a large bronchus in a case with adherent pleura, a very much higher negative pressure might be developed. The accidental entry of gas is perhaps niost likely to occur where a vein is surrounded by disease which stiffens its walls and hinders its collapse. The Source of Gas Embolism. — The lung being wounded there are only two sources, in a properly conducted operation, from which gas can enter a pulmonary vein — one is the alveolar air, and the other is the tubing leading to the manometer and gas apparatus. Alveolar Air. — As to the first, its possibility is vouched for by such authorities as Brauer and Spengler, and has the support of Saugman. These authorities point out that the alveolar air is at a slight positive pressure, while the venous pressure is negative, especially during inspiration. Saugman has experienced even mediastinal emphysema from mere wounding of the lung, though here there could be no negative pressure to assist. Brauer had himself a case of fatal embolism during the use of a Paquehn's cautery whereby a vein only the size of an ordinary hat-pin was wounded. In this case gas bubbles were observed in the retinal ves- sels. Brauer remarks that this accident is especially liable to happen where the vein is stiffened by sur- rounding infiltration ; he also insists that but a small volume of air may be required, and inspiratory suction, apart from the cardiac suction, may suffice to bring it about. On this account Saugman insists that the patient be warned against deep breaths and coughing during the early stage of the operation. Begtrup- Hansen has recorded a case with adherent pleura where the lung was wounded and symptoms began two or three minutes after withdrawal of the needle, coughing having led to fatal embolism. In a case of Mayer's a deep inspiration displaced the needle into an adherent tongue of lung tissue and a fatal embolism followed. From the Manometer Tubing. — Gas from the mano- meter tubing seems a possible source of embolism. -— ^^^:zi::^i——--^—^-:eatment 159 particularly if such tubing and the manometer are of needlessly wide bore. Not only must the amount of gas displaced from one arm of the manometer by a rise of say 5 or 6 cm. be considered, but also the expansion of the whole volume of gas in the tubing and apparatus as a result of the negative pressure. Ihis might supply a dangerous volume of gas in apparatus of wide bore. On this account and also for other reasons (see pp. 28 and 29) the writer advises strongly that apparatus of narrow bore, '3 or "4 cm. diameter, be used, but it is questionable whether even this affords complete protection against gas embohsm if the lung is pierced or torn. Saugman, to avoid this danger, uses at the initial operation a water manometer close to the needle and of such bore that a fluctuation of —14 cm. allows the entry of only 1 c.c. of air or gas. This measure was devised and recommended by Wiirtzen. Volume and Nature of Gas. — In considering the danger of gas embolism it is obvious that the volume and the nature of the gas are of importance. Forlanini, in operating on dogs, found that 6 to 8 c.c. of N could be put into the left ventricle, or 2 to 3 c.c. into the carotid, without symptoms of embolism resulting — more than this gave serious symptoms. Saugman considers that 20 to 30 c.c. are needed to produce fatal embolism in humans, but Brauer records a death after 15 c.c. of gas, though the diagnosis of embolism was unverified by post-mortem examination. Brauer holds the results to depend more on the part of the brain implicated than on the quantity of gas introduced. Many operators use O2 at the initial operation on the recommendation of Deneke with the belief that it is quickly absorbed in the circulation and hence less likely to lead to embolism than is N or air. Brauer, however, points out that though O2 is quickly absorbed in venous blood, and hence in the greater circulation, this does not apply to the pulmonary veins, the real source of danger, since in these the blood is arterial. Experiments carried out by him, with the assistance of Wever and Stargardt, on 160 PNEUMOTHORAX TREATMENT rabbits, dogs, and apes, go to show that the kind of gas matters Httle, and that embolism occurs alike with O2, with air, or with N. Symptoms. — The main symptoms of gas embolism are those already described under the heading of Pleural Shock — any nervous symptoms may occur according to the area of the brain or cord affected, and they are generally indistinguishable from those which occur as the result of pleural reflex. Possibly the onset is less abrupt in embolism, and the patient may have time to complain of feeling ill before unconsciousness supervenes. Brauer describes the appearance of skin " marbling " where the gas reaches the small peripheral vessels, and says that this is followed by the appearance of pale areas and cyanotic patches, and later by patches of hyperaemia like an exanthem. Characteristic, according to Brauer, is the immediate cessation of respiration and circulation, and he has seen recovery after one minute of such cessation, and after one hour of unconsciousness. Prophylaxis. — Gas embolism has become a rare com- plication since the introduction, by Brauer, of the water manometer, and a few simple precautions should render its occurrence well-nigh impossible. These precautions may be briefly enumerated. Avoid puncture or tearing of the lung. The first is assured by abstention from the use of a sharp needle at the initial operation — the second by extreme gentleness in the use of the blunt needle or cannula, particularly in the presence of adherent pleura. Tearing of the lung tissue may also result from the use of very high intra- pleural pressures. Never open the connection with the gas bottle until, or unless, a characteristic manometer reading with respiratory fluctuations is obtained. This precaution must be as scrupulously followed at refills as at the first operation, since it can never be certain but what the needle has entered a pleural adhesion. Mayer ex- perienced a fatal gas embolism from the withdrawal of the needle during a refill into a tongue of adherent THE ACCIDENTS OF TREATMENT 161 lung tissue through which it had passed into the pneumo- thorax cavity. Let the gas flow in at a decided negative pressure at the first operation by lowering the water-bottle level some 7 or 8 cm. below the level of fluid in the gas bottle. Avoid any attempt to obtain manometer fluctuations, where these are absent, by getting the patient to take deep breaths or to cough (see, however, p. 68). During unavoidable coughing tjie tubing should be closed with finger pressure near the needle, partly to cut off the gas, and partly to prevent the manometer fluid being coughed out. Never pass the blunt stilette for clearing the needle without the use of a stuffing box or a rubber cap. Saugman experienced a fatal accident where this pre- caution was not adopted. Do not use rubber and manometer tubing of needlessly large bore, nor have tubing of needless length. Treatment. — The treatment of gas embolism must be in many respects similar to the treatment of pleural shock {q.v.), since it is genei'ally impossible to diagnose between them. The needle must be immediately with- drawn and the patient inverted or the head hung low, since gas tends to rise. Diffusible stimulants — -ether, camphor, etc. — ^must be injected and faradisation of the heart may be tried after the advice of Saugman. Arti- ficial respiration must be strictly avoided, as likely to pump more gas into the vessels, but rhythmical tongue traction may be tried. SURGICAL EMPHYSEMA Superficial. — This is apt to occur where a large needle is used, where a high intrapleural pressure is maintained, where the pleura is stiffened by disease, where the patient's tissues are very deficient in tone, and especially where much coughing follows the filling. The manoeuvre of lifting and manipulating the tissues leading down to the pleural puncture, so as to obliterate the path of the 11 ]GJ PNEUMOTHORAX TREATMENT iiccdlc us this is witlulrawu (sec p. J-T), luis practically always sufficed to prevent this accident in the writer's experience. When emphysema occurs a puffy swelling appears, giving the characteristic crackling sensation to finger or stethoscope. A pad and pressvu'c and pre- vention of cough, so far as this is possible, will generally help to check its further sjiread. Deep or Mediastinal. — This occvn-s where the gas gets under the visceral pleura into the interstitial lung tissue. From there it tracks into the mediastinum and appears in the neck above the sternum. The symptoms are sore throat and a sensation of tightness rovmd the neck, difficulty in swallowing and breathing, and a nasal intonation of voice. Pain in the chest may be present and the patient may become cyanosed, greatly to the alarm of himself and his medical attendant. Never- theless, no harm ever results, though the condition may recur and even prevent the continuation of treatment. PERFORATION OF THE LUNG This, the most serious accident to which a pneumo- thorax patient is liable, is fortunately of comparatively rare occurrence. Forlanini cxjjerieneed 8 perforations among 139 patients ; all were in the stage of " clinical recovery," before anatomical healing was attained. Jacot fotmd 5 cases, or 20%, among his collection of pru'ident effusions. The conditions Avhich predispose to this accident are : (1) The presence of large cavities or caseous areas, and this forms one of the arguments against needless delay when once the jDatient has passed beyond the reach of other methods of treatment. (2) The presence of an incomplete pneumothorax in a patient who seeks to live an ordinary life, and this appears to be the case particularly where uncoUapsed cavities are present. Saugman experienced this accident in a girl patient of his as the resvdt of riding and dancing. THE ACCIDENTS OP TREATMENT 163 Those who practise pneumothorax tlicrapy will have experience of patients with large superficial cavities, which separate only partially or with difficulty from the chest wall, and generally maintain firm adhesions else- where, and these patients form a constant source of uneasiness on this account. Spontaneous perforation seems to be especially common where advanced disease is present in the lower lobe ; W'Ounding of adherent caseous kmg at a first operation seems a possible pre- disposing cause to a subsequent perforation. Forlanini considers the perforation in all cases due to rupture of a softened caseous area and not to the mere progress of disease — he thinks the risk greatest where the pressure has been allowed to fall too low before a refill. All his cases, save one, were in men, and all considered themselves well and pursued an ordinary life. Jacot particularly insists on the i61e of strong untearable adhesions in the causation of lung perfora- tion ; he has commonly formd the pleural rupture at the point of attachment of a stretched adhesion. He admits, however, the occasional responsibility of a quiet ulceration. The rupture, he points out, is not neces- sarily into a cavity. Symptoms, — The symptoms are sudden and violent pain in the side followed by a hectic temperature wdth a rise of about 3"5° F. (2° C), and the rapid appearance of a pleural effusion. The signs may not differ much from those already present. When the perforation remains open an intense whistling sound may accompany the respiratory murmur. A drop of intrapleural pressure and inability to re-establish it by the introduction of gas is evidence of considerable value. Often a plug of tissue makes the opening valvular ; the manometer may then show a step-like rise of pressure. In addition, a gas analysis, when this can be undertaken, will show an abnormal percentage of oxygen — proof of an open fistula which was demonstrated by Leconte and Demarquay as long ago as 1863. Prognosis. — Rupture of the lung is generally of fatal 164 PNEUMOTHORAX TREATMENT issue, for whereas iu spontaneous pneumothorax the resulting collapse of the lung tends to close the wound and favours healing, in these cases, where the lung is already collapsed so far as this is possible, the rupture does not close but even tends to gape still further. The pleura suffers a constant reinfection and a fatal empyema generally leads to death ^vithin a year. Prophylaxis. — ^Most important is it that the activity of patients with partial pneumothorax and cavitation be kept within reasonable limits, and this particularly at the time when a fresh refill begins to be required. Re- strictions should be put on violent and sudden move- ments rather than on physical exertion of a quiet con- tinuous nature. Patients with complete collapse of the lung do not run this risk of perforation. Treatment. — Where cases do not die quickly of acute infection, the pleural exudate accumulates and is coughed out from time to time through the lung fistula, either accidentally or as the result of purposeful changes of position. Hectic fever occurs, the patient grows weaker, and amyloid degeneration is hkely to follow. For such cases ordinary rib resection offers no hope, as the lung cannot expand and the cavity cannot close. A hopeful method has, however, been devised by L. Spengler, a method which has indeed led to the recovery of seven among thirteen cases so treated. Its aim is the closure of the cavity by a series of aspirations each followed by a plastic operation. The fluid is drawn off and one or two days later an extrapleural thoracoplasty is done over the lower part of the chest. Two or three weeks later a further aspiration followed by a thorax resection of the upper chest is joerformcd. A third operation may be needed, and in a few^ cases the cavity is then com- pletely closed. In the remainder the marked reduction in size of the suppurating cavity allows a great imjsrove- ment in the patient's general health, and after a time a Schede's operation, removal of the parietal pleura and soft parts, assures a further closing down of the cavity. Only three operation deaths occurred among THE ACCIDENTS OF TREATMENT 165 the thirteen cases as against a mortality of 80% for Sehede's operation done at an earlier stage. A further three cases died of tubei'culosis after weeks or months, leaving seven recoveries, of which four became fit to return to work. CHAPTER XIV GENERAL RESULTS OF PNEUMOTHORAX TREATMENT In a disease so protean in its forms, and of such different outlook at various stages and in different individuals as is phthisis, no matter needs more cautious and judicious handling than do the statistical results of any mode of treatment. It may be said Avith much truth that no individual case of phthisis is strictly comparable to any other case, and this fact is especially brought home to him who sets out to generalise on a collection of phthisis material. Nevertheless, generalisation must be attempted, and some sort of statistical compilation is, in the long run, inevitable, since in this way alone can a large material be brought before the eye. With this in mind the writer has collected such figures as are avail- able from the works of certain experienced operators, and puts them forward for what they are worth. Since each author has his own special method of correlating his results, the figures are not often comparable one with another and each must be taken as it stands. The matter is further complicated by the fact that the number of completed cases of any one operator is not, as yet, large, and that the periods since cessation of treat- ment vary and in many cases are too short to form a basis for conclusions. It may, indeed, be said that the time for a statistical summing up of the results of pneumothoi'ax treatment has hardly yet arrived. Never- theless, though the ultimate percentage of complete cures is hardly yet to liand, the enormous volume of 16G GENERAL RESULTS OF TREATMENT 167 clinical improvement clearly emerges from the available figures, and even if the results were not permanent, yet the prolongation of life, increased comfort and Avell being, and return of power to work — and this for the most part among cases of advanced disease — entitles the operation of pneumothorax to a very high place among successful therapeutic measures. Saugman (1913) introduces his results with the follow- ing remark : ■' If you are to judge of the results of the treatment with artificial pneiimothora.x, it must at once be under- stood that this treatment deals with verj' severely attacked third-stage patients, of whom a great part, indeed most of them, Avithout this treatment would have l^ractically no chances of recovery and healing." After recording his agreement with Brauer as to the im- possibility of dealing statistically with these cases, he offers a " kind of statistics " in the following form : Thirty-seven of his cases were " quite hopeless " owing to disease in the better lung being of considerable extent (covering more than half the surface), or pro- gressing, or owing to complications which themselves threatened Mfe. In 8 of these he failed to produce a pneumothorax and these have all died, 6 within six months, 1 after one year, and 1 after two years. Of the remaining 29 in whom a pneumothorax was produced, 25 are since dead. In some there was great temporary improvement with loss of fever and relative health for some months, with disappearance of the tubercle bacillus from the sputvim. Four are still alive, 2 in a rather hopeless condition, but 2 with a possibility of ultimate recovery. One hundred cases are classified as suitable for pneumothorax treatment owing to the relatively good condition of the better lung, " though often attacked over nearly half its surface," but then with less active disease. Among these, followed up for periods varying ))etween If and years, the effects of pneumothorax treatment arc compared with that of sanatorium treat- (64). thorai 32 8 18 12 12 14 1 o 1 168 PNEUMOTHORAX TREATMENT ment among those (36 in number) in which adhesions prevented the production of a pneumothorax : Pneumothorax Korneumo- J . Fit for ordinary or lighter -vvork 2. Unable to work on account of tubercu- losis ....... 3. Dead from tuberculosis 4. Unknown ...... 5. Dead from acute complications Neglecting the last three cases tlic folio Aving percentagts are given : 1. Able to work 50% 222% 2. Unable to work 28-1% 33-3% 3. Dead from tuberculosis . . . 18-7% 38-8% 4. Freed from tubercle bacilli . . .50% 8"6% Among the treated cases 50 are alive, and in 30 of them the treatment is completed. Of these, 13 arc cured and 2 free of symptoms, but the time interval is still too short for safety. Five are partly able to work and 10 are still ill. In these last 15 cases treatment was suspended because disease was progressing in the functioning lung, and this sufficed to secvu'e its arrest in several cases. Saugman's mode of dealing with pneumothorax material is probably the best that can be devised, the groups chosen for contrast being very fairly comparable. Indeed the only obvious criticism which comes to mind is the likelihood that the more advanced and less favour- able cases might tend to fall into the second group where universal pleuritic adhesions prevented pneumothorax, and, unless the cases were all at closely similar stages of disease, such criticism must be admitted to be valid. Begtrup-Hansen (1913) follows a similar classification to that of Saugman, but refrains from presenting his material in any statistical form. He divides up his 100 cases as follows : In 31 no effective pneumothorax Avas possible. Among 25 of these followed up later, 10 died (40%), 9 got worse GENERAL RESULTS OF TREATMENT 169 (36%), 6 improved (24%), but only 8 lost their tubercle bacilli. In 69 an effective pneumothorax was produced ; 24 of these Avere really unsuitable owing to advanced disease of other organs or of the othe^ lung, and 23 of these died, though many showed a temporary improvement. Of the remaining 45 cases — all treated for over ll years, all with severe disease and tubercle bacilli in the sputum, and 38 '\\ith fever — 8 were completed and cured and 7 of these have remained so for periods of three months to five years, but 1 got a return of disease in the same lung one year later and had a thoracoplastic operation done, since the pleura was adherent. The duration of treatment in these 8 cases varied from eight months to four years. Fourteen other cases are ap- parently cured, giving a good result in 22 of the 45, or nearly 50%. A further 17 improved (37-7%), 6 got worse; of all, 10 are now dead (22 '2%). Of Zink's (1913) 110 cases, in 81 effective pneumo- thorax was achieved ; in 35 of these it was complete and in 45 partial, though in most of these adhesions were eventually reduced to long strings and nearty full collapse obtained. Of 67 cases, the remainder being too recent for in- clusion, 25 entirely lost their sputum, and in addition 29 became and remained entirely bacillus-free. Thirteen continued to void tubercle bacilli. Thirty-one became fully fit for work ; in 9 of these the lung has been long re-expanded ; 22 go about their business with the lung still collapsed. All these give promise of complete recover)'. In the 9 cured eases the duration of treatment was 6 months, 9 months, 12 months, 17 months, 20 months, 23 months, and 25 months in 2 cases. In 6 the cure is expected to be permanent, but in 3 the treatment was very short and the future un- certain, although they appear to be healed. Sixteen cases have got worse, in 5 of these because treatment was not maintained sufficiently long (I year, f year in two, 1 year in two), though the first effects were 170 PNEUMOTHORAX TREATMENT excellent. Of the remainder of these, in 5 disease spread in the other lung, in 3 in the (? partially) col- lapsed lung, and 1 died of tuberculous enteritis. The remaining 20 cases have all attained the stage of " clinical recovery " and expect to get well, though in one or two this result is less likely. L. Spengler (1913) discusses the results in 88 patients, of whom 23 (26%), with treatment completed for periods of i to 5 years, are apparently cui'cd ; a further 10 (ll-3%) have improved, but in 19 (21-6%), mostly with partial and ineffective pneumothorax, the results were negative. Thus he obtained a " positive " result in 69 of the 88 cases (78-33%) and a " negative " result in 19 cases (21-66%). Among Reliefs (1912) 40 cases, in 23 a complete pneumothorax was achieved and in the remaining 15 no efficient gas collection could be produced. Among these 15 he records a 60% mortality as against 20% (5 cases) among the 25 pneumothorax cases. His cases were operated on by Brauer's open method. Brauns notes that among 80 of his cases, 49 (61-2%) were healed, and 20 (23-7%) dead in spite of the pneumo- thorax. Twelve still remained under treatment. Carpi (1914) records the following results among 30 cases treated : Six cases completed for 1^ to 4 years were cured: all but one of these were bad cavity cases. Twelve have been treated for 1 to 2 j-ears : all in stage of " clinical healing " and 6 at work. Two cases treated some weeks with improvement. Three cases began well but treatment had to be suspended on account of the development of disease in the opposite lung or elsewhere. Thirteen cases died under treatment, 6 of intestinal tubercle which became worse under pneumo- thorax treatment, 2 of disease in the other lung. 2 of hnig perforation, 2 of empyema, and 1 of I uberculous nuiiitmitis. GENERAL RESULTS OF TREATMENT 171 Thvis in 55% a good, in 8 '3% a doubtful, and in 37% a negative resvilt Avas obtained. A. Pisani (1914) divides his cases into one-sided and double cases. One-Sided Disease. — Thirty-two cases of which 18 have been fully healed for four or five years, 6 have shown temporary improvement, and 8 have died ol' complications. Double-Sided Disease. — Seventy-two cases; in 33 of these a decided temporary improvement lasting in some a long and in others but a short time. In the remaining 39 great improvement occurred and symptoms vanished — but later disease progressed in the opposite lung. So much for the individual records of European workers. The American figures have been conveniently collected by Sachs, who is thus enabled to tabulate the results of pneumothorax treatment among a very large material. One thousand one hundred and forty- five cases from the records of 24 American observers are analysed, 75 of these having been under his own personal care. " It is apparent," he says, " from the large proportion of far advanced cases, 88 "7% (with bilateral involvement present in 77%, cavities in 62%, and serious tuberculous complications in 23%), that artificial pneumothorax is being tried at present in this country chiefly in very advanced cases unimprovable under ordinary sanatorium regimen. This must be taken into consideration in judging the ' immediate ' and ' subsequent ' results obtained in the 1,145 cases analysed in this paper." The " immediate " results are conveniently set forth in the following table : Quiescent . . . . 119 or 10-8%^ Arrested or apparently arrested 105 or 9-5%'r 21-7% Cured 16 or 1-4%J Improved .... 323 or :.9'~,% Failure . . ■ .162 or 14-6%| Unimproved . . . 203 or 18-3%U9-i% Dead ISO or 16-2%j 172 PyEUMOTHOKAX TREATMENT The favourable results, he points out, were mostly, that is in 29"2%, of a "palliative" character — in 21'7% they were " apparently durable." A comparison with the combined results of Zink, Brauer, and Spengler shows the European and American residts to be fairly closely in accord : American. Earopean. T'nimproved, failures, deatlis . 49-lOn 42-9% Improved .... 29-2% 31-2% Quiescent .... 10-8% 9-8% Arrested and cured . . 10'9% 16-1% Among the " apparently durable" results, 21"7% of the American figures, Sachs finds that not much over one- half stand the test of time, as shown by the figure, 12 '3%, obtained for cases remaining well after re-ex- pansion of the lung. These figures, be it remembered, cover the whole of a very mixed material and one in which, with bilateral involvement in 77 °o of cases, the indications for pneumo- thorax treatment appear to have been interpreted in a very liberal spirit. The results are also burdened with cases of " partial " pneumothorax, a condition which varies, according to its volume, from fair utility down to complete uselessness. The " immediate results " showed only 18'2% quiescent and arrested for " partial," as against 47% for " complete " pneumothorax. Again, among 162 " bilateral " cases there were only 24 '7% quiescent or arrested and 43% unimproved or dead, while among 91 "unilateral" cases 50 "5% were qui- escent or arrested and only 12 "^g with no improvement or death. These marked discrepancies show clearly how very different are the results under the best conditions of selection and achievement, from those attained in eases unsuitable from the beginning, or where the pro- duction of an efficient pneumothorax has demonstrably failed. In discussing the indications for treatment (p. 12) the writer expressed a hope that with ex- tended knowledge fewer and fewer cases would be allowed to drift into those stages Avhere pneumothorax GENERAL RESULTS OF TREATMENT 173 treatment is but doubtfully nidicated, and adhesions arc almost certain to limit its utility. To achieve a sure success the operation must be performed at the suitable moment ! Under such circumstances, exemplified to some extent in the selected figures presented above, pneumothorax treatment could lay claim to a success little short of marvellous when the nature of the material (see Indications and Contraindications) was fully taken into account. The striking success of the treatment in comparatively advanced cases, more especially the immediate success, impresses itself very forcibly on all those w^ho practise this treatment, and has led to a chorus of praise with never a dissentient voice. " All the physicians who have applied artificial pneumothorax on an adequate scale," says R. Burnand, " agree in assigning to it an eminent therapeutic value. We have failed to find such unanimity up to now, in the appreciation of any other antituberculous remedy, and this agreement alone is of a kind to give us encourage- ment." Burnand's own experience allows him to subscribe to this favourable opinion of a method which " is capable of bettering very decidedly, perhaps of curing patients seriously attacked, where no other system of treatment would have a chance of saving them." It is to the " lost " case of phthisis that pneumo- thorax treatment offers a chance of life and recovered health, and this, as one of its exponents has justly affirmed, is its " high aim." BIBLIOGRAPHY Begtrup-IIaxsi-.x, Th. (\>jlcfjorcl Saiiatoriimi, Denmark) : Pages 121, ]•_'.->, 130, 108: " Der kiinstliehc Pneumo- thorax in der Plithisisbehaiidlung." (CoiDcnhagcn, 1912 ; reviewed in Internat. Ccntralblott fur Tuherk-Fnrschmis,. 1913, vii. 362.) BuAUEU, L. (Hamburg) : Pages 3, 20, 30, 152, 156, 158, 159, 160 : " Die Behandlung der einseitigcn Lungenphthisis mit ktinstlichem Pneumothorax (nach Murphy)." {Aerzil. Verein z. Marburg Dec. 1905, Munch, med. Wochenschr., 1906, liii. 337.) " Wcitcre klinische und experimentelle Erfahrungen liber arterielle Luftembolie." {Verhandl. dcs Deutsch. Kongress. f. inn. JMediz., 1913, xxx. 347.) BiiAUXS, H. (Hanover) : Pages 52, 55, 121, 170: " Meinc Erfahrungen mit der Eorlaninischen Stichmethode in der kiinstlichen Pneumothoraxtherapic." {Zeitsch. f. Tuherkulose, 1912, xviii. 549.) Bruns, O. (Marburg) : Pages 8, 95, 104, 109, 110, 114, 117 : " Ucber die praktische Bedeutung der Zirkulationsanderung durch einseitigcn Lungenkollaps bei therapcutischen Eingriffen an der Lunge." (Beitriige z. Klin, der Tuberk., 1914, xxix. 253.) Carlstrom, G. (Golenburg, Sweden) : Page 117 : " Beitrag zur Frage der Wirkung dcs kiinstlichen Pneumothorax auf das Herz und die Zirkula- tion." {Beitriige z. Klin, der Tuberk., 1912, xxii. 243.) Carson, James (Liverpool) : Pages 1, 6 : Essays, Physiological and Practical. (Liverpool, F. B. Wright, 1822.) 175 176 BIBLIOGRAPHY Cayluy, \V. (London) : Page 2 : "A ease ol' htemoptysis treated by the induetion of pneumothorax so as to collapse the lung." {Clin. Soc. Trans., 1885, xviii. 278.) Cloetta, M. (Zurich) : Pages 114, 115 : " Beitriige zur Physiologic und Pathologic dcr Lungencirculation und deren Bcdeutung fiir die intrathorakale Chirurgie." {Archiv. f. Klinische Chirurgie, 1912, xeviii, 835.) D.vviES, MoRRisTox (London) : Pages 27, 32, 35 : " A note upon the complete removal of pleural effusions by the regulation of intrathoracic pressure during aspira- tion. (Oxygen replacement.") {Lancet, 1912, ii. 1774.) " Recent advances in the surgery of the lung and pleura." {Brit. J. of Surge nj, 1913," i. 228.) Davy, J. (London) : Page 32 : " Observations on air found in the pleura in a case of pneumato-thorax, with experiments in the absorption of different kinds of air introduced into the pleura." {Philosoph. Trans., 1823, cxiii. 496.) Dluski, K. (Zakopane Sanatorium) : " Beitrag zum ktinstlichen Pneumothorax." {Beitriige z. Klin, der Tuberk., 1914, xxxiii. 1.) Emerson, C. P. (Johns Hopkins University) : Page 48 : " Pneumothorax ; a historical, clinical and experimental study." {Johns Hopkins Hosp. Reports, 1903, ii. 1-450.) Fagiuoli, a. (Berlin) : Page 18 : " Versuche einer doppelseitigen Pneumo- thoraxbehandlung." {Zeitsch.f. Tuberk., 1914, xxii. 547.) Forlanini, C. (Pavia) : Pages 2, 3, 17, 20, 23, 27, 80, 55, 62, 74, 93, 95, 96, 98, 100, no, 115, 117, 122, 135, 143, 147, 151, 152, 155, 159, 162, 163 : " A contribuzione della terapia chirurgiea della tisi ablazione del polmone ? Pneumotorace artificiale ? " {Gazzetta degli Ospedali, agosto, settembre, ottobre, novem- bre, 1882.) " Primi tentativi di pneumotorace artificiale nella tisi polmonare." {Gazz. meet, di Torino, 1894, Ixv. 381, 401.) BIBLIOGRAPHY 177 FORLANINI, C. {V-AY\&)~contd. : "Apparate und Operationstechnik fiir den kiiiist- lichen Pneumothorax." {Deutsch. med. Woch., 1911, xxxvii. 2313.) " Die Behandlung der Lungenschwindsucht mit dem kiinstlichen Pneumothorax." {Ergebnisse der inn. Mediz. und Kinderheilk., 1912, ix. 621.) Graetz, F. (Marburg) : Pages 21, 109, 118: "Der Einfluss des kiinstlichen Pneumothorax auf die tuberkulose Lunge." {BeitrUge z. Kim. der Tuherk., 1908, x. 249.) Hamman, L., and Sloan, M. F. (Johns Hopkins University) : Page 52 : " Induced pneumothorax in the treat- ment of pulmonary disease." {Johns Hopkins Hosp. Bulletin, 1913, xxiv. 53.) Houghton, James (London) : Page 1 : " Account of a remarkable case of pneumo- thorax." (Dublin J. of Med. and Cheni. Science, 1832, i. 313.) Jacot, M. (Leysin) : Pages 124, 125, 126, 127, 162, 163 : " La pleuresie purulente complication du pneumothorax artificiel." (Revue rned. de la Suisse Romande, 1915, xxxv. 117, 185, 241.) KiJss, M. G. (Paris) : Pages 8, 27, 35 : " La technique et les resultats immediats du pneumothorax artificiel dans les formes avancees unilaterales de tuberculose pulmonaire." (Bull et Mem. Soc. Med. des Hop. de Paris, 1910, xxx. 3f<= serie, 88.) Mayer, A. (Berlin) : Pages 106, 121, 122, 127, 128, 129, 130, 132, 140, 141, 149, 158 : " Experimen telle und klinische Mitteilungen liber die nach Pneumothoraxoperationen auftretenden Pleuraergiisse." (Beitr'dge z. Klin, der Tuberk., 1914, xxix. 51.) V. MuRALT, L. (Davos) : Pages 120, 121, 122, 124, 130, 132, 135, 142, 144: " Erfahrungen tiber Exsudate bei kiinstlichem Pneumo- thorax." (VII Supplement-Band der Beitrage z. Klin, der Tuberk., 1914, 356.) 12 178 BIBLIOGRAPHY MuHPHY, J. B. (Chicago) : Pages 3, 35 : " Surgery of the lung." (Journ. of the American Med. Assoc, 1898, xxxi. 151, 208, 281.) Parker, R. W. (London) : Page 145 : " Suggestions for the treatment of special cases of empyema by thoracentesis and the simultaneous injection of purified air." {Lancet, 1882, i. 689.) Piper, H. (Berlin) : Page 157 : " Uebcr die Beziehungen zwischcn venosen Blutdruck und intrathorakalem Druck." {Arch, f. Anat. und Physiol, Phys. Abt. 1913, s. 385.) PoTAix (Paris) : Pages 2, 145: " Des injections intrapleurales d'air sterilise dans le traitement des epanchements pleuraux consecutifs au pneumothorax." {Bull, de VAcademie de Med. 1888, S""' serie, xix. 537.) Powell, K. Douglas (London) : Page 114 : " Notes on the Pneumothorax occurring in Phthisis." {Med. Times and Gaz., 1869, i. 112, 166, 194.) llisT, E. (Paris) : Page 108 : " Critical Review, Artificial Pneumo- thorax." {Quarterly J. of Med., 1913, vi. 259.) RiST, E., and Maikgot (Paris) : Page 107 : " Examen radiologique ct pneumothorax therapeutique." {Bull, et Mem. Soc. Med. des Hop. de Paris, 1912, xxxiv, 3""= serie, 377.) Robinson, S., and Floyd, C. (Boston) : Pages 3, 47 : " Artificial pneumothorax as a ti-eat- ment of pulmonary tuberculosis." {Arch, of Int. Med., 1912, ix. 452.) Sachs, T. (Chicago) : Pages 52, 153, 171 : " Artificial pneumothorax in the treatment of pulmonary tuberculosis ; results obtained by 24 American observers." {J. of Amer. Med. Assoc, 1915, Ixv. 1861.) BIBLIOGRAPHY 179 Saugman, Che. (Vejlefjord Sanatorium, Denmark) : Pages 20, 27, 28, 29, 30, 43, 46, 48, 52, 62, 77, 80, 95, 96, 97, 117, 120, 121, 129, 141, 152, 155, 158,159,162,167: " On the results of the pneumothorax treatment of phthisis." {Trans, of 17th Internal. Congress of Med. London, 1913, Med. Sectn., vi. 463.) " Zur Technik des kunstlichen Pneumothorax." (Beitrage z. Klin, der Tuberk., 1914, xxxi. 571.) Spenglee, L. (Davos) : Pages 17, 144, 164, 170: "Der Ablauf der Lungen- tuberkulose unter dem Einflusse des kiinstliehen Pneumothorax." {Correspondenz-Blatt. f. Schweizer Arzte, 1909, xxxix. 801.) Spengler, L. (Davos), and Sauerbruch (Zurich) : Pages 123, 124, 142, 143: "Die ehirurgische Behand- lung der tuberkulosen Pleura exudate." (Munch, med. Woch., 1913, ix. 2825.) Straub, H., and Otten, M. (Tubingen) : Page 14 : " Einseitige vom hilus ausgehende Lungen- tuberkulose." {BeitrUge z. Klin, der Tuberk., 1912, xxiv. 283.) Webb, G. B., Gilbert, G. B., James, T. L., and Havens, L.C. (Colorado) : Page 32 : " Artificial pneumothorax with report of gas analyses and experiments to determine the use of air or nitrogen." (Arch, of Int. Med., 1914, xiv. 883.) Weiss, A. (Gorbersdorf) : Pages 17, 20, 23, 52, 79, 117, 138, 147, 149 : " Ueber Komplicationen bei der JBehandlung mit kiinstlichem Pneumothorax." (Beitrage z. Klin, der Tuberk., 1912, xxiv. 333.) Wilms (Heidelberg) : Page 62 : " Die Pfeilerresektion der Rippen zur Verengerung des Thorax bei Lungentuberkulose." (Die Therapie der Gegenwart, 1913, liv. 17.) Zesas, D. G. : Page 135 : " Zur Frage der pleurogenen Reflexe." (Zentralbl. f Chirurg., 1914, xli. 371.) 180 BIBLIOGRAPHY ZiNK (Davos) : Pages 17, 23, 47, 52, 98, 123, 156, 169 : " 110 Falle von kiinstlichem Pneumothorax und die daran gemachten Beobachtungen." (Beitrdge z. Klin, der Tuberk., 1913, xxviii. 155.) INDEX Acute forms of tubercle, their suitability for treatment Adherent pleura (see Pleura) Adhesions (see Pleural adhesions) Altitude, effect of, on pneumothorax Anaesthetics and anodjmes ..... 42, 43. Antiseptic : solutions for gas bottles preparation of skin Apical phthisis : its suitability for treatment early phthisis a disqualification Apparatus : general features . . ' . its preparation for use . ' . Pearson and LilUngston's . de Carle Woodcock and Lister's . Kiiss' filling the gas bottle . Asthma as a contraindication .... Autoinoculations after fillings Autotoxaemia, its removal by pneumothorax . ijacteriology of pleural effusions Bed at beginning of treatment Bilateral pneumothorax .... ,, tuberculosis, as a contraindication Blood pressure in artificial pneumothorax Bottles, "gas" and "pressure" . advantage of movable fixed .... Chemistry of pleural effusions Collapse, complete Complete pneumothorax Contraindications to treatment special Cough, effect of pneumothorax on Cytology of pleural effusions . Diabetes, as a contraindication Diaphragm, effect of pneumothorax on paradoxical movements of 181 PAOE 21,21 80 44,65,154 24,37 44 16 19 24 25,37 26 27 27 37,38 22 88, 80, 103 8, 9 129 87 18 172 118 24 28, 147 27, 147 17, . 127 71 56 11 22,23 90, 103 . 129 23 . 106 . 107 182 INDEX Dyspepsia, at onset of pleurisy- Dyspnoea of toxic origin liiffect of lung collapse : mechanical serological Embolism (see Gas embolism) Emphysema, as a contraindication contributing to heart failure surgical Expectoration, effect of pneumothorax on h ever, effect of pneumothorax on as symptom of plevirisy showing autoinoculation PAGE . 136 22, 90 . 119 . 161 92, 169 89, 90, 91 liG, 135 88, 89, 148 (jas : at first operation at refills ..... comparison of air and nitrogen . putting in, at initial operation at refills amount of, at first operation at refills . rate of absorption of . volume and nature in gas embolism Gas embolism ..... mechanism of intravenous pressure as factor in source of . . . volume and nature of gas in symptoms prophylaxis treatment Gas pressures (see also Manometric readings) at first operation at refills .... in complete pneumothorax in partial pneimiothorax . effect of altitude on optimum pressure . excessive pressures . influence of pleurisy on . Gas replacement .... instruments needed the operation with a full pleural cavity Hasmoptysis, as an indication for treatment Heart and circulation .... heart disease as contraindication . displacement 31 31 33 45 70 46 71 7.->, 76 159 156 156 157 158 159 160 160 161 46, 66 67,71 77 78 80 103 113 137 145 145 145 149 21 116 22 117 78 INDEX 18i Heart murmms .... embarrassment enlargement Hiluni tuberculo.sis History of pneumothorax treatment Indications for treatment Intervals between fillings, at early stage at later stages in partial pneumothorax Intestinal tuberculosis a contraindication Intrapleural pressure (see Gas pressure) Intravenous pressure and gas embolism . Ividney disease as a contraindication r.uiE 117 117, 118 118 13, 15 1, 2, 3 11 ()4 74 70 23 170 107 -L/aryngeal tuberculosis anti pneumothorax Lung circulation . Lung, perforation of symptoms . prognosis . prophylaxis treatment . Lung, The compressed . re-expansion of . forcible re-expansion of fibrosis of . unexpanded blood supply Lung, The functioning or bettor effects of treatment on disease in . blood supply Manometers, water most suitable gauge mercury air pressure safety Manometric readings . absence of manometric response incomplete manometric response absence of respiratory fluctuations effect of pleural pockets on abdominal puncture . at refills .... respiratory fluctuations significance of various types of effects of pleurisy on . U.3, U '.)5, 22, 23 114 127, l(i2 103 163 164 164 108 0,98,90,111 133, 142 96, 108, 109 100 115 17 HI 17, 111, 112 17 112 111 110 o 5,28 29 30 30 29 4 J, 47, 60 49 50 50 5U, 51 51 60 2 9, 69 70 130 137 , 138 184 INDEX Mediastinum ..... effect of open pneumothorax on effect of closed pneumothorax on hernial protrusions of . excessive displacement of ilixed infections, risk of . PAGE . 104 . 104 . 105 . 106 112, 113 93 Needles, for initial operation. writer's trocar and cannula for refills the Lillingston the Saugman . sterilisation and drying . tiiidema of functioning lung . Operation, Initial (see also Refills) technique of . . . sites for puncture preparation of patient of instruments, etc. position of patient passing the needle entrance of gas removing the needle difficulties and complications Jrain, a symptom of pleurisy Paravertebral triangles and quadrangle . Partial pneumothorax, ineffectual . effective Perforation of lung (see Lung) Peribronchial tuberculosis Physical signs, general after initial filling . of complete coUapse in re-expansion of lung . in pleurisy Phthisis (see Apical phthisis) Pleura, adherent, its diagnosis its incidence its reduction stripping of surfaces . Pleural adhesions — causing lung perforaton Pleural fistula ..... Pleural pockets (see also Partial pneumothor Pleural shock or reflex symptoms conditions of appearance prophylaxis treatment 34 34, 45, 149 35 36 36 43, 44 102, 112, 113 39 40 42 42 43 43 45 47 50. 51 48, 49, ax) 134 S5, 139 53 55 13 S2 48 73 99 138, 139 41, 60 61 79 163 149 50, 53 151 151 152 154 155 INDEX 185 Pleurisy, its effect on gas absorption prevalence causation dry pleurisy . pleurisy with effusion classification of effusions purulent effusions . chemistry of effusions cytology of effusions bacteriology of effusions . effects of pleurisy, beneficial unfavourable symptoms of . effects on gas pressure physical signs .... treatment of . chronic effusions . acute serous effusions . acute purulent effusions Pressures (see Gas pressures) I'ulmonary relaxation (see also Mediastinum) ixadjography (see X rays) Refills : operation technique puncture spot . manometer readings putting in gas . early fillings complete collapse intervals . intrapleural pressure: Replacement (see Gas replacement) Results of pneumothorax treatment oafety syringe of Forlanini .... Succussion splash ...... Suitable cases for treatment (see Indications) their number .... Symptoms, showing complete collapse . disappearance of — under pneumothora.x of pleurisy ..... of pleural shock .... of gas embolism .... of lung perforation i emperature (see Fever) Termination of treatment .... Tidal percussion in diagnosis of adherent pleura Toxaemia (see AutotoxEemia) Ti'eatment of pleural effusions of pleural shock . 13 PAOI? 76 . 120 121 . 123 123 123, 125, 130 . 125 . 127 . 120 . 129 . 131 . 133 . 134 137,138 138, 13n, 140 . 140 . 141 . 142 143 112, 113 01 04 OG 70 71 71 74 77 16G 36 150 139 11 73 88 134 151 160 163 97 40, 61 , 99 125, 127 155 186 INDEX Treatment of gas embolism .... of lung perforation general results of pneumothorax . Tubercle bacilli, disappearance under treatment Tubing, rubber, for pneumothorax apparatus as source of gas embolism PAGE . 161 . 164 . 166 92, 168, 169 33, 38 . 158 Unilateral tuberculosis its identity with lulum tuberculosis its incidence 12 13 14 Weight, effect of pneumothorax on 92 ^-rays, in unilateral tuberculosis . value in initial stages of treatment in diagnosis of adherent pleura . complete collapse in pleviral effusions . 13 39 61 . 72 139, 140 11: -;«' rmi' A '\s¥"'^;^:' »