33

RESEOTIONAL SURGERY IN PULMONARY TUBEROULOSlS*

By J. V. CUSSEN,M.B., B.Ch. ' N T I L the advent of streptomycin, the treatment of pulmonary tuberculosis may be said to have consisted of rest and collapse therapy. Over the years many other treatments had been at various times advanced. Although some of these evoked much hope, few withstood the light of experience and keen observation. Some of us will recall the interest which was aroused by the introduction of gold treatment. Many apparently excellent results were reported. The sanatorium doctor spent much of his time injecting into the veins of his patients varying quantities of gold. Patients imbued with new hope accepted a stricter rest regime. The radiological findings often indicated remarkable improvement. Final analysis, however, left little doubt that the apparently good results in certain cases were attributable to the stricter rest r~gime rather than to the gold. In the immediate pre-antibiotic period, the accepted classical treatment of pulmonary tubreulosis had, however, reached a high degree of perfection. The improvement in the quality of pulmonary radiographs had allowed of more accurate and precise diagnosis. Serial x-ray films and fluoroscopy permitted close observation of the course of the disease. Fluoroscopy facilitated the rational maintenance of temporary collapse treatment. The further development of the method of Jacobeus by Maurer and Michetti made it possible to attain the technically perfect artificial pneumothorax. The introduction in 1938 by Monaldi of the method of direct drainage of tuberculous lung cavities was a further notable advance. Many cases were and still are being rendered operable by this method of suction drainage. Phrenic paralysis was much employed for a considerable period. Whatever results were obtained, and they were few, many unfortunate consequences of such treatment are still observed. Pneumoperitoneum, with or without phrenic paralysis, although still employed as a curative measure, is undoubtedly losing ground. Plombage, at one time, knew much favour. The frequent intolerance of the foreign body employed led to many disastrous results and to the general abandonment of the method. Plombage with improved techniques is still under trial. Surgical collapse methods had undergone much development througll the introduction of improved techniques, all aiming at the more economic selective collapse. Extrapleural pneumothorax was re-introduced with improved technique by Graf and Sehmidt in I936. Of all the methods of surgical collapse, this gives the best collapse and offers the best possibility of functional recuperation after cure. Thoracoplasty, first executed by Sauerbruch in 1911, had undergone much d~velopment. The ll-rib paravertebral operation had been super-

U

*Communication made to the Irish Tuberculosis Society a t its meeting in Galway, March, 1955.

34

I R I S H JOURNAL OF MEDICAL SCIENCE

seded by the more economical selective collapse achieved by the techniques of Semb and others. The sanatorium doctor thus could offer, to a very limited number of his patients, the hope of some real lasting benefit from the limited therapeutic armament at his disposal. Unfortunately, to many he could offer little beyond sympathy and encouragement. Streptomycin was the first definite advance in the drug treatment of tuberculosis. Dramatic results were soon observed. Unfortunately the development of bacillary resistance was a disappointing and discouraging element in relation to this new treatment. The introduction of paraamino-salicylic acid and later of isonicotinic acid largely neutralised the difficulty in regard to resistance, which, it was demonstrated, was retarded by combining one or other or both of these drugs with streptomycin. The introduction of antibiotics revolutionised the treatment of pulmonary tuberculosis. The striking results obtained permitted the development of lung resection, whilst bringing within the range of collapse treatment a high percentage of hitherto unsuitable indications. Of all the developments in the treatment of pulmonary tuberculosis, that of lung resection must be considered as possibly the.most interesting. Not alone has it allowed of the treatment of previously'untreatable cases, but by permitting the examination of resected lungs and lobes it has also contributed to more accurate diagnosis and to the better interpretation of pulmonary radiograms and tomograms. In 1891, the resection of lung tissue for tuberculosis was first carried out by Tufter. Fr, om 1934 onwards an ever-increasing number of these operations was reported by Alexander, Churchill, Chamberlain, Sellors, Price Thomas, Overholt and others. With the present development of anaesthetic and surgical techniques, lung resection has become an integral part of the treatment of pulmonary tuberculosis and has superseded certain previous methods. The generally accepted indications for resection procedures in pulmonary tuberculosis include the failure of collapse therapy, certain dense loci or solid lesions, destroyed lung, certain types and localisations of lung cavities, bronchial stenosis and bronchiectasis, certain emergency indications and pleural complications. The cavity which has failed to close under thoracoplasty affords an excellent indication for resection. Although such cases may offer certain technical difficulties, the postoperative course is usually very satisfactory and the end-results strikingly good. The best approach to some cases requiring resection treatment is by way of preliminary thoracoplasty. The inert cavity under the technically satisfactory endo- or extrapleural pneumothorax is suitable for resection and the technical aspect is often simpler. The failure of other methods of collapse therapy such as plombage may also be adequately treated by resection. The destroyed lung, which is the seat of multiple cavities involving all lobes, often associated with important caseous and fibrotic elements, renders pneumonectomy imperative. It is, nevertheless, frequently the case that appropriate preliminary treatment comlJined with chemotherapy and antibiotics will allow of such clearance of the disease as to permit of a less drastic procedure than pneumonectomy.

SURGERY IN PULMONARY TUBERCULOSIS

35

The giant cavity is a p r i n c i p a t ~ a t i o n for excisional surgery. Where removal of the lesion is not feasible, eventual collapse therapy should be preceded by Monaldi drainage treatment. The "bullous " cavity may require resection. This thin-walled round or oval cavity was not a frequent finding before the advent of antibiotics and chemotherapy. It has to be distinguished from the clean hard-walled cavity of reduced dimension which had been commonly observed prior to the introduction of chemotherapy. The bullous cavity is distinguished by the thinness of its wall, the generally important often increased dimension, and the apparent absence of pericavernous infiltration. The manometric readings in true bullous cavities have been found to be atmospheric, indicating that no valvular mechanism was involved. This cavity will frequently appear to be relatively isolated in healthy lung tissue and to be remarkably stable, I t may be that further observation will show that surgical treatment is not essential. It has been our attitude to advise resection for these cavities, where feasible. Nevertheless, in a number of instances in which radical treatment was not possible, these bullous cavities showed little variation when observed over long periods. The lower lobe cavity previously constituted a frequently insoluble therapeutic problem. It now represents one of the most suitable indications for reseetional surgery. Resection procedures in regard to the lower lobe generally give very satisfactory results. Where a choice of precedure is possible, and this is not always the case, with the advanced lesions admitted to our sanatorium I would generally prefer resection to surgical collapse. Cavities of moderate to small dimension of the postero-apical zone of the upper lobe are suitable for limited thoracoplasty. Cavities of the anterior zone of the upper lobe are not good indications for surgical collapse and should preferably be resected. Irrespective of the localisation, resection is preferred for lesions exceeding 3-4 cms. in diameter. Dense lesions of the tuberculoma type, circumscribed caseous loci, or inspissated cavities may call for resection procedures. Canetti (Madrid Conference) stated that some round caseous loci are often favourably influenced by antibiotics and chemotherapy, whereas large loci of the" tuberculoma type nearly always fail to respond. Parenchymatous disease associated with bronchial stenosis should not be treated by collapse therapy, medical or surgical. Stenosis by impeding bronchial drainage causes purulent retention, usually characterised by febrile incidents and increasing toxic effects. Ectatic elements may develop, or deteriorate if already formed. Collapse treatment is obviously illogical in such conditions. Bronchiectasis, not associated with bronchial stenosis, is a not uncommon sequela of pulmonary tuberculosis and may require resection surgery. There are divergent schools of thought in regard to the question of resection treatment of certain minimal lesions such as Small isolated dense loci. Torning, (Madrid Conference), cited the study, made by" Duroux and Janiou, of 170 patients with round loci. This study, carried on over a period of one to nineteen years, showed that in only 4 per cent. of loci of less than 2 cms. in diameter was there progression, ~here~as

36

I R I S H JOURNAL OF MEDICAL SCIENCE

deterioration was noted in 27 per cent. of foei of 3-4 cms. and 70 per cent. of loci of more than 6 cms. Small round loci observed to be stable and associated with negative clinical and laboratory findings do not require resection. Conversely, such loci, giving rise to positive laboratory findings, and not responding to conservative treatment, are indications for resectional surgery. The pleural complications of pulmonary tuberculosis have possibly demonstrated, more than any other form of the disease, the striking coY~tribution of resectional surgery. Earlier methods of treatment" had been frequently ineffective and discouraging not alone to the patient, but: also to the sanatorium doctor. Nevertheless, good results were frequently obtained, prior to the introduction of resection surgery, with aspiration or drainage combined with adequate lavage, but only in the absence of broncho-pleural fistulae. Antibiotics and chemotherapy have improved the results of censervative treatment, and have in some cases brought about the closure of fistulae. Radical treatment of pleural complications should not be resorted to without a reasonable trial of conservative therapy, especially in view of the frequently disappointing functional results. The empyema which has failed to respond to conservative treatment calls for radical surgery where feasible. The persistence of the suppuration is generally related to broncho-pleural fistula, although its presence may not be demonstrable. Lobectomy and decortication or pleuropneumonectomy may be the desirable procedure according to the distribution of the parenchymatous disease. Certain emergencies provide indications for resection. Rupture of a lung cavity in the course of pneumolysis, whether endoor extra-pleural, may call for resection. A similar emergency might arise in regard to thoracoplasty. Lung haemorrhage of a grave form, or of intractable type, may under certain conditions justify lung resection. It should, however, be resorted to only for the gravest reasons, as in the presence of copious haemoptysis it is evident that the possibility of dissemination in the eontralateral lung is very great. Amongst the contra-indications may be cited active contra-lateral disease and inadequate respiratory function. Nor should a resection be carried out in the presence of active lesions in the opposing lung. Stabilised regressive disease in the contra-lateral lung does not exclude resection. Such disease may, in fact, be favourably influenced by the appropriate resection procedure in regard to the other lung. Inadequate respiratory function is an absolute contra-indication. The more frequently observed elements of diminished respiratory function are emphysema, pleural sequelae, previous extensive irreversible collapse procedures, medical or surgical, diaphragmatic paralysis and cardiac insufficiency. The decision to proceed with resection and the time at which the operation should be carried out must be based on the most careful Study of the case. The aim in the treatment of every sufferer from pulmonary tuberculosis must be to attain the maximum result with the minimum risk or loss of function.

SURGERY IN PULMONARY TUBERCULOSIS

37

Where conservative treatment has effected improvement and offers a reasonable possibility of a final satisfactory result, it is not justifiable to submit the patient to a major surgical procedure. ,Even in the most favourable conditions, the functional loss with resection may be important. Eerland, analysing the results of his first 300 segmental resections, found that the functional loss was disproportionate to the amount of lung tissue removed in 22 per cent. of the cases. Contrary to what is often asserted, conservative treatment and non-surgical collapse measures can lead to excellent, complete and durable results, even in regard to important ulcerative lesions. The apparent failures of conservative treatment are often related to inadequate investigation of the degree of healing. Examination of apparently good results calls for iomograms of high technical quality. Investigation of the genuine closure of cavities is greatly facilitated by tomography while the cavity is still clearly visible on the synthetic radiograph. By repeating at appropriate intervals the significant tomogram, the closure of the cavity may be accurately followed. Canetti, in his recent address to the International Union against Tuberculosis, showed that such cavity closure was satisfactory and that surgical procedures were not essential. Artificial pneumothorax, when the correct indication is present, must be considered before major surgery is decided on. The technically perfect endo-pleural collapse will, where the indication is correct, give complete and durable results with minimal risk and loss of function. Cavities of the lower lobe, notably of its apical segment, may respond to pneumothorax treatment. This was frequently observed when there was no better alternative to pneumothorax. Where conservative treatment has effected maximal, but incomplete, benefit, or in those cases in which resection is imperatively indicated, operation should not be deferred, and greas care should be exercised in selecting the most favourable moment for the procedure. Conservative treatment, unless it offers a reasonable hope of a satisfactory result within a reasonable time limit, should only be pursued to the point where it has realised the most favourable indication for surgery. Chemotherapy and antibiotics should from the beginning of treatment be employed prudently so as to conserve their value in relation to eventual radical surgery. It is true that this is not always possible. Many, if not most, of our patients have presented such advanced disease on admission that there has been no alternative but to administer large doses of antibiotics in order to render them suitable for any form of surgical treatment. The choice of the optimal time at which to intervene may be a question of considerable difficulty. Ill-timed surgery, too early or too late, especially with possible prior abuse of antibiotics, may be followed by unsatisfactory results. Partial resection, in the form of lobectomy, segmental or wedge resection, is the desirable procedure. All the published statistics indicate that pneumonectomy is attended by a significantly higher mortality rate. The value of partial resection procedures cannot be over-emphasised in relation to the filling of the residual thoracic space, thereby eliminating certain of the most smdous complications of resection surgery. There is also the functional aspect, although results in this respect are often disappointing. Pneumonectomy should only be undertaken where a lesser

38

I R I S H JOURNAL OF MEDICAL SCIENCE

procedure is excluded, The mortality rate is higher, and the problem of the residual space is more delicate. The choice of patient should be based on the anamnesis, clinical examination, radiological investigation, examination of the bronchial tree and determination of the respiratory function. The ease history is of primary importance. The time of onset of the disease should be accurately ascertained, where possible. Patients with old-standing disease are less suitable for resection surgery. Careful elicitation of the history will often suggest previous disease or pleural involvement of the contralateral lung. Although the disease may have satisfactorily cleared, the history of such incidents must be of significance in regard to respiratory function or possible postoperative evolution. The previous treatment undergone by the patient must be carefully analysed. The dosage and manner of previous administration of antibiv~ics may be of considerable importance in regard to any proposed resection procedure. The age factor should be considered. Although each case should be studied on its merits, special thought should be given to the age question, when patients have exceeded 40 years. It is, of course, true that resection surgery is frequently carried out on older patients suffering from carcinoma. This analogy, however, may be dangerous. The patient undergoing resection surgery for carcinoma, a condition of relatively rapid development, will frequently have had no previous lung disease. The tuberculous patient aged 40-50 years will almost certainly present some impairment of lung elasticity and function. Clinical examination based on close observation over a sufficient period is essential to the proper assessment of the suitability of the patient for resection surgery. The value of careful observation cannot be overemphasised. The most careful investigation to reveal extra-pulmonary disease or concomitant disorders is called for. Extra-pulmonary tuberculosis does not necessarily exclude lung resection. Amyloid disease of a mild degree, with no significant modification of hepatic, renal and suprarenal function is not a contra-indication. Diabetes does not exclude resection. In certain cases improvement ~ of the diabetic condition has been observed to follow removal of the tuberculous focus. Radiological examination should include careful study of all available radiograms taken since the diagnosis of the disease. This study is especially important in relation to decortication procedures. In effect, the nature and degree of previous disease may render a partial lung resection an imperative supplementary procedure to decortication. Radiological investigation calls for frequent fluoroscopy, and synthetic radiographs and tomograms, both postero-anterior and profile. Tomograms of good technical quality are essential to the correct interpretation and loealisation of pulmonary disease. They should include both lung fields. Only by tomography can the absence of cavitation in contralateral lungs or ipsolateral lobes be confirmed. Limited resection procedures are often rendered possible by the partial clearance of disease under the influence of rest and drug treatment. When a patient first comes under observation, tomograms should be taken in regard in regard to cavities visible on the synthetic radiograph. This allows of cavities being

SURGERY IN PULMONARY TUBERCULOSIS

39

accurately followed to closure by regular repetition of the significant cut. If tomograms are not carried out prior to the apparent healing of the cavity localisation of any small residual element may be difficult and uncertain. Lateral tomograms may be of great assistance in the localisation of the disease in regard to limited resection procedures. Accurate pre-operative localisation of the disease will allow of a well defined plan of operation. During the course of the surgical intervention this plan should be adhered to, and it should be altered only on imperative indication revealed by the thoracotomy. Bronchoscopy has a major role, not alone in regard to the post~ operative treatment, but also in regard to diagnosis and verification of the projected site of bronchial section. Endobronchial disease with ulceration at the level of the projected site of bronchial section must exclude resection until satisfactory healing of the bronchial lesion has been obtained. The presence of stenosis, confirmed by bronchoseopy, must strengthen the indication for resection as opposed to collapse treatment. Bronchography is a valuable, indeed often essential, procedure in regard to the accurate assessment of surgical interventions on the lung. Negative radiological findings do not exclude bronchial disease. Bronchoscopy allows of examination only to a limited extent beyond the main bronchi. Whilst recognising the value of bronehography, it should only be employed where essential in the investigation of tuberculous lung disease. Bronchography can lead to dissemination of the disease. Assessment of the respiratory function may be of some difficulty. In the majority of cases fluoroscopy and clinical examination will suffice. Fluoroscopy is of prime importance. The freedom of the eostophrenic angles, as also the diaphragmatic excursion on inspiration and expiration, is of the greatest significance. The degree of costal movement and mediastinal displacement should also be carefully observed. ]n doubtful cases broncho-spirometry and measurement of the pressure in the pulmonary artery may be called for. Where pulmonary function appears to be just sufficient, or just inadequate, full functional tests are indicated. The results of such tests should be interpreted in relation to the other findings. Where clinical and fluoroscopic examination appear to exclude resection, satisfactory functional tests should not take precedence or be opposed to them. In the assessment of eases of doubtful respiratory function, consideration should be given to the possibi!ity of a more important resection becoming essential for technical reasons during the course of the operation. ,Yudicious rest and antibiotic treatment are an essential preliminary to resection surgery. Such treatment, correctly carried out, w~ll improve the prognosis and may reduce the amount of any tissue that requires to be resected. Postural treatment may also be of value in the preparation of the patient for operation. Treatment prior to operation should be pursued with the greatest prudence so as to avoid the development of bacillary resistance and thus allow of the resection being carried out under cover of antibiotics and chemotherapy. The technique employed in resection surgery is not within the province of this paper. I would, however, refer to the position of the patient during operation. In my opinion, the prone position introduced By

40

I R I S H JOURNAL OF MEDICAL SCIENCE

Overholt is the most desirable. The drainage and aspiration of bronchial secretion is facilitated by this position and the risk of bronchogenic dissemination diminished. It also facilitates the technique of the quick approach to, the bronchus which can be dissected with minimal delay. Study of the arterial blood has shown that acidosis is less marked with the prone position. The major consideration in the postoperative treatment of patients submitted to a less extensive procedure than pneumonectomy should be to obtain the re-expansion of the remaining lung tissue with the least possible delay. The obliteration of the residual space by the re-expanded remaining lung is the most potent factor in the prevention of haemorrhage and suppuration. The postoperative care of patients subjected to pneumonectomy is less delicate, but the complications inherent to the residual space may be more formidable. Although the immediate postoperative care calls for adequate oxygenation, oxygen administration should be interrupted as soon as feasible, as it may lead to a certain degree of paresis of the pulmonary parenchyma, with resultant unsatisfactory re-expansion and retention of bronchial secretions. This in turn may favour atelectasis. Although resection surgery is generally well tolerated by the tuberculosis patient, operated on according to a well founded and reasonable indication, the incidence of complications is not negligible. They may be referred to under three groups: (1) operative, (2) immediate postoperative, (3) late postoperative. Injuries to the major bloodvessels constitute one of the l~rineipal operative complications. Although the haemorrhage may be adequately controlled, the injury may have serious consequences. It may, in effect, become necessary to carry out a more extensive resection thml had been originally planned. The possibility of such a complication should always be borne in mind in the pre-operative assessment of cases of limited respiratory function. During operation inundation of the contralateral bronchial tree may occur. This complication is undoubtedly less frequent with the patient in the prone position. The technique of the quick approach to the bronchus is also favoured, especially in regard to cases with much bronchial secretion. The more important immediate postoperative complications include shock, haemorrhage, atelectasis, oedema of the upper respiratory tract. Adequate blood replacement should be ensured both during and after operation. Occasionally grave haemorrhage calling for immediate reoperation may occur. In the case of pneumonectomy or the persistence of an important space in lesser resection procedures, clot formation in a haemorrhagic effusion may lead to serious signs of mediastinal pressure. In such circumstances, surgical evacuation of the clot should be resorted to. This is a relatively frequent complication of extrapleural pneumothorax, and the results of surgical clearance of the space are dramatic. In a personal observation of this complication in relation to pneumonectomy, surgical intervention immediately relieved a serious condition when the use of Streptokinase had failed to render aspiration feasible. Atelectasis is one of the principle complications of lung resection. It may be caused by accumulation of bronchial secretion, bronchial kinking or oedema due to trauma. Atelectasis should be carefully watched for

SURGERY IN PULMONARY TUBERCULOSIS

41

in the postoperative period. Frequent radiograms should be taken. Should it develop, it should be energetically treated by bronchial aspiration. The post-resection patient should be placed in a bed so constructed as to facilitate bronchoscopy. Tutoring of the patient in regard to eoughing, expectorating and appropriate positioning will contribute to the prevention of this complication. Phrenic paralysis should be avoided in resection procedures. The conservation of diaphragmatic movement is important to the elimination of bronchial secretion, as indeed also to the maintenance of maximum respiratory function. Oedema of the upper respiratory tract is more liable to occur in young subjects and may call for tracheotomy. Of the later complications, bronchial fistula is the most formidable. With the improvement in technique, the early or immediate postoperative bronchial fistula is very rare. When a fistula develops, it often occurs between the tenth and fifteenth days after operation, but it may make its appearance at a much later stage. Its advent is often insidious. It may be suspected from a rise in temperature, blood-stained sputum, or the purulent transformation of the fluid in the post-resection space. It may be necessary to carry out bronchoseopy, or inject a colouring agent into the residual space to obtain confirmation. The prognosis of bronchial fistula is related to the importance of the residual space. It will be grave in regard to pneumonectomy, whereas it may be good in relation to small post-lobectomy cavities. The treatment of bronchial fistulae is very difficult. Cases have been reported of successful re-suturing of the bronchus. In one case of pneumonectomy, personally observed, immediate surgical drainage followed by a 9-rib thoracoplasty gave a satisfactory result. Persistent fistulae complicating partial resection procedures may respond to aspiration and drainage or require thoracoplasty. In certain circumstances it may be necessary to resort to pneumonectomy. In a personal observation, a fistula of the right lower lobe bronchus was successfully dealt with by re-suture through a large injected space six months after lobeetomy. For empyema unassociated with bronchial fistula, aspiration or closed drainage with adequate lavage may lead to a satisfactory result. Infection of the thoracic wall is a complication of grave import, as a cutaneous fistula generally develops. The incidence of this complication was reported as high as 5--15 per cent of cases in earlier statistics. More accurate assessment of indications and improved technique, with more efficacious drug combinations, have almost eliminated parietal infection. The management of the residual space after lung resection is of vital importance. Lobectomy should generally be followed by appropriate corrective thoracoplasty. Over-distension of the remaining lung should not be allowed to develop. In lower lobectomy, phrenie paralysis may be employed, but except in certain special cases should be avoided. Segmental resections rarely call for corrective thoracoplasty. When there is no significant mediastinal displacement, with overdistension of the contralateral lung, a corrective procedure may not be essential in regard to the residual space after pneumonectomy. Corrective thoracoplasty may, however, be necessary where the patient is resistant to antibiotics and chemotherapy. Recent statistics suggest that the introduction of iso-nicotinic acid has had a notable influence on the incidence of complications. Berard, of

42

I R I S H JOURNAL OF MEDICAL SCIENCE

Lyons, referring to 950 resection operations, has noted that bronchial fistula has largely disappeared as a complication since its introduction. Although resection surgery, as applied to the treatment of pulmonary tuberculosis, is a relatively recent procedure, a number of important statistics have already been published. Holmes Sellors, reporting on 538 resections, has noted 72 per cent. early good results, 14 per cent of complications, and a mortality rate of 7 per cent. Eerland, analysing his first 447 patients, followed up over periods from 2 to 8 years after operation, has reported 76"6 per cent. of pneumonectomies, 86 per cent. lobectomies, and 96'7 per cent. segmentectomies leading an active life. The mortality rate over the same period was 13'1 per cent. for pneumonectomies, 3"9 per cent for lobectomies, and 0 per cent. for segmental procedures. Overholt has recorded a significantly higher mortality rate, viz., 15 per cent. for right sided pneumonectomy as compared with 5"4 for left pneumonectomy. Analysis of the last 105 resection procedures carried out for pulmonary tuberculosis at St. Patrick's Chest Hospital, Castlerea, gave the following figures. Deaths

Lobectomies

42

Pneumonectomies

37

3

Segmentectomies

22

2

Decortication Total

4 105

Fistulae

--

-Mortality 4.76%

Satisfactory Results

Unsatisfactory or, as yet, insufficient

4

37

3

34

--

20

--

~ 1 4.6%

5

3 Good Results 89.5%

4.76%

The maximum follow-up period for any of the cases included in these figures is two years and eight months. Of the three deaths with pneumonectomies, two were in relation to procedures on the right lung. Of the 105 cases, at least 90 per cent. had initially extensive, often bilateral, 9llsease requiring prolonged bed rest and drug therapy to render them suitable for surgery. M a n y important statistics already published have included resection procedures in regard to minimal lesions. Such minimal lesions are not i~mtuded i n the 105 cases in the above series. All published results confirm the major advance in the treatment of pulmonary tubereu]osis realised by the development of lung resection. It has provided the solution to many previously insoluble problems in the management of the disease. The judicious exploitation of this new surgical contribution calls for its correct integration with the already proven methods of treatment. Resection surgery has its definite place in the treatment of pulmonary tuberculosis. What that place is to be calls for ever more careful ,observation and examination of the patient.