extrapleural pneumothorax: a review of 128 cases

Thorax (1948), 3, 166.

EXTRAPLEURAL PNEUMOTHORAX:A REVIEW OF 128 CASES

BY

A. T. M. ROBERTS

From the Brompton Hospital, London

The first extrapleural pneumothorax induced atthe Brompton Hospital, London, was performedby Mr. J. E. H. Roberts, on May 10, 1937, andthis was the first operation of its kind to be per-formed in England or America since the intro-duction of the modern operation by Graf. Duringthe following five years, up to May, 1932, theoperation was attempted on 147 occasions at theBrompton Hospital. This paper is an analysis ofthese cases, based on a study of the pre-operativehistory in each case, and as complete a follow-upas possible.

ANALYSIS OF CASES REVIEWEDOPERATIVE PROCEDURES.-The operative pro-

cedures are summarized in Table I. On 16 (11 percent) of the 147 occasions the operation was aban-

TABLE IOPERATIVE PROCEDURES IN 147 CASES IN WHICHEXTRAPLEURAL PNEUMOTHORAX WAS ATTEMPTED

BETWEEN MAY, 1937, AND MAY, 1942

Procedure Number

Abandoned on table .. .. .. 12Thoracoplasty performed .. .. 4Case notes missing .. .. .. 3"Standard" operation .. .. .. 100"Combined" operation .. .. .. 28

Total operations attempted .. 147

doned on the operating table due to the density ofadhesions in the extrapleural plane; in four ofthese cases the surgeon proceeded to perform a

formal thoracoplasty, while in the remainingtwelve the patient was returned to the ward afterclosure of the wound. A further three cases havehad to be omitted from this review because thecase notes were missing. The remaining 128patients have been traced as far as possible.

The cases fall into two main groups which havebeen studied separately:

1. One hundred cases in which a straight-forward extrapleural strip and apicolysis wasperformed-the " standard " operation.

2. Twenty-eight cases in which the extra-pleural and intrapleural spaces were joined,either at or shortly after operation, and refillscontinued to the combined intra-extrapleuralspace-the " combined " operation.In five cases an extrapleural pneumothorax was

induced on each side. at separate operations, but inonly two cases were both operations performedat the Brompton Hospital.AGE DISTRIBUTIoN.-The youngest patient was

a girl of 13 years, while the eldest was a man of58 years. In 12 cases the patient was under 20years old. The ages of the patients, male andfemale, at the time of operation are shown inTable II.

TABLE IIAGE AND SEX DISTRIBUTION

Age in years 10-19 20-29 30-39 40-49 50-59 Total

Male .. 3 19 26 10 1 59Female .. 9 38 18 4 - 69

LENGTH OF FoLLOw-UP.-Some of the patientsare still under the care of the Brompton Hospital,but in the majority the follow-up has only beenpossible with the help of- tuberculosis officers,superintendents of sanatoria, and general practi-tioners throughout the country. Details of thelength of follow-up, together with the numbersknown to be living or dead, are summarized inTable III.Of the 128 patients, 118 have been traced either

to death or to-within a year of the date of follow-up, while the remaining ten have been traced for

group.bmj.com on March 25, 2018 - Published by http://thorax.bmj.com/Downloaded from

http://thorax.bmj.com/

http://group.bmj.com

EXTRAPLEURAL PNEUMOTHORAX

varying periods up to eight years from the date ofthe operation. In only five (4 per cent) was thefollow-up less than five years. Of the 128 opera-tions, 108 were performed during 1937 and 1938,so that an eight- to nine-year follow-up has been

possible in most cases. Fifty-six patients areknown to be dead. The overall average follow-up of the remaining seventy-two patients, knownto be living at the time up to which they weretraced, is 8.0 years.

TABLE IIILENGTH OF FOLLOW-UP IN YEARS

Year of Total Dead -______ ______ Length of follow-up, in yearsoperation cases 9 8 7 6 5 4 3 2 !1

1937 .. 46 24 20 1 _- 1 -1938.. 62 25 11 21 1 22 - - 1 11939.. 9 6 - 3 - - - - - 21940. 6 - - - 1 3 - 1 - - 11941. 4 1 - - - - 3- - - -1942. 1 - - - - 1 - - - -

Totals. 128 56 31 22 5 3 6 1 - 2 2

TABLE IV

TABLE VCAUSES OF FOURTEEN POSTOPERATIVE DEATHS AFTER THE "STANDARD"1 OPERATION

Time Condition ofAge Sex Cause of death after contralateral

operation lung30 M Postoperative shock. Extrapleural pneumothorax induced on Extrapleural

contralateral lung 6 weeks before .. .. .. .. .. 1 day pneumothorax25 F Haemorrhage and bronchopleural fistula .. .. .. 2 weeks Infiltration28 M Tuberculous infection of extrapleural space .. .. . 5 mths. Normal34 F Haemorrhage into extrapleural space and massive collapse of lung 2 days Infiltration17 F Tuberculous infection of extrapleural space .. .. .. 31 mths. Infiltration

and cavity29 F Acute bronchitis and cardiac failure.. .. .. .. .. 3 days Re-expanded

A.P.23 M Pyogenic infection of extrapleural space I.... .. .. 1 days A.P.36 M Tuberculous infection of extrapleural space .. .. .. lj yrs. Infiltration32 M Tuberculous toxaemia and cardiac failure .. .. .. .. 7 days Infiltration26 F Apical cavity opened at operation, tuberculous infection of extra-

pleural space .. .. .. .. .. .. .. .. 5 weeks A.P. and cavity27 M Tuberculous infection of extrapleural space .. .. 1 year Infiltration

and cavity30 F Pyogenic infection of extrapleural space .. .. 5 mths. Infiltration32 M Acute bronchitis. Very inefficient contralateral lung .. .. 1 mth. Re-expanded

A.P.30 F Tuberculous infection of extrapleural space .. .. .. 6 mths. Normal

167




A. T. M. ROBERTS

TABLE VICAUSES OF FOURTEEN LATE DEATHS AFTER THE "STANDARD OPERATION

Time |Condition ofAge Sex Cause of death after contralateral

operation lung

25 F Spread of disease in contralateral lung .. .. .. 4 mths. Infiltrationand cavity

19 F Spread of disease under contralateral A.P. .. .. .. 9 mths. A.P.38 M Cavity not closed-thoracoplasty-spread of disease .. .. lI yrs. Infiltration19 F Late spread of disease under extrapleural pneumothorax .. 1 year Infiltration29 --F Contralateral tuberculous empyema .. .. .. .. 7 mths. A.P.20 M Haemoptysis-spread of disease .. .. .. .. .. 1 yrs. Infiltration

and cavity22 M Spread of disease .. .. .. .. .. .. .. 2 yrs. Infiltration

and cavity35 F Contralateral thoracoplasty .. .. .. .. .. .. 2 yrs. Infiltration

and cavity22 M Spread of disease II.... .. .. .. .. .. yrs. Normal48 F Tuberculous infection of extrapleural space-recovery. Later

spread .. .. .. .. .. .. .. .. .. Ij yrs. A.P.23 F Haemoptysis-spread of disease .. .. .. .. .. 3 yrs. Infiltration19 F Spread of disease .. .. .. .. .. .. .. 2 yrs. Infiltration44 F Spread of disease .. .. .. .. .. .. .. 1 yrs. Infiltration36 F Spread of disease .. .. .. .. .. .. .. I year Old A.P.

RESULTS iN ONE. HUNDRED PATIENTS SUBMITTEDTO THE " STANDARD " OPERATION

In sixty-six of a hundred cases submitted to the".standard" operation the disease was bilateral atthe time of operation. In seventy-eight an arti-ficial pneumothorax on the operated side had beenattempted and failed or had been abandonedbecause of contraselectivity.DEATHS.-Forty-five of the patients are known

to be dead. The deaths, which occurred at varyingintervals after operation as shown in Table IV, andwhich in many cases could not be attributed to theoperation, have been divided into three categories-postoperative, late, and remote.

Postoperative deaths (Table V).-Fourteendeaths can be attributed to the effects of the opera-tion or its immediate complications.

Late deaths (Table VI).-Fourteen deaths weredue to spread of the disease in spite of the opera-tion, although the operation itself was not a factor.Remote deaths.-Fifteen patients enjoyed a

period of normal health after apparent control oftheir disease by the operation, only to relapse laterand succumb -to spread of the disease. Twofurther deaths were from non-tuberculous causes-gastric ulcer in one case and nephritis in the other-but in neither was the tuberculous disease com-pletely under control at the time of death.

LIVING PATIENTS.-Fifty-five of the hundredpatients in-the " standard " series were alive after

an average period of eight years. Forty-six werefree of symptoms and were doing full-time workwith their disease quiescent or arrested. Refillswere still being given to three, a thoracoplasty hadbeen performed in seven, and an oleothorax intwo. In the remaining thirty-four cases the extra-pleural pneumothorax had obliterated or had beenvoluntarily abandoned.Nine patients still had symptoms of active

disease. Two of these were still having refills, onehad had an oleothorax, and six had dischargingsinuses five following thoracoplasty and onefollowing drainage of a late pyogenic infection of

TABLE VIIPROGRESS OF ONE HUNDRED PATIENTS SUBMlED TO

THE "STANDARD" OPERATION

Postoperative .. .. 14Deaths .. Late .. .. .. 14 45%

Remote .. .. .. 17

Continuing refills .. 3Alive and Voluntarily abandoned 24

well .. Obliterated .. .. 10 46%Thoracoplasty .. 7Oleothorax .. 2

Continuing refills .. 2Alive with Voluntarily abandonedsymptoms Obliterated .. .. 1 9%

Thoracoplasty .. .. 5Oleothorax I.;. 1

i

168





the extrapleural space. The progress of the hun-dred patients submitted to the " standard" pro-cedure is summarized in Table VII.POSTOPERATIVE COMPLICATIONS OF THE " STAND-

ARD " OPERATION.-Tuberculous Empyema of theExtrapleural Space.-This occurred in thirteenpatients, of whom ten subsequently died (six post-operative, one late, and three remote deaths). Twopatients were alive and well after thoracoplastyand one was alive and well after the infection hadsubsided spontaneously. Cases in which only anoccasional tubercle bacillus was found in the extra-pleural fluid are not included in this group.Pyogenic Infection of the Extrapleural Space.

-Infection with streptococci, staphylococci,or pneumococci developed in fourteen patients.Intercostal drainage or rib resection was performedon eleven of these patients. Four were dead (twopostoperative and two remote deaths), four werealive and well, and six were alive with dischargingsinuses. Of the nine patients subsequently sub-mitted to thoracoplasty, only one has since died.

Bronchopleural Fistula.-Bronchopleural fistulacomplicated infection of the extrapleural spacein seven cases. Three of these patients sub-sequently died. Of the other four, two recoveredspontaneously after abandonment of reffils, onewas alive and well after thoracoplasty, and onehad a dicharging sinus after thoracoplasty.Haemorrhage.-Some oozing of blood into the

extrapleural space after operation is inevitable. Ineight cases, however, serious haemorrhageoccurred, and in two it was the cause of post-operative death. The remaining six patientsrecovered eventually after thoracoscopy withremoval of the blood and blood clot and adequatetransfusions.

Atelectasis.-The degree of relaxation of thelung after the operation varied considerably andin twelve cases amounted to a complete and per-sistent atelectasis of the whole lung. Six of thesepatients eventually died, four from tuberculousempyema, one from added haemorrhage, and onefrom the resultant anoxia. In the remaining sixcases re-expansion slowly took place over a periodof months and the patients were alive and well atthe time of follow-up.Effusion.-In most cases a sterile effusion.

developed after operation, but in the majority iteventually absorbed leaving a dry space. In four-teen the effusion persisted, slowly turned to sterilepus and necessitated the abandonment of thepneumothorax.The total of sixty-eight complications listed

above actually occurred in forty-nine patients,

since some patients suffered more than one com-plication. Thus approximately 50 per cent ofpatients developed some postoperative compli-cation.

RESULTS IN TWENTY-EIGHT PATIENTS SUBMITTEDTO THE " COMBINED" OPERATION

Nineteen of the twenty-eight "combined"operations were performed in the presence of acontraselective artificial pneumothorax, the septumbetween intrapleural and extrapleural spaces beingdivided widely at operation. In the remainingnine cases an artificial pneumothorax was inducedat operation, deliberately in six instances butaccidentally in three. The septum was divided atoperation in five of these cases but in the otherfour it was divided by cautery through a thora-scope several weeks later.DEATHS.-Eleven patients are known to be dead

(six postoperative and five remote deaths), givingan overall mortality of 39 per cent and a post-operative mortality of 21 per cent. Five of thepostoperative deaths were due to tuberculousinfection' of the combined intra-extrapleural space,while the sixth followed a first-stage thoracoplastyperformed to deal with a bronchopleural fistula.LIVING PATWENTS.-Twelve patients, of whom

five were still having refills, were alive and well.Five patients, of whom two were still having re-fills, were alive with symptoms. The progress ofthe twenty-eight patients submitted to the " com-bined" operation is summarized in Table VIII.

TABLE VIIIPROGRESS OF TWENTY-EIGHT PATIENTS SUBMITTED TO

THE "COMBINED" OPERATION

Postoperative .. 6 (21 %/)Deaths .. Late.. .. ..39%

Remote_..5 (18%)

Continuing refills .. 5Alive and Voluntarily abandoned 3well Obliterated .. .. 2 43 %Thoracoplasty .. 2

Continuing refills .. 2Alive with Voluntarily abandoned -symptoms Obliterated .. .. 1 18%

Thoracoplasty .. 2

POSTOPERATIVE COMPLICATIONS OF THE "COM-BINED " OPERATIoN.-Tuberculous infection of theextrapleural space occurred in eight patients, repre-senting 28 per cent of the total and, as mentionedabove, was the cause of postoperative death in fivecases. Pyogenic infection of the extrapleural space

169




A. T. M. ROBERTS

and bronchopleural fistula occurred in one caseeach, and persistent effusion in four cases. Aswould be expected, atelectasis of the whole lungwas not infrequent-it occurred in ten patients,five of whom developed tuberculous infection ofthe combined space, which led to the patient'sdeath. No case of severe haemorrhage was seenin the " combined" series.These twenty-four complications occurred in

sixteen patients, so that 66 per cent of patientsdeveloped one or more complications.

THORACOPLASTY AFTER EXTRAPLEURALPNEUMOTHORAX

In the whole series of 128 cases, thoracoplastywas eventually performed in twenty-four (19 percent) for the following reasons:

1. Tuberculous infection of the extrapleuralspace-five cases. One patient died after opera-tion, three were alive and well, and one wasalive with a sinus.

2. Pyogenic infection of the extrapleuralspace-ten cases. One remote death, five aliveand well, and four alive with sinuses.

3. Bronchopleural fistula-one patient whodied after the first stage of the thoracoplasty.

4. Disease not controlled by extrapleuralpneumothorax-seven cases. One postoperative,one late, and three remote deaths. One aliveand well and one alive with a sinus.

5. Closure of the extrapleural space-onepatient, seven and, a half years after the originaloperation, now alive and well.*The -progress of the twenty-four patients on

whom thoracoplasty was eventually performed issummarized in Table IX.

TABLE IXRESULTS IN TWENTY-FOUR PATIENTS SUBMITTED TO

THORACOPLASTY AFTER EXTRAPLEURAL PNEUMOTHORAX

Deaths No. % Alive No. %

Postoperative 3 Well 10 42Late .. 1 33 WithRemote .. 4 symptoms 6 25

THE FATE oF THE ExTRAPLEuRAL SPACE ATERVOLUNTARY ABANDONMNT oF REFILLS

Thirty-one (24 per cent) of the 128 patients areknown to have had their refills abandoned volun-tanrly after an average period of five years; anaverage follow-up of a further three and a halfyears after abandonment has been possible.

fThree and three and a half years respectivelyafter abandonment of refills two patients died fromspread of disease in the contralateral lung; inboth cases the lesion for which the extrapleuralpneumothorax had been performed was controlledat the time of death. In two cases thoracoplastywas performed on the extrapleural pneumothoraxside, one immediately after abandonment of avery large extrapleural space and the other threeand a half years later. In both cases the thoraco-plasty was necessitated by reactivation of theoriginal lesion, but both patients were alive andwell at the time of follow-up.The remaining twenty-seven patients (twenty-

four " standard " and three " combined ") hadremained well and working for an average periodof three and a half years. No relapse of the tuber-culous process or any other complication hadoccurred after the last refill in these cases. Inthirteen of the twenty-seven cases the extrapleuralspace was filled by a " cap " of sterile fluid and infive by an area of fibrosis. In the remaining ninedetails were not available.

FAcroRs AFFEcrING PROGNOSISThe incidence of postoperative deaths and of

complications is high in this series as a whole, andit is not surprising that the procedure fell intorelative disuse-as is strikingly borne out by thediminishing frequency with which the operationwas performed between 1937 and 1942 (Table Ill).However, many of these operations were per-formed soon after the operation was introducedinto this country-in fact 108 were performedduring 1937 and 1938 alone. It was inevitable, inthose early days, that many patients were operatedon who would now be considered unsuitable.Turning to the credit side of the picture, more-over, nearly half the patients were alive, symptom-less, and working eight years on the average aftertheir operations-46 per cent in the " standard"group and 43 per cent in the " combined " group.These results were achieved in a group of patientswho would have had a poor prognosis, for mosthad positive sputum tests and often considerablecavitation, and they could not be treated by artifi-cial pneumothorax because of adhesions. Manyalso had bilateral disease.

It would obviously, therefore, be of great advan-tage to deduce from the available data any infor-mation about factors influencing prognosis fteroperation. To this end the hundred cases treatedby the " standard " procedure have been dividedinto two large groups: (a) successful cases, inwhich no serious or lasting complications arose,and in which 'the patient returned to a normal life

170)





with the disease well under control; and (b) unsuc-cessful cases, in which the operation either led todeath or left the -patient worse or no better off thanbefore. Any doubtful case has been deemedunsuccessful but nearly every case fell readily intoone or other group. Not all the patients who diedwere necessarily unsuccessful cases-four of theremote deaths occurred after the extrapleuralpneumothorax had been well stabilized for severalyears, and these cases have been included in thesuccessful group., Conversely, four patients whoare still alive but with marked symptoms have beenincluded in the unsuccessful group.By these criteria there were, in the " standard"

series, fifty-five successful and forty-five unsuccess-ful cases.The cases were then divided into various cate-

gories according to age and sex, length of historybefore operation, radiological appearances, andextent of disease in the contralateral lung. In eachcategory the percentage of successful and unsuc-cessful cases was determined. Difficulties arose inassessing the radiological features. Judgment ofthe degree of fibrosis from postero-anterior radio-graphs is open to so much error that no attempthas been made to assess the cases from this aspect.The cases were, however, divided into two cate-gories according to whether the cavity appeared to

TABLE XANALYSIS FACTORS ASSOCIATED WITH SUCCESSFUL ANDUNSUCCESSFUL RESULTS IN ONE HUNDRED PATIENTS

SUBMITfED TO THE "STANDARD" OPERATION

Successful UnsuccesfulCategory

No. % No. %

Total male .. 24 56 19 44Total female .. 31 54 26 46Total male plusfemale .. .. 55 55 45 45

0-1 year .. 31 74 11 26Length 1-2 years .. 11 50 11 50of 2-3 years .. 5 55 4 45history 3-5 years .. 4 44 5 56

>5 years .. 4 22 14 78

Size <2 cm. .. .. 24 71 10 29of 2-4cm. .. 20 49 21 51cavity >4 cm. .. .. 9 39 14 61

Situation "Subpleural" .. 20 40 30 60ofcavity Not "subpleural" 33 69 15 31

Normal .. 26 87 4 13Contra- Infiltration .. 20 43 26 57lateral Cavitation .. 5 38 8 62lung A.P. .. .. 4 36 7 64

be subpleural or not, though it was realized thatthis criterion involves a fallacy in that a cavity mayappear to be central in the postero-anterior radio-graph and yet may well be subpleural eitheranteriorly or posteriorly.Radiographs were not available at the time of

review for two male patients in the " successful"group. The results of this analysis, which aretabulated in Table X, were as follows.AGE AND SEx.-The numbers in the various age

groups are too small for the results to be of anysignificance and therefore they are not tabulatedin Table X. No difference at all is apparent in theresults in the two sexes.LENGTH OF HIsToRY.-The prognosis appears to'

vary directly with the length of history of symp-toms before operation-the shorter the history thebetter the prognosis. In those cases with a historyof less than one year, 74 per cent were supcessful,while in those with a history of five years or over,only 22 per cent were Auccessful.

SIZE OF CAvinT.-The greatest transverse dia-meter of the cavity or system of cavities wasmeasured in centimetres. The figures show thatthe success of the operation varied inversely withthe size of the cavity. With cavities of 2 cm. orless in maximum diameter 71 per cent were suc-cessful, while with cavities of over 4 cm. in dia-meter only 39 per cent were successful.

POSITION OF CAviY.-Cases in which the cavitywas subpleural as judged by the postero-anteriorradiograph had a considerably worse prognosisthan those in which the cavity was more central.EXTENT OF DISEASE IN THE CONTRALATERAL

LUNG.-As would be expected, those cases with nodisease in the contralateral lung at the time ofoperation fared better than those in whom thedisease was already bilateral. Of the completelyunilateral cases, 87 per cent were successful.The criteria suggesting a good prognosis appear

therefore to be (1) a short history; (2) a smallcavity which is not obviously subpleural; and(3) the absence of important disease in the contra-lateral lung.

These criteria are in conformity with those ofMonod (1938) and of Dolley, Jones, and Skillen(1940). Monod considered that " the recent anddiscretely developed lesion with a small cavitysituated at a distance from the periphery" was theideal type of case for the operation. He warnedagainst the dangers of performing the operation inthe presence of fibrous disease or of large cavities,especially when these are subpleural, or of long-standing lesions " with a rigid shell." Dolley,

171




A. T. M. ROBERTS

Jones, and Skillen divided their indications -intothree grades: absolute, broad, and conditional.Their " absolute" indications were: that artificialpneumothorax had failed, that cavities were neitherfibrotic nor stiff-walled, were not more than 4 cm.in diameter, and did not extend below the sixthrib posteriorly, that the contralateral lung was freeof active tuberculosis, and that the lesions werenon-acute, recent, and limited in extent.

REVIEW OF THIRTY "GOOD RISK" CASESThe number of cases in which all the above

criteria were obeyed was very small-only nineamongst the hundred cases submitted to the"standard " procedure. There were, however,thirty patients in the " standard" group with ahistory of symptoms for less than two years andwhose cavities were less than 3 cm. in maximumdiameter. These thirty "good risk" cases havebeen assessed separately.They consist of fifteen men and fifteen women

with ages ranging from 13 to 46 years (average 28years). In all but four an artificial pneumothoraxcould not be induced or had been abandonedbecause of contraselectivity. In only eight wasthe cavity deemed to be subpleural. The extent ofdisease in the contralateral lungs was: no disease,15 cases; infiltration only, 14 cases; cavitation,1 case. In none of these cases was there an artifi-cial pneumothorax for treatment of disease in thecontralateral lung.

Five patients are known to have died, and theremaining twenty-five have been traced for anaverage follow-up period of eight and a half years.There was only one postoperative death, six anda half months after operation, from pyogenicinfection of the extrapleural space. There wereno late deaths. Three remote deaths occurredfrom spread of pulmonary tuberculosis, four anda half, six and a half, and eight years respectivelyafter operation, and one from nephritis one yearafter operation. Of the twenty-five living patients,thirteen had had their pneumothoraces voluntarilyabandoned after an average period of 4.8 years; allthese were symptomless after an average periodof 3.9 years since abandonment. Nine were- aliveand symptomless after early obliteration of theirextrapleural spaces after an average period of 1.1years; the average follow-up after obliteration wasa further seven years. Two patients eventuallyrequired a thoracoplasty, one because of spread ofthe disease four years after operation, and onebecause of pyogenic infection of the extrapleuralspace six months after operation. Both are aliveand well, though the latter has a discharging sinus.One patient had an oleothorax induced shortly

after operation; but after three months, duringwhich he was progressing well, he was lost sight ofand has not been traced.

COMPLICATIONS IN THE " GOOD RISK" CASES.-There were no instances of tuberculous infectionof the extrapleural space and only two of pyogenicinfection, both associated with bronchopleural

TABLE XIPROGRESS OF THIRTY "GOOD RISK" PATIENTS SUB-

MITED TO THE "STANDARD" OPERATION

Postoperative .. 1 (3.3%YO)Deaths Late ..16.7%

Remot e 4 (13.3%)Well: 80%Continuing refills . .

Alive . . Voluntarily abandoned 13Obliterated .. .. 9Thoracoplasty .. 1Oleothorax . . .. 1With sinus after thor-acoplasty: 1 3.3%

fistula. Seven patients developed a persistentsterile effusion which led to obliteration of theextrapleural space. There were two instances ofpostoperative haemorrhage, but both patientsrecovered satisfactorily after thoracoscopy andevacuation of blood clot. The progress of thethirty patients in the " good risk " group is sum-marized in Table XI.

CONCLUSIONS AND SUMMARY1. One hundred and twenty-eight patients who

were treated by extrapleural pneumothorax at theBrompton Hospital for pulmonary tuberculosisbetween May, 1937, and May, 1942, have beenfollowed up.

2. All except five patients (4 per cent) have beentraced either to the time of death or for at leastfive years. Fifty-six (44 per cent) were dead; 58(45 per cent) were alive, symptomless, and work-ing; and 14 (11 per cent) were alive with symp-toms. The average follow-up period of theseventy-two living patients was eight-years.

3. In twenty-eight cases the extrapleural pneu-mothorax was joined at or shortly after operationto an intrapleural pneumothorax on the same side(the "'combined " operation); in a hundred casesthe extrapleural pneumothorax alone was estab-lished (the " standard " operation).

4. The postoperative mortality was 14 per centWith the "standard" operation and 21 per centwith the "4combined" operation. The incidence

172





of tuberculous infection in the extrapleural spacewas 13 per cent in the " standard " and 28 per centin the " combined " series. The incidence of otherpostoperative complications, was also high in bothseries.

5. Forty-five per cent of the patients, however,were alive, symptomless, and working, and veryfew late complications had arisen once the extra-pleural pneumothorax had been satisfactorilystabilized.

6. In thirty-one cases the extrapleural pneumo-thorax had been voluntarily abandoned after anaverage period of five years.

In 27 (84 per cent) of these cases no relapses orcomplications had arisen during a further averageperiod of three and a half years since abandon-ment.

7. A study of the hundred " standard " casesrevealed that the main factors leading to a success-ful outcome for the operation were: (a) a shorthistory; (b) a small cavity which was not obviously

subpleural; (c) minimal disease in the contralaterallung. The findings are in agreement with the indi-cations of Monod (1938) and the "absolute"indications of Dolley, Jones, and Skillen (1940).

8. There were thirty " standard " cases with ahistory of symptoms for less than two years anda cavity of less than 3 cm. in maximum diameter.Twenty-four (80 per cent) of these patients were,alive, symptomless, and working and only one diedpostoperatively.

I wish to thank Dr. Geoffrey Marshall, whosuggested and guided this investigation; Prof. A.Bradford Hill for advice on the statistical value ofthe figures; and the Medical Committee of theBrompton Hospital for permission to carry out theinvestigation and to publish this paper.

REFERENCESDolley, F. S., Jones, J. C., and Skillen, J. (1940). -Amer.

Rev. Tuberc., 41, 403.Monod, 0. (1938). J. thorac. Surg., 8, 150.

N

173




Review of 128 CasesExtrapleural Pneumothorax: A

A. T. M. Roberts

doi: 10.1136/thx.3.3.1661948 3: 166-173 Thorax

http://thorax.bmj.com/content/3/3/166.citationUpdated information and services can be found at:

These include:

serviceEmail alerting

the online article. article. Sign up in the box at the top right corner of Receive free email alerts when new articles cite this

Notes

http://group.bmj.com/group/rights-licensing/permissionsTo request permissions go to:

http://journals.bmj.com/cgi/reprintformTo order reprints go to:

http://group.bmj.com/subscribe/To subscribe to BMJ go to:


http://thorax.bmj.com/content/3/3/166.citation

http://group.bmj.com/group/rights-licensing/permissions

http://journals.bmj.com/cgi/reprintform

http://group.bmj.com/subscribe/



extrapleural pneumothorax: a review of 128 cases

Documents