RECURRENT AND CHRONIC SPONTANEOUS PNEUMOTHORAX* · RECURRENT AND CHRONIC SPONTANEOUS PNEUMOTHORAX...

Thorax (1948), 3, 88.

RECURRENT AND CHRONIC SPONTANEOUSPNEUMOTHORAX*

BY

R. C. BROCK

London

The general problem of spontaneous pneumo-thorax in the apparently healthy, or "pneumo-thorax simplex" as it has been termed byKjaergaard (1932), has been dealt with fairly ex-tensively by a number of authors. A certainamount has been written about recurrent attacksand somewhat less about chronic spontaneouspneumothorax, but nothing really authoritativehas been published. This paper deals with thesetwo manifestations of spontaneous pneumo-thorax; it is based on a study of seventy-onepersonal cases.From a study of the literature the incidence of

single attacks of spontaneous pneumothorax wouldnot appear to be high. One of the most compre-hensive papers is by Kjaergaard (1932), whosefifty-one cases were gathered from hospitals inCopenhagen and from about fifty others in Den-mark, and covered a period of twenty years. Perry(1939) based his interesting and useful study of thecondition on eighty-five cases which were admittedto the London Hospital between 1924 and 1937, aperiod of fourteen years. Ornstein and Lercher(1941-2) discuss fifty-eight cases occurring in theSeaview Hospital. In practice " pneumothoraxsimplex" is found to be commoner than this,especially when the condition is being looked forand when patients are radiographed more fre-quently. It is, for instance, diagnosed in studentsand nurses often enough to suggest that many morecases are overlooked in the population at large.Perry found many extra cases in the out-patientdepartment when he was deliberately on the look-out for the condition.The majority of cases described by authors are

of simple, single attacks; in the literature theincidence of recurrence and chronicity is extremelysmall. Thus Kjaergaard's fifty-one cases includedseven which had recurrent attacks and four chroniccases. Perry records only four recurrent and two

*Based on a paper read before the Thoracic Society in February,1948.

chronic cases amongst his eighty-five ; in his surveyof the literature he found thirty-three singlecases of recurrent pneumothorax. Ornstein andLercher's series of fifty-eight cases included seven-teen with recurrent attacks. Nikolski (1912) hadnine recurrent cases in his series of ninety. Wood(1931) states that, in seventy-one patients withspontaneous pneumothorax at the Mayo Clinic, 21per cent gave a history of multiple attacks and in11 per cent both sides had been affected at differenttimes; in three cases both sides had been affectedsimultaneously. There are a certain number ofreports in the literature of chronic cases, mostlysingle cases, occasionally two or three and nevermore than a handful; a number of these have beenfound accidentally, and from the radiographs pro-duced it is certain that some are not examples ofchronic pneumothorax but of giant tension bullaeor cysts.

I have been unable to discover any comprehen-sive account of a large series of recurrent andchronic cases of spontaneous pneumothorax, andit is fair to state that the series of seventy-onecases that forms the basis of this study is unique.I would stress that these are all of the chronic orthe recurrent type and include none with simpleisolated or short attacks; all have been in my ownpractice, seen and treated by me, and they coverthe years 1936-47. While it is true that a numberof friends and colleagues, knowing my interest inthis condition, have kindly referred their cases tome, it is nevertheless worthy of comment that sucha large number has been seen by one surgeon inonly eleven years. It suggests that recurrent andchronic cases are far commoner than has hithertob:en supposed. It is certain that they constitutea real and important problem and one that it iswrong to dismiss lightly as is so often done in dis-cussion of the condition. From my experience ofthe disability caus2d, I find it difficult to accept thecomplacent comments often made that there i, noneed to worry about treatment; that, given time,

on 15 August 2019 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.3.2.88 on 1 June 1948. D

ownloaded from

http://thorax.bmj.com/

RECURRENT AND CHRONIC SPONTANEOUS PNEUMOTHORAX

the lung will always expand; that the patientshould be rested in bed or, in recurrent cases,merely advised to avoid doing anything strenuous.Such remarks betray complete failure to under-stand the problem as it really exists.

DEF.NITIONAlthough the lung in " pneumothorax simplex"

usually expands in two or three weeks, it may taketwo or even three months to do so. If it is stillcollapsed after three months it is fair to classifyit as a chronic case. Chronicty occurred in forty-six of my own series, and the average duration inthese was fifteen months; the longest was nine anda half years; in one the duration was three and ahalf years, in two it was two and a half years,and in a fifth two years. Amongst these forty-fivechronic case; the condition was entirely chronicin twenty-nine; in seventeen, chronicity super-vened upon recurrent attacks (Table I).

TABLE ICASES OF CHRONIC PNEUMOTHORAX IN A SERIES OFSEVENTY-ONE CASES OF SPONTANEOUS PNEUMOTHORAX

Chronic super-Age Chronic vening or Total chronic

recurrent

1-10 2 0 211-20 0 2 221-30 3 7 1031-40 6 6 1241-50 13 2 1551-60 5 0 561-70 0 0

l 921 men 10fmen 31 menTotal 29 8 women] 17{ 7 womeni46 15 women

Duration| Average 15 months; longest 94, 34, 21, 24,and 2 years

Twenty-five in my series presented as purelyrecurrent cases; in a further seventeen recurrencesended in chronicity, making a total of forty-twoin which relapses occurred (Table IL). The recur-

rences may be all on one side, or they may bealternating bilateral (fifteen in my series) as firstreported by Goodhart (1896), or both sidesmay be involved simultaneously (Plate XXIa)(S examples). In others a chronic pneumothoraxhas been present on one side and recurrent attackshave occurred on the Oppposite side. The averagenumber of attacks was four; the greatest numberin one patient was fifteen: another patient hadcleven attacks, and a third had ten. Locke (1929)mentions a recurrent case in which eighteen attacks

occurred over a period of seven years; both sideswere affected at different times.

TABLE IICASES OF RECURRENT PNEUMOTHORAX (IN SEVENTY-ONE CASES OF SPONTANEOUS PNEUMOTHORAX)

kge Recurrent iRnegcurrent end- Totaling in chronic Toa

1-10 0 0 011-20 4 2 621-30 10 7 1731-40 6 6 1241-50 2 2 451-60 2 0 261-70 1 0 1

Total 25 {21 men 17 42

Bilateral: 16(10 menAlternating { 5 women 15

Simultaneous I 8

Average number of attacks 4: largest numbers 15, 11, 10

SIGNIFICANCEThese figures emphasize the seriousness of the

condition, and this is amply borne out by first-handknowledge of the cases themselves. I have alreadyreferred to misguided statements that chronicitydoes not seem to matter and that the lung willusually expand in time; or that relapses can betreated by further rest in bed or by prohibitingexercise. The first statement is simply not true;most patients with a chronic pneumothorax sufferconsiderable disability, and many are chronic in-valids. This is quite apart from the very real dangerof a further pneumothorax on the opposite side,and also the danger of an acute respiratory infec-tion on the good side. One of the few cases in theliterature studied post-mortem (Brunner, 1921) wasof a patient who died from an attack of influenzalpneumonia in the opposite lung. Many of thesepatients with a chronic pneumothorax have wastedmonths or even years resting in bed or in sanatoria.If tuberculosis is not the cause of the condition,it seems illogical, unkind, and wasteful to condemna patient to spend from six to twelve months rest-ing in bed for what is usually a simple mechanicalevent which can be quickly relieved by activemeasures.The same unnecessary restriction is often applied

to the relapsing cases, and it is particularly mis-guided to enjoin these patients to live a life ofrestricted activity Many of them are otherwise

89



j.com/


ownloaded from


R. C. BROCK

healthy, and such advice moreover reveals ignor-ance of the fact that many of the attacks occurwhile the patient is at rest: it is by no means un-common for them to happen when the patient islying in bed, or when he first gets out of bed in themorning. One of my patients, a man who had hadfifteen attacks on both sides and who had beentold by a doctor to avoid strenuous exercise, said,"My last attack occurred when I lifted up mybaby daughter; I am not to be allowed to do eventhat? " It is, of course, true that violent exerciseis more likely to cause an attack, and cycling is acommon precipitating cause.Although spontaneous pneumothorax is often

designated as something that occurs in theapparently healthy, my experience with andanalysis of these patients with recurrences orchronicity reveals that the condition is often anindication of the presence of notable disease in thelungs for which treatment may be needed. In factit is my impression that in chronic and recurrentpneumothorax we have not so much a clinicalentity as a manifestation in one aspect of thenatural history of many lung diseases. Thisreceives support from the fact that five of mypatients have since died of their primary disease.

ANALYSIS OF MATERIALStatistics are usually tedious and often disap-

pointing in what they convey, but it is still neces-sary to present them, albeit in as brief and simplea way as possible. Table III shows the relative

TABLE IIICOMPARISON OF THE PRESENT SERIES WITH 358 CASES

FROM THE LITERATURE

Present series (young-Age est patients 1l years, 358 cases collected41 years; oldest patient by Perry (1939)

64 years)

1-10 2 1411-20 6 5121-30 20 12031-40 18 5741-50 17 3951-60 7 16Over 60 1 5

Total 71 f52 men 302 in which age19 women recorded

that all decades are represented, the two youngestpatients being 14 and 44 years, and the oldest 64.The age distribution is, however, significantlydifferent from that of pneumothorax simplex, inwhich the greatest incidence is between 20 and 30.The higher incidence of chronic and recurrentcases in the third and fourth dccades reflects thediffering aetiology in these cases.

Clinical features.-The clinical features of pneu-mothorax simplex have been amply and frequentlypresented and there is no need to elaborate on thewell-known facts. Certain points deserve mentionin the chronic and recurrent cases. The greatertendency for the older age groups to be affectedhas just been mentioned, and this is associated witha greater incidence of the degenerative changes ofchronic bronchitis and emphysema. In very youngpatients congenital cystic disease may be a pos-sible cause, but the most frequent is staphylo-coccal pneumonitis with lung abscess. It is nowgenerally accepted that tuberculosis, although acommon cause of spontaneous pneumothorax, isbut rarely the cause in the group known as"pneumothorax simplex" or "pneumothoraxoccurring in the apparently healthy." Ornsteinstates that three of his fifty-eight patients developedtuberculosis about two years later; only one ofKjaergaard's fifty-one patients developed thedisease, and that was several years later and afterdefinite exposure to infection. One of my seventy-one patients has developed symptoms (five and ahalf years after) suggesting pulmonary tuberculosis,and so far even this is not proven; another patient,a boy of 12, with bilateral chronic pneumothoraxfrom severe congenital polycystic disease, died ofacute pulmonary tuberculosis two years after theonset of his pneumothoraces and eighteen monthsafter they had been cured. Pneumothorax intuberculosis is a late, often subterminal, complica-tion rather than an early manifestation. This isborne out in my series of chronic and recurrentcases in which pleural tuberculosis was found tobe the cause in only one instance. Although thispatient presented as a case of chronic pneumo-thorax of two and a half years' standing, and thepleura was dry when examined, there was a historyof several bouts of fever and of fluid formationin the air-space; this indicated the probability ofthe tuberculous basis which was easily confirmedat thoracoscopy. Fluid formation is rare in thenon-tuberculous cases, and when it does occur itis little more than a trace.

It is not uncommon for the presence of a pneu-mothorax to be noticed accidentally. For instancea doctor was found to have a pneumothorax whenhe was being examined for insurance; the only

incidence in men and women (52 men, 19 women),which compares closely with the figures collectedfrom the literature by Perry (301 men, 57 women).The age incidence is of interest, for it will be seen

90



j.com/


ownloaded from



PLATE XXI.-(a) To show bilateral spontaneouspneumothorax in a young woman who sufferedfrom asthma, chronic bronchitis, and emphysema.(b) Right chronic pneumothorax of seven months'duration in a woman aged 26 who was anasthmatic and bronchitic. (c) Radiograph of thesame patient as in (b) after right pleurodesis withsilver nitrate; the lung is now fully expanded.

(")

^@,,, t-A _

e __ ....... ..* _3: ....... : }|. .S S.s j_-_ -_. sa ._Iw

(h)

91

c)



j.com/


ownloaded from


R. C. BROCK

XCa)0

o

C v)o-

c'S

C.

0

t_on

+._

Ca)

CZ

92



j.com/


ownloaded from



(b)

PLATE XXIII.-(a) A radiograph of a giant bullouscyst simulating a chronic pneumothorax. (b) Thesame case as in (a); the bulla has increased insize and caused symptoms suggesting a tensionpneumothorax. The huge cyst was removedeasily and satisfactorily by operation. (c) Grossdestructive or bullous emphysema of the left lungsimulating chronic pneumothorax in a managed 56. The bullae were removed at operationwith an excellent result.

(a)

IC)

93



j.com/


ownloaded from


R. C. BROCK

.r. N fA...' _.1

W-W 4..) 14

x q-. -.. W

><5 Cd ed J

S.;.. .2 .:=s CL -,Lli u ed "f-

0444

94



j.com/


ownloaded from


RECURRENT AND CHRONIC SPONTANEOUS PNEUMOTHORAX 95

______I ____ff ' I'

(a)(/

(c)

PLATE'XXV.-(a) Radiograph frcm a case of recurrent, alternating pneumothorax in a young man. (b) Right broncho-gram of the same patient as in (a); shrunken, atelactatic and bronchiectatic middle and lower lobes are revealed.(c) Thoracoscopic view of a large bulla to show a perforation.



j.com/


ownloaded from


R. C. BROCK

PLATE XXVI.-(a) Bilateral congenital polycystic disease in a young man; a right pneumothorax is present. (b) Thesame case. A right pleurodesis has been induced. A left tension pneumothorax has supervened, and an indwellingneedle was inserted in order to save life. (c) The same case. The right lung has expanded but a cap of pleuralthickening or fluid still remains; the left lung has also fully expanded after pleurodesis. Note the thin-walledlarger cysts at the left base. (d) The same case. The cysts at the left base have now become grossly distended soas to simulate a localized tension pneumothorax. They subsided spontanecusly.

96



j.com/


ownloaded from



R6 ........... .. . . . . .............. . :'. ' . j ee 3 iQ.:. ::':.£ii:i4ESSu,_u___ i i_

FLATE XXVII.-Radiograph of a right tension pneumothorax which had recurred fivetimes and then became chronic. A huge mediastinal hernia is present. The edgeof a large thin-walled cyst can be seen above the right diaphragm.



j.com/


ownloaded from


R. C. BROCK

,--'01~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~"

PLATE XXVIII.-A woman, aged 30, with chronic spontaneous pneumothorax; showing the appearances seenat thoractomy.

98



j.com/


ownloaded from



-i.

*.,..

PLATEXXIX.-ShowingthecysticSlower|lobe.(Same patient as in PlateXXVIII.)

P'LATE XXIX.-Showing the cystic lower lobe. (Same patie-nt as in Plate XXVIII.)



j.com/


ownloaded from


R. C. BROCK

PLATE XXX.-Showing the expansion of the upper and middle lobes that began as soon as the cystic lower lobe hadbeen removed and the partial volvulus corrected. (Same patient as in Plates XXVIII and XXIX.)

100



j.com/


ownloaded from



information he could give was that some sixmonths earlier he had noticed a vague discomfortin the affected side, and during the last few yearshe had occasionally found he had been unable totake a deep breath. This pneumothorax was stillpresent four months later and was then treated bypleurodesis. It is by no means uncommon toobserve a small pneumothorax on the " good " sidewhich has given no indication of its presence, achronic or recurrent pneumothorax on the otherside dominating the clinical picture.A striking feature of the severe cases of chronic

pneumothorax is loss of weight. This is best illus-trated by Plates XXIb and c, and XXII, whichshow a patient (a woman aged 26) who had hada chronic pneumothorax for seven months; shesuffered from chronic bronchitis and asthma. Herweight dropped from 7 stone 4 pounds to 4 stone13 pounds, a fall of 2 stone 5 pounds. As willbe seen from the photographs, she regained herlost weight as soon as expansion of the lung wasachieved. Incidentally, a year after the chronicattack on the right side she suffered a spontaneouspneumothorax on the left side; it is just as wellthat her right side had been treated and not leftto take care of itself.Methods of investigation.-It is scarcely neces-

sary to do more than mention the importance oftaking a careful history of previous respiratorydisease. Inquiry about tuberculosis is not likely tobe omitted, but it is important to remember to askspecifically about asthma; information aboutattacks in childhood is often not volunteered.Evidence of lower respiratory infection, includingbronchitis or bronchiectasis, must be sought.Twenty patients in my series gave a history ofchronic bronchitis. In a further five, bronchiectasiswas present.

Plain radiography may show nothing more thanthe pneumothorax, especially if the lung is com-pletely collapsed. The opposite lung may showemphysema; it should always be carefully in-spected so that a shallow pneumothorax is notoverlooked.The first thing to consider is whether the con-

dition is a true pneumothorax or whether it is agiant cyst or bulla simulating one. This may attimes be very difficult. Plate XXIIIa and b showan example in which differentiation was not easy,although one felt fairly confident the condition wasa giant cyst. It will be noted that in Pfate XXIIIbthere is a - great increase in size of the air-space;this occurred in conjunction with an acute attackof bronchitis and closely simulated a tensionpneumothorax; the patient was admitted tohospital as an emergency. Presumably the cough-

ing from the bronchitis had caused the cyst tobecome violently distended. In this case the cystseemed to arise in connexion with the upper lobe;at operatioti it was found to spring from the lowestlateral fringe of the lower lobe. Plate XXIIIcshows another example of what might be diagnosedas a tension pneumothorax; this patient also wasextremely dyspnoeic. At operation he was foundto have two enormous bullous cysts; the lungexpanded to fill the chest immediately they wereremoved.

This recognition of giant bullae or cysts is ofgreat practical importance. It has already beenmentioned that the radiographs of some examplesreported in the literature as bilateral chronicpneumothorax strongly suggest bilateral apicalbullae. This is so in a case recorded by Lewis(1933); in the case report it is stated that needleswere inserted in both " pneumothoraces" to ob-serve the pressures, and the patient died suddenlyand unexpectedly ten hours later; there was noautopsy. It is almost certain that an autopsywould have revealed bilateral apical bullae, eachwith a puncture and accompanied by tensionpneumothorax on one or both sides. In fact deathin such a case should not be " unexpected ": if aneedle is inserted, a tension pneumothorax isalmost certain to follow. The cyst itself usuallydevelops as a result of a check-valve action, andif it is punctured it may continue to deliver airinto the pleura with each breath, unless the pleurahappens to be completely obliterated. This dangerassociated with puncture of a bullous cyst is by nomeans generally appreciated. If it is decided thatdiagnostic puncture is needed, provision should bemade for treatment of the tension pneumothoraxthat may follow. It is even better, in a case ofdoubt, to be prepared to do a thoracotomy at once,if radiographs show that a pneumothorax has beencaused, without waiting for a severe tensionpneumothorax to develop; an event which may behazardous in the extreme to many of these patientsand may be slow to respond even to prompt treat-ment with an indwelling needle and a water-seal.

It is most important to inspect the plain radio-graphs, in a case of true pneumothorax, for evi-dence of cystic disease or bullae in one or bothlungs. Failure to recognize the presence of largeremovable cysts may result in prolonged and need-less invalidism. Plate XXIVa shows a small apicalbulla in a case of recurrent bilateral pneumothoraxin a young man ; the bulla was confirmed at thora-coscopy. Experience shows that it is exceptionalto detect small apical bullae on plain radiographyalthough they may more often be demonstrated ontomography. Tomography may also reveal other-

101



j.com/


ownloaded from


1R. C. BROCK

wise hidden cystic disease or bullae in one or bothlungs; it should not be omitted although it-maynot help. The plain radiographs should also beinspected carefully for evidence of apical scarringor of old calcified foci, both of which may besignificant and causal (Plate XXI-Vb).

Bronchograms are often revealing and, althoughperhaps not necessary as a routine, are usuallydesirable. On three occasions they have revealedquite unsuspected bronchiectasis.

Plate XXVa and b shows such a case, a young

man of 21 who had had four attacks of spontan-eous pneumothorax on the right side and three on

the left; he had virtually no cough or sputum. Thevery shrunken atelectatic right lower lobe was

overlooked in the plain films. It is a fair assump-

tion that the considerable compensatory emphy-sema of the right upper lobe may have renderedit more liable to give way and leak; this does not,however, explain the attacks on the left side, whichshowed no bronchiectasis. Bronchograms mayalso be of help, in differentiating between a

pneumothorax and a giant bulla; in one case theygave conclusive indication of a pneumothorax;conversely the information they have given hasbeen at times uncertain.

Bronchoscopy is not needed as a routine but issometimes helpful. In one case considerable rota-tion and obstruction of the main bronchus wascaused by a large cyst moving about inside a hugetension pneumothorax -with a large mediastinalhernia.

Intrapleural pressure readings may be valuablein demonstrating conclusively the presence of a

fistula, as shown by rapid return of the pressuresto the previous reading after withdrawal of air.Most cases of chronic pneumothorax will have hadrepeated aspirations of air and pressure readings inan attempt to secure lung expansion. Estimationof the oxygen, nitrogen, and carbon dioxide con-tent of the pleural air may also indicate a fistula;this has been done in a few of my patients.The most valuable and often conclusive exam-

ination is thoracoscopy, and this should never beomitted; in a bilateral case it may be permissibleto waive the examination on one of the sides. Itis a sound principle throughout the whole of medi-cine and surgery to examine by direct inspectionwhenever possible; here we have an example ofpleural disease, readily amenable to inspection, andyet thoracoscopy has been used only relativelyrecently and has still by no means become a

routine. With its aid it is often possible to makea complete and certain diagnosis at once,; the infor-mation thus given may be negative but will never-theless be useful. Moreover, in many cases it is

the medium of successful treatment in that the firststeps towards an artificial pleurodesis can be taken.The information afforded by thoracoscopy will bedescribed in the later parts of this paper. PlateXXVc shows a large bulla with a perforation dis-closed at thoracoscopy.

Types of case.-Analysis of seventy-one casesinvestigated shows that about nine groups or types

TABLE IVSPONTANEOUS PNEUMOTHORAX: AETIOLOGY

Type of case Number

1 Generalized emphysema, 12 casesBullous emphysema 13 ,

Total . .. .. .. 25la Asthma and bronchitis with emphysema 82 Large solitary bullae or cystic disease.. 113 Diffuse polycystic disease .. .. 34 Small bullae (mostly apical) .. .. 155 Apical scar .. .. .. .. 66 Leak or tear seen .. .. .. 47 Areas of "cuckoo-spit".. .. .. 48 -Various.. .. .. . .. 4

(Tooth extraction 1Staphylococcal abscess 1Drainage ofempyema 1Tuberculous pleurisy 1)

9 No cause .. .. .. .. 6

can be identified, although there is some inter-dependence between them (Table IV). Fig. 1illustrates these groups diagrammatically.

Although some of these types may have a com-mon mechanical factor they differ pathologicallyand illustrate the rich variety of disease conditionsof the lungs represented in chronic and' recurrentspontaneous pneumothorax; the condition, farfrom being an entity, is a natural event in aconsiderable number of lung diseases. It will besimpler to consider the significance of these variousconditions when discussing the mode of productionof the pneumothorax and the reasons for chroni-city and recurrence.Mechanics of production and chronicity of

pneumothorax.-A number of authors have, con-tributed interesting, useful, and informative dis-cussions on the mode of production of "pneumo-thorax simplex": notably Kjaergaard (1932 and1935), Perry (1939), Ornstein and Lercher (1941-2),and authors who have written on isolated cases orthe comparatively rare cases that come to autopsy.The easiest and most obvious explanation, and

the one most generally accepted, is rupture of asmall emphysematous bulla or cyst. There have.been sufficient post-mortem demonstrations ofruptured bullae in fatal cases to make it certain

-102



j.com/


ownloaded from



EMPHYSEMA

POLYCYSTIC LUNG

I CaseAPICAL BULLA

15 Cases

BULLOUS EMPHYSEMA

6 Cases

CYSTIC LUNG

6 Cases1 4 -

TORN LUNG

2 Cases

" CUCKOO-SPIT "

4 CasesFIG. 1

5 9

I

1-03



j.com/


ownloaded from


R. C. BROCK

that they can be the cause. A tear or rupture hasbeen observed, in only four of my -cases, but asmall bulla of the characteristic type has beenobserved in fifteen cases, a number sufficientlylarge to be significant. Kjaergaard (1932) has dis-cussed the formation of these small bullae and thevalvular mechanism that may be demonstrated intheir basal parts. Two illustrations in his papershow this valvular mechanism in cysts taken frompost-mortem specimens in which the bullae werefound accidentally and had not ruptured to causea pneumothorax. It is easy to see how such a bullamay be progressively distended by each act ofrespiration until it bursts; this bursting can occurwhile the patient is at rest, as indeed happens; therupture is also liable to be precipitated by any actof straining or violent exercise. According toKjaergaard the credit for first describing thismechanism belongs to Fischer (1922), and to hispupil Hayashi (1915), who described three cases.Orth had already pointed out that there must be anobstruction to the outlet of air from these bullaeas it is difficult or impossible to force the air outof them by pressure. Hayashi demonstrated at thebase of a bulla a valve-like structure which wasformed of deformed atrophic lung tissue.

Kjaergaard mentions two types, in one of whichthere is no scar-tissue in the base of the bulla,which abuts directly on healthy or emphysematousalveqli, and another in which the bulla rests on alayer of scar tissue.

This mechanism of production and of rupture ofthe bulla also offers a simple explanation of whythe pneumothorax may remain chronic or mayrecur; indeed, when one considers the valvemechanism it is strange that spontaneous expan-sion of the lung occurs so often.The isolated apical bullae or cluster of small

bullae in an otherwise normal lung are generallyaccepted as being due to healing of an old tuber-culous lesion. In some cases (six in this series) ascar may be present without any demonstrable orvisible bulla, and it is a matter of assumption thatthe escape of air has come from a smaller emphy-sematous area or minute bulla formed secondary tothis scar tissue. This is a very reasonable assump-tion and also introduces the question that will bereferred to again later, whether, even when an un-ruptured bulla of moderate size is observed, it isthe actual cause of the leak, since the latter maywell have occurred from a nearby, easily over-looked, smaller area of dilated alveoli.These areas of so-called "local emphysema"

lead naturally to a consideration of true emphy-sema as a cause of pneumothorax. A number ofauthors have referred to the fact that pneu-

mothorax is a relatively rare complication ofemphysema. Kjaergaard -has suggested that thecommunication between the bullae and theadjacent alveoli in emphysema is not usuallyvalvular but is wide and free and that this mayexplain the rarity of rupture. Nevertheless, ifrupture does occur it would be anticipated thatit would be a more serious condition with a greatertendency to chronicity and recurrence. This sup-position receives support from my cases in whichgeneralized emphysema was present in twelveinstances; it is more understandable and signifi-cant that gross bullous emphysema was present ina further thirteen, making a total of twenty-fivecases. The differing role of emphysema inpneumothorax simplex and in the chronic and re-current groups is further reflected in the greaterrepresentation in the higher age groups (Table III).The fact that eight of these patients also sufferedfrom asthma and chronic bronchitis is of furthersignificance, and it is also noteworthy that oneof these patients was aged 17, two were in thesecond decade, and three were in the thirties. Ina bare inspection of the age incidence they wouldnot be differentiated from the cases of benignpneumothorax in young adults.When gross bullous disorganization of the lung

accompanies generalized emphysema, the natureof the bullae is certain. There is less certaintyof the nature of the large bullae or cysts that mayoccur in otherwise healthy lungs. It is easy toclass them all as congenital, and there is littledoubt that many of them are. If other congenitalanomalies, such as aberrant rib formation, areobserved, this gives strong support to the congeni-tal origin. In others there is no such ready con-firmation and the association of irregular clustersof bullae of various sizes with dense interlacingbands of scar tissue and dense pleural plaquesstrongly suggests that they are the end resultof some severe destructive inflammatory process inthe lungs. They may even be gross examples ofpost-tuberculous scarring such as is commonlyseen in smaller foci at the apex. Multiple cystsoccurring without fibrosis are more likely to betruly congenital.

These cysts or bullae occur in either the upperor the lower lobe. Even if the cysts themselvesare not congenital they may arise from a con-genital weakness of structure of the lungs thatfavours their production and therefore their con-tinued increase in size and final rupture. In oneof my patients, a woman aged 23, a number ofclusters of very thin-walled bullous cysts werepresent in the lower lobe, and at .thoracotomy avery striking phenomenon was seen; the cysts

104



j.com/


ownloaded from



were not tense but were lax and collapsed com-pletely on expiration and became distended withinspiration. It required little imagination to seehow a sudden cough, strain, or violent exertioncould lead to their rupture.

Allison (1947) has recently pointed out that cer-tain of the giant cysts or bullae are caused by adefect in the wall of a bronchus of moderate sizeand that the hole can be seen in the floor of thecyst.

In addition to single congenital cysts, diffusecystic or polycystic disease may be present; thismay be unilateral or bilateral. Such was the case inthe patient whose radiographs are shown in PlateXXVI. This patient came into hospital with a-pneumothorax on the right side (Plate XXVIa);it was the eighth on this side, and he had had twoon the left side. In spite of the small cystic natureof the disease, bronchograms showed no bronchialdilatation, and at thoracoscopy the lung could beseen studded with innumerable thin-walled cystsof varying size, but all small. Clearly the cysticdegeneration or malformation affected chiefly thealveolar tissue or bronchioli.

Artificial pleurisy was induced with silver nitrateon the right side, and while this was proceedinghe developed yet another spontaneous pneu-mothorax on the left side, making eleven attacksin all. This was a tension pneumothorax, andonly prompt insertion of a needle attached to anunder-water seal saved his life (Plate XXVIb).After the right lung had expanded he sufferedyet another attack on the left side and this wasthen treated with silver nitrate. The conditionafter bilateral pleurodesis can be seen in PlateXXVIc; it should be noted that some larger cystsare present at the left base. A most interestingand significant phenomenon has since beenobserved on the left side. The large- cystic con-dition at the left base increased in size, and onone occasion when he-complained of distress andshortness of breath a radiograph (Plate XXVId)showed gross distension of the cysts in the leftlower lobe. In a later radiograph these becamesmaller, and since then they have contracted stillmore so that the present appearance resembles thatseen in Plate XXVIc. This is clear proof that thecysts were responsible for the recurrent pneumo-thoraces, and suggests that these big basal cystswere the cause of the left tension pneumothorax.Once his pleura had been obliterated they wereunable to rupture, and so they distended to ahuge size and presumably may do so again. Theformper spontaneous pneumothoraces may haveacted as a safety-valve and spared the lungs. Itmay become necessary to operate and resect this

I

cystic area at the left base in order to spare himprogressive compression of the relatively healthylung above.

This progressive distension after pleurodesis hasbeen seen in two other cases in this series. Onewas a patient who had a small apical bulla about2 cm. in diameter. At a recent follow-up this wasnoticed to have grown to some 6 cm. in diameter.Presumably the pleurodesis spared him a succes-sion of recurrent pneumothoraces; the bulla is asyet too small to be of significance. In the otherpatient, a woman aged 30, who had had threeattacks on the right side over a period of twelveyears and whose last attack had lasted six months,thoracoscopy revealed a moderately large thin-walled cyst in the right lower lobe. After somehesitation it was decided to try the effect of pleuro-desis rather than lobectomy. After pleurodesisthis cyst not only persisted but grew larger. Asshe complained of pain and discomfort and ofrecurrent pyrexial attacks with bronchitis, the lobewas removed. She has since been quite well.These experiences and observations indicate the

natural history of the condition and the fact thatradical treatment will usually be needed whena cyst is present. Kjaergaard (1935) amplifies hisearlier paper by describing two cases studied postmortem in which a pneumothorax was caused byrupture of congenital valve vesicles on the surfaceof the lungs; in both cases cysts were presentdeep in the lungs as well as on the surface. Thecondition was comparable with the case illustratedin Plate XXVI. Of great interest in connexionwith cystic disease is the occurrence 6f familialpneumothorax, a number of examples of whichare recorded in the literature and are detailed byKjaergaard. Gotzsche (1933), for instance, des-cribes a remarkable series of five in the samefamily. Bachmann (1940) reports the occurrence.in a father and a daughter. Two patients in myseries had near relatives (in one case a sister)who had had a spontaneous pneumothorax. Theassumption is that the attacks are due to ruptureof hereditary congenital lung cysts.

It is surprising that a tear or hole was seen inonly four cases, but in many there were sufficientadhesions present to render complete examinationimpossible. These adhesions themselves may infact be a cause of chronicity or recurrence, inthat they may cause the lung7to tear as it collapsesand the tear is then held open. This has actuallybeen observed at thoracoscopy; and in one case,in spite of the application of silver nitrate, thelung would not expand; the adhesion was accord-ingly divided and the lung at once expanded. If acontrolling adhesion is seen next to an area of

105



j.com/


ownloaded from


R. C. BROCK

scarring or bullous change, it should'be dividedso as to allow a tear or rent to become relaxedand to heal.

In four of my cases I was able to observe aremarkable condition that has not been describedbefore, although at least one author has hinted atit. In none.of these four could bullous or cysticchange be seen either radiographically or at thora-coscopy; one patient (a man aged 58) had general-ized emphysema; another (a man aged 34) hadbronchiectasis and mild emphysema; a man aged32 had no other evidence of abnormality in thelungs; and the fourth (a man aged 35) had emphy-sema, asthma, and bronchitis. The most carefulinspection of the pleura at thoracoscopy failed toshow any rupture or leak, but a number of scat-tered foci could be seen in which a few tiny airbubbles (each less than a millimetre in diameter)could be seen on the lung. It is not uncommon tosee small'air bubbles on the moist lung surface ina pneumothorax, but when the patient was in-structed to hold his nose and blow up his cheeksthese clusters could be observed time and timeagain to increase not only in size but also in thenumber of bubbles. Each cluster would grow soas to form a tiny area resembling the froth oneoften sees in plants in early summer and, which isknown as "cuckoo-spit." I have applied thisname to the condition. These areas are causedby a minute alveolar leak; this leak is certainlyin some cases only into the subpleural tissues; but,with an unsupported pleura, rupture can occurand an intrapleural leak follows. These foci arenumerous and probably changing, and the observa-tion strongly suggests that in certain cases ofchronic and of recurrent pneumothorax in whichno definite or gross leak, such as a ruptured bulla,can be demonstrated, the mechanisri of escape ofair is multifocal, changing from time to time, andit is so- minute as to escape detection except withthe closest inspection with a thoracoscope heldalmost up to the lung surface. It is doubtful if themechanism could be'observed in the post-mortemroom, and it may explain cases such as those des-cribed by Rolleston (1900), Priest (1937) and othersin which death occurred from pneumothorax andyet a most careful inspection failed to reveal anyabnormality in the lungs.

This mechanism may also be the one actuallyresponsible even when a large bulla or cyst is seen,because I have often observed that the bullae aretense and intact, not collapsed and leaking. Pitt(1900), in addition to describing the post-mortemfinding of a torn adhesion attached to a torn bullain a fatal case, mentions another fatal case: thatof a man aged 54 who had chronic bronchitis and

asthma and whose lungs contained many bullae,but no actual hole could be demonstrated eventhough the bronchi were carefully distended withwater.

Schmincke (1928) reports an autopsy on a youngman who died of bilateral spontaneous pneumo-thorax and had many emphysematous b!ebs inboth lungs. He believed that a peripheral zoneof embryonal tissue persisted which did not differ-entiate into alveoli; he describes and depicts suchan embryonal area which in some areas did notprogress to its final development into maturealveoli. This observation seems to support theconclusion I have come to that in certain cases theincidence of the spontaneous pneumothorax is duenot to any gross disease, either focal or general inthe lungs, but to a congenital (or possibly acquired)defect of quality that renders the pleura liable toleak or rupture easily in many isolated, minute,and changing places. In other words these patientshave " leaky lungs " or " porous pleura."

This would seem to be the only explanation forthe six cases in my series in which no abnormalitycould be observed or demonstrated either radio-graphically or at thoracoscopy. In one (a womanaged 35 who had three attacks on the right side,four on the left, and also a chronic pneumothoraxon the right lasting six months until obliterated bypleurodesis) I inspected the pleura four times witha thoracoscope and tried a variety of ways, such ascoughing, straining, and applying constant suction,to try and demonstrate a leak. The lung seemedperfectly healthy.That a defect in the quality or strength of the

lung tissue probably exists is shown by anothercase, that of a woman aged 44 who had had apneumothorax for four months. Fig. 2 shows theappearance of bullous or cystic change observedat thoracoscopy. A small hole, which may havebeen caused by the local anaesthetic needle, wasseen in the upper lobe. As the lung was beingpainted with a swab moistened with silver nitratethe swab-holder slipped and fell gently a distanceof 2 or 3 cm. on to the surface of the lung;another hole appeared at once. No normal lungwould rupture in this way, and it is beyond reason-able doubt that the lung tissue, including thepleura, was extremely fragile. Incidentally thelung was very slow to expand in this case; severalapplications of silver nitrate were needed, and ittook six months to fill the chest.

This- defect in the tensile strength of the lungscan well be due to imperfection of the elastictissue, which may vary considerably in quantityand is noticeably absent or defective in emphy-sema. The formation of bullae is probably closely

106



j.com/


ownloaded from


RECIJRRENT AND CHRONIC SPONTANEOUS PNEUMOTHORAX

FIG. 2

connected with rupture of the elastic supportingand strengthening layer. Zahn (1891) measuredthe combined thickness of the alveolar wall andpleura and showed that whereas in the normal itis 0.13 to 0.24 mm., in emphysematous lungs it maybe as little as 0.05 or even 0.03 mm.; he suggestedthat this atrophy of the visceral pleura might beone cause of pneumnothorax.

Summary of the causation of pneumothorax.-The cause of pneumothorax, considered with par-ticular reference to recurrence and chronicity, maylie in any one of several anatomical sites. It maybe within thr pleura, as when an adhesion causesthe lung to tear and holds the rent open. It may bepleural as when the elastic supporting layer isdefective in quality or quantity. It may be sub-pleural, when there is rupture associated withbullae arising either in connexion with an old scaror with destructive emphysema; subpleural leak-age may also occur from defective alveoli, possiblycongenitally maldeveloped. The major diseasemay lie in the alveoli throughout the lung, as ingeneralized emphysema, or it may be in the bron-chioli or smaller bronchi as in congenital cysticconditions. Finally the larger bronchi may beresponsible, as in the case of giant valvular bullaeresulting from defects in the wall of the bronchusitself. Chronic diseases of the bronchi, such aschronic bronchitis, asthma, and bronchiectasis, alsoplay their part.

TREATMENTIn the past the treatment of chronic and recur-

rent pneumothorax has been governed largely bylaissez faire; or perhaps, equally unsatisfactory,

a mistaken assumption of an under-lying tuberculous basis has resulted inthe patient being condemned to wast-ing many weary months in bed. Evento-day patients stand in peril of beingsent to a sanatorium. Conservativeexpectant management of a first attackof "pneumothorax simplex" issound: the lung usually expandsquickly and safely in its own time. Ifthe rate of expansion is slow and pro-longed and the condition passes intothe stage of chronicity, or if relapsesoccur, then there can be no questionthat active treatment is needed. It iswasteful of the patient's time and ofvaluable beds to keep him immobilewhen the cause of his disability restson purely mechanical factors thatcan be corrected. I appeal, therefore,

for active treatment of these cases.All good treatment is founded on a knowledge

of the cause of the disease, and often rests largelyupon a consideration of the morbid anatomicalchanges present. It follows that treatment shouldnot be empirical, for the analysis of these casesshows that a number of different causes may beacting. Each case must be fully investigated inthe manner already detailed, and tho&hcoscopy inparticular should never- be omitted.The most important first step is the recognition

or exclusion of localized large cystic or bullousdisease, as distinct from the simple small apicalbulla. It is useless to persist in conservative treat-ment if a large cyst or bulla is present; it is alsowrong to rely upon pleurodesis, because the cystremains and may increase in size. The correctthing to do is to excise the cyst or cysts either bylobectomy, or, if they are conveniently sited, bylocal resection. I have had to do this in eightcases, which, so far as I know, is the largestnumber treated in this way.

Bigger (Horsley and Bigger, 1937) describes acase in which he resected a small, leakingemphysematous bulla in a case of chronicpneumothorax. Sycamore (1936) and later Tysonand Crandall (1941) report a single case of resec-tion of a small, unruptured bulla in a man aged30 who had had five attacks of spontaneouspneumothorax. Hoyer and Clagett (1946) per-formed middle lobectomy for cystic diseaseassociated with chronic pneumothorax of sixmonths' duration in a man aged 41. My own firstcase was operated upon by right lower lobectomyin May, 1939, and is so rich in features of practical

107



j.com/


ownloaded from


R. C. BROCK

value and interest that it deserves recording inmore detail.

Mrs. S. M., aged 30, was first taken ill in February,1934, with an attack of pain above the right breastfor three weeks; a few months later she had a secondattack, and in December, 1935, a third one, in whichshe was dyspnoeic. The dyspnoea persisted, and twomonths later she had a further relapse and the pre-sence of a spontaneous pneumothorax was recognizedfor the first time. A diagnosis of tuberculosis wasmade, and a four months' pregnancy terminated; shewas kept in bed until May, by which time the rightlung had expanded. In July, 1937, she had anotherslight attack of pain, and in April, 1938, when she waseight months pregnant, she had yet a further attackof severe dyspnoea with fever and cough. Thedyspnoea persisted and was very severe when herchild was born in June. After this she continuedan invalid existence with a troublesome cough andcrippling dyspnoea; her weight fell by a stone and ahalf.

In April, 1939, she was seen by Dr. C. H. Hoyle,who noted that she was so short of breath as to beunable to converse for long at a time. Her vitalcapacity was 1,000 c.cm. and she had a right totalpneumothorax with a large mediastinal hernia.Dr. Hoyle recognized what had been consistentlyoverlooked up till then, that the outline of a thin-walled bulla or cyst could be made out in the lowerpart of the chest (Plate XXVII). This could be seento move medially and laterally on inspiration andexpiration, or with change of position.At thoracoscopy the presence of a large thin-walled

bullous cyst was confirmed; it sprang from the lowerlobe and had a white, opaque, smooth surface with ascar-like patch near the centre. Nearby two smallerbullae could be seen.At bronchoscopy the carina was deformed to a

Z-shape and the right bronchus so flattened and dis-torted, as if compressed and rotated forwards anddownwards to the left, that the bronchoscope couldnot be passed into it.Thoracotomy was performed in May, 1939, and a

large tri-lobed cystic condition was displayed in thelower lobe; the huge mediastinal hernia allowed thecyst to fall forwards medially so that the right bron-chus was twisted through a right angle (Plate XXVIII).A condition of partial volvulus was thus produced,and this explained the bronchoscopic findings. Theupper and middle lobes did not aerate, even withconsiderable pressure by the anaesthetist. The lowerlobe was removed with the contained cysts, and assoon as this was done the partial volvulus was cor-rected and the upper and middle lobes at once beganto aerate (Plate XXIX). A small cyst about 2.5 cm.in diameter attached to the hilar region of the upperlobe was removed. The main cystic mass is shownin Plate XXX. She made an uninterrupted recoveryfrom the operation, although the upper and middlelobes seemed to be unable to fill the whole hemi-

thorax; for, though they expanded to the chest wall,some mediastinal shift to the right occurred and haspersisted.The change in her condition after this operation

was dramatic; she soon resumed a busy normal life,and within five months regained her lost weight. Twoyears later she went through a normal pregnancy and-delivery. During the war she looked after threeevacuated children in addition to her own family.She was also able to resume singing, which had beenquite impossible while she had her pneumothorax.

She continued well until September 10, 1947, whenshe experienced pain in her left chest; later dyspnoeaoccurred and she was found to have a spontaneouspneumothorax on the left side. When the radio-graphs of her earlier illness were being reviewed it wasnoticed, for the first time, that she had had a smallpneumothorax on the left side during convalescencefrom her operation. This had been quite symptom-less, but it emphasizes how easily a shallow pneumo-thorax can be missed unless it is deliberately lookedfor; this should always be done on the contralateralside when a spontaneous pneumothorax is present.Thoracoscopy revealed two small bullae, one

attached to the upper lobe and about 1 cm. indiameter; the other attached to the lingula andwith a haemorrhagic wall. Both were painted withswabs of 20 per cent silver nitrate solution, whichwas also applied to the surface of the upper and lowerlobes. An obliterative pleuritis was thus set up andthe lung expanded after two aspirations of air andfluid.

This case has many points of interest; the longhistory of recurrent attacks (at least five in all)over a period of five years had resulted in muchsuffering and disability and had brought thepatient to a state of invalidism. Because of anerroneous diagnosis of tuberculosis, her child hadbeen sacrificed. The policy of laissez faire inregard to the pneumothorax is typical. She hadbeen seen by a number of physicians over severalyears, and the presence of the cyst had been over-looked in the radiographs: this would not havemattered if she had been submitted to thoraco-scopy, for the cysts must then have been revealed.The successful treatment by lobectomy is of in-terest for, so far as I know, this is the first occasionon which this operation was used in the treatmentof chronic or recurrent pneumothorax. Hoyerand Clagett's (1946) case was not operated onuntil 1945, by which time I had performed lobec-tomy on six more such cases. Finally, there is thedevelopment of the contralateral spontaneouspneumothorax and its treatment by the inductionof a chemical pleuritis.

I have, in addition to these eight cases, per-formed lobectomy or local excision of giant bullaeor cysts in a further six cases without pneumo-

108



j.com/


ownloaded from



thorax, although several of these simulated chronicpneumothorax. It would seem that the very largecysts are not so liable to leak or rupture as thesmaller ones, possibly because they receive sup-port from their firm apposition to the chest wall.Their very size protects them.When the presence of a cyst or large bulla has

been excluded the logical treatment would appearto be pleurodesis; that is, obliteration of the pleuralspace by some artificial means so as to preventrelapse, or to achieve expansion of the lung in achromnc case.

Table V shows the methods of treatment usedin this series. For various reasons, such as refusal,

TABLE VSPONTANEOUS PNEUMOTHORAX: TREATMENT

Lobectomy.. .. .. .. .. 8Pleurodesis .. .. .. 53 patients

11 bilateral .. .. .. 64 sidesSilver nitrate .. .. .. 52 patients

9 bilateral .. .. .. 61 sidesPoudrage .. .. .. 3 sides

Cauterization ofadhesion .. .. 2 patientsNo treatment .. .. .. .. 12 patients

no treatment was given in twelve patients. Intwo cases adhesions were divided. In fifty-threepatients pleurodesis was done; eleven of thesewere bilateral cases, and so sixty-four sides weretreated in this way. Poudrage (using iodized talcpowder, after the method of Bethune, 1935) wasused in three cases, but in fifty-two silver nitratewas used. Two patients had poudrage on one sideand silver nitrate on the other and were emphaticthat the silver nitrate was less disturbing. It hasalso the great advantage that after the initialapplication at -thoracoscopy further treatmentscan be given very simply by injection while thepatient is in bed. Moreover, in bilateral cases itis not obligatory to thoracoscope both sides anda simple intrapleural injection can then be givenon the second side. Poudrage requires thoraco-scopic application in the theatre.

I first used a solution -of silver nitrate to pro-duce pleurodesis in January, 1936, in a case ofchronic pneumothorax. Adams (1933) describedits sclerosing action when applied to bronchialfistulae, and it occurred to me that it might beequally effective in closing the hypotheticalalveolar-pleural fistula. Actually no hole couldbe seen, but the lung was swabbed with a solutionof 20 per cent silver nitrate and a progressiveobliterative pleurisy followed. As a result of thisI used it in other cases of chronic and recurrent

I*

pneumothorax and also in producing pleurodesisbefore lobectomy (Brock, 1942). I have sincelearnt that Morlock described its use in a caseof chronic pneumothorax in 1933; he used 5 ml.of a 1 per cent solution.

According to Hennell and Steinberg (1939), theidea of induction of chemical pleuritis in cases ofchronic or recurrent pneumothorax first occurredto Spengler (1901). In 1906 Spengler suggestedusing 0.5 per cent silver nitrate; in 1919 he advisedusing 30 per cent glucose solution for the samepurpose; but it was not until 1923 that he reportedhis first clinical case. In this he used 30 ml. ofa 30 per cent solution of glucose in repeated dosesand obtained a cure in four months. In his reporthe suggested that either larger amounts of 30 percent (up to 100 ml.) or smaller amounts of 50 percent should be used. Kenner (1932) succeededwith 20 ml. of 0.5 per cent silver nitrate afterfailing with 30 per cent glucose.Harvey (1938) reports two cases of spontaneous

pneumothorax in asthmatics; in one the pneumo-thorax had been present for three months and wastreated with 40 ml. of 50 per cent glucose, whichwas administered twice; on each occasion severepain and a febrile reaction was caused.

Hennell and Steinberg report five cases ofchronic or of recurrent pneumothorax treated bythe induction of a chemical pleuritis; this is thelongest series recorded up till now. Three of theirpatients were treated with iodized poppy seed oil;one was treated with a hypertonic glucose solu-tion, but this failed, and oil was then used; thefifth was successfully treated with hypertonicglucose solution. They state that it is best to usea 67 per cent (saturated) solution in doses of 50or 60 ml. A 50 per cent solution- may be inade-quate, but the saturated solution is more likelyto be effective and is just as safe. They have usedas little as 25 ml. and as much as 60 ml. withequally satisfactory results; the size of - thepneumothorax has been the rough guide to theamount used; 50 ml. was the average dose. Whena glucose solution failed, iodized oil succeeded,but with a much more acute reaction, with moreconstitutional symptoms, and more fluid.

Hennell and-Steinberg state that their review ofthe literature disclosed reports of no more thantwelve cases of spontaneous pneiimothorax treatedby induction of chemical pleuritis. The largestsingle series before their five cases comprised threecases and was reported by Schott (1934, 1935),who used 0.5 ml. of oil of turpentine in one caseand a 50 per cent solution of glucose in two cases.

Other oily preparations have been used in iso-lated cases by various workers; gomenol in olive

109



j.com/


ownloaded from


R. C. BROCK

oil (Chandler, 1939) or merely plain olive oil hassucceeded. I have used gomenol in olive oil with-out success on two patients and have then suc-ceeded with silver nitrate. So far no author hasdescribed a sufficiently comprehensive series treatedwith simple oil to prove the claims of this sub-stance. Iodized oil, as used by Hennell and Stein-berg, seems to provoke just as severe and painfula reaction as silver nitrate. The effectiveness ofhypertonic glucose solutions is also in some doubt;several injections are needed, and they are certainlypainful and disturbing (Kjaergaard); some authorsstate that morphia should be given before theglucose injection.Another alternative is the use of venous blood;

I do not know what quantity is needed to ensuretotal obliteration, but Watson and Robertson(1928) record a case in which 300 -ml. of blood wasused.While the method of pleurodesis by a chemical

irritant is undoubtedly practical and efficient, Ido not press the claims of silver nitrate againstother substances; but I have found it so effectivein treatment that I have been loth to change. Thepain and general reaction caused are the oneserious objection to its use. I am planning to trycopper sulphate as an alternative, but to producea satisfactory obliterative pleurisy a fairly highdegree of acute pleural inflammation must beinvoked, and this will inevitably cause pain anddiscomfort. Hypertonic glucose solutions, iodizedoil, and blood also cause similar severe pain andgeneral disturbance. The amount of pain causedby silver nitrate varies; it may be severe andrequire morphia for its control; it may be rela-tively slight and be relieved by a simple analgesicsuch as aspirin. When the silver nitrate is applieddirectly to the lung under thoracoscopic controlno pain is caused unless the solution comes intocontact with the parietal pleura. If it is injected,pain is more likely to occur.For injection into the pleura I use 5 or 10

minims of a 10 per cent solution; for direct appli-cation to the lung at thoracoscopy three or foursmall swabs are used, dipped in a 20 per centsolution. A febrile reaction follows, rising andfalling over about five days and reaching a maxi-mum of 1000 to 1010 F.; an effusion of varyingsize accompanies it. Sometimes a superaddedspontaneous pneumothorax.occurs, due to super-ficial necrosis, but this has never caused anyanxiety or been more than temporary. The resultshave been extremely satisfactory in all the 61 cases(total number of sides) in which it has been used.In only two has the rate of expansion been slow,but further injections and patience have succeeded.

The method must be used intelligently and care-fully. Many people seem to imagine that all thatis necessary is to inject the solution into the pleuraand do little or nothing else. First of all it isnecessary to ensure that active silver nitrate reachesthe pleura, and this is done by inserting a needleand then injecting the solution from an all-glasssyringe; a glass-metal syringe must not be used.Next, air and liquid must be aspirated intelligentlyto get the pleural surfaces in apposition so thatthey will adhere while they are inflamed. If thepleural surfaces are left separated by liquid, theymay become fibrous and inert by the time theliquid has been absorbed. Moreover it is askingtoo much to expect total obliteration of the wholepleura with one injection of silver nitrate ; this mayoccur, but it is uncertain. After four weeks anattempt should be made to reinduce a pneumo-thorax, and if a pocket is obtained a further injec-tion should be made. Unless the pleura is testedin this way relapse may follow.

In cases of chronic pneumothorax it is usuallynot possible to aspirate liquid to achieve rapidlung expansion, and slower obliteration must beexpected. In such cases it is best to achieve asummation of effect by repeating the injection atten- or twenty-day intervals, aspirating air andliquid meanwhile, until obliteration is achieved.Some of my colleagues tell me they have greatlyaccelerated the rate of lung expansion by intro-ducing a small tube through the thoracoscopiccannula and leaving it in place for twenty-fouror forty-eight hours so that continuous suction canbe applied.

FOLLOW-UPOf the fifty-nine patients in whom active treat-

ment was carried out, eight had lobectomy or localresection of a large cyst (one case), and fifty-threewere treated by pleurodesis; two of the lobectomypatients underwent pleurodesis as well (one bnthe lobectomy side, one on the opposite side), andthis explains the apparent discrepancy in thefigures. One of the lobectomy patients died threemonths after operation from a tension spontaneouspneumothorax (which was unfortunately not diag-nosed) on the opposite side; the remaining sevenpatients are alive and well. The patients treatedby pleurodesis have been followed up to date, withthe exception of eight who have been lost sight ofafter periods ranging from six months to two years.In all these eight cases the state of the pleura hadbeen satisfactorily tested after pleurodesis in anattempt to confirm obliteration; one of these, aman who had had bilateral pneumothoraces andwho refused to allow the left side to be obliterated,

410



j.com/


ownloaded from



had a further attack on the left side during thetwo years he was observed; he has not been seensince 1942. None of the remaining forty-fivepatients, all of whom have been followed up todate, has had a relapse. Four have died; one, aboy aged 12 years at the time of his chronic bila-teral pneumothoraces due to congenital polycysticdisease, died of acute pulmonary tuberculosis twoyears after the onset of the pneumothoraces. Theremaining three patients died from broncho-pneumonia, or acute bronchitis and cardiac failureat four years, eighteen months, and six monthsafter the pleurodesis. Two of these had sufferedfor years from asthma, chronic bronchitis, andemphysema, the third suffered from chronic bron-chitis and emphysema.The treatment of a disabling chronic pneumo-

thorax or a troublesome relapsing or recurrentpneumothorax by the pleurodesis given by silvernitrate is one of the most satisfactory and effectivemethods we have at our disposal, and I can con-fidently recommend it as the method of choiceonce gross cystic disease has been excluded. Itis almost entirely free from risk ; the principle ofprod4ction of a chemical pleurisy is a sound one;the major objection of the painful reaction thatit can cause may well be answered by the intro-duction of a kinder pleural irritant. The funda-mental principle, namely that of the production ofpleurodesis, will, however, remain.

I am greatly indebted to many of my colleagueswho have asked me to see and treat so many of thecases which form the basis of this paper. I owe aspecial debt of gratitude to Dr. C. H. Hoyle, who hasnot only provided fourteen of the cases in this seriesbut has also been most helpful and encouraging inhis general interest in the problems they present, andhas contributed many stimulating and valuableobservations.

REFERENCESAdams, W. E. (1933). 1. thorac. Surg., 3, 198.Allison, P. R. (1947). Thorax, 2, 169.Bachmann, H. (1940). Dis. Chest, 6, 77.

Bethune, N. (1935). J. thorac. Surg., 4, 251.Brock, R. C. (1942). Guy's Hosp. Rep., 91, 99.Brunner, A. (1921). Mitt. Grenzgeb. Med. Chir., 33,

124.Chandler, F. G. (1939). Lancet, 2, 638.Fischer, B. (1922). Z. klin. Med., 95, 1.Fishberg, A. M. (1932). "Pulmonary Tuberculosis."

Philadelphia.Goodhart, J. F. (1896). Trans. clin. Soc. Lond., 29,

109.Gotzsche, C. (1933). Ugeskr. Laeger, 95, 765.Harvey, C. (1938). Med. J. Aust., 2, 950.Hayashi, J. (1915). Frankfurter Z. Path., 16, 1.Hennell, HI., and Steinberg, M. F. (1939). Arch.

intern. Med., 63, 648.Horsley, J. S., and Bigger, I. A. (1937). " Operative

Surgery." St. Louis. Vol. 2, p. 605.Hoyer, L. P., and Clagett, 0. T. (1946). J. ihorac.

Suirg., 15, 418.Kenner, A. (1932). Beitr. Klin. Tuberk., 80, 169.Kjaergaard, H. (1932). Acta med. Scand., Supplt. 43.Kjaergaard, H. (1933). Acta med. Scand., 80, 93.Kjaergaard, H. (1935). Acta med. Scand., 86, 407.Lewis, J. (1933). Brit. med. J., 2, 779.Locke, E. A. (1929). Med. Clin. N. Amer., 13, 75.Morlock, H. V. (1933). Proc. roy. Soc. Med., 26, 525.Nikolski, S. (1912). "Uber den Spontanen Pneumo-

thorax, Giessen" (quoted by Fishberg, A. M.(1932)).

Ornstein, G. G., and Lercher, L. (1941-2). Quart.Bull. Seaview Hosp., 7, 149.

Perry, K. M. A. (1939). Quart. J. Med., n.s., 8, 1.Pitt, G. N. (1900). Trans. clin. Soc. Lond., 33, 95.Priest, R. (1937). Brit. med. J., 2, 321.Rolleston, H. D. (1900). Trans. clin. Soc. Lond., 33,

90.Schmincke, A. (1928). Beitr. path. Anat., 80, 692.Schott, E. (1934). Munch. med. Wschr., 81, 716.Schott, E. (1935). Miinch. med. Wschr., 82, 1751.Spengler, L. (1901). Z. Tuberk., 2, 27.Spengler, L. (1906). Beitr. klin. Chir., 49, 68.Spengler, L. (1919). In Brauer, L., Schroder, G., and

Blomenfeld, F. (1919). " Handbiuch der Tubercu-lose." Leipzig. Vol. 3, p. 232.

Spengler, L. (1923). Schweiz. med. Wschr., 53, 309.Sycamore, L. K. (1936). Amer. J. Roentgen., 36, 844.Tyson, M. D., and Crandall, W. B. (1941). J. thorac.

Surg., 10, 566.Watson, E. E., and Robertson, C. (1928). Arch. Surg.,

16,431.Wood, H. G. (1931). Minnesota Med., 14, 550.Zahn, F. W. (1891). Virchows Arch., 123, 197.

ill



j.com/


ownloaded from


RECURRENT AND CHRONIC SPONTANEOUS PNEUMOTHORAX* · RECURRENT AND CHRONIC SPONTANEOUS PNEUMOTHORAX...

Documents

Transcript of RECURRENT AND CHRONIC SPONTANEOUS PNEUMOTHORAX* · RECURRENT AND CHRONIC SPONTANEOUS PNEUMOTHORAX...