decortication of the in tuberculous disease · type of empyema is a contraindication, but in this...

Thorax (1955), 10, 293.

DECORTICATION OF THE LUNG IN TUBERCULOUSDISEASE

A STUDY IN 43 CASES

BY

T. SAVAGE AND H. A. FLEMINGFrom Sully Hospital, Glamorgan

(RECEIVED FOR PUBLICATION JULY 15, 1955)

Constrictive pleuritis is a not uncommon com-plication of pulmonary tuberculosis. It may occuras an effect of primary tuberculosis or, less com-monly, of tuberculosis at any stage, or as a com-plication, unhappily frequent, of artificial pneumo-thorax treatment. Once established, this conditiongives rise to considerable loss of respiratory func-tion, to danger from an unobliterated pleural spacewhich, if not already infected, is liable to becomeso, and, occasionally, to symptoms from a dis-placed mediastinum.

Established methods, other than decortication,of treating such a chronic condition are far fromsatisfactory. Frequent aspiration provides noanswer; simple drainage means a permanent" tube lifes"; an oleothorax as a rule becomesinfected. Thoracoplasty may further reducefunction, and though it may apparently obliteratethe pleural space an empyema may lie quiescentunder it only to discharge itself through a bron-chus or on to the skin years later. Operationsof the Schede type are both mutilating and func-tion-destroying.

Decortication of the lung is obviously a mostattractive form of treatment, combining a removalof this thickened pleura and empyema with anopportunity for improving the respiratory func-tion, allowing the mediastinum to return to thenormal position, and permitting resection of thediseased segments of the lung if this is indicated.

Decortication has been and is practised in manycentres, but with varying results. These have beenreported by many workers, including Watermanand Domm (1951), Weinberg and Davis (1949aand b), and Riley and Kaunitz (1951), who dealwith the unexpanded artificial pneumothorax, andGurd (1947), O'Rourke, O'Brien, and Tuttle(1949), Quinlan, Schaffner, and Hiltz (1952), Lam

(1948), and Stead, Eichenholz, and Stauss (1955),who deal with the general problem of tuberculouspleurisy. In these reports of series of cases thehistorical aspects of the subject and the detailsof the operations used are fully covered.The main reasons for the indifferent results

have been summarized in an annotation in theLancet (1954). These are the presence of old fociof tuberculous disease and fibrosis in the lung,bronchostenosis, and difficulty in removing the"peel " from the tuberculous lung, giving rise tomany air leaks which make subsequent expansionof the lung difficult and sometimes impossible.Thoracoplasty was not infrequently needed toobliterate persistent post-operative air spaces (Ellis,Hedberg, and Krueger, 1953). Another cause offailure is the collection of blood and fluid in theunobliterated pleural space with the eventualformation of another cowl over the lung.A number of instances of thickened pleura and

empyema are seen in cases of complicatedprimary tuberculosis where the pulmonary paren-chymal element of the complex is very small andhealed and the rest of the lung perfectly healthy, orin cases where an artificial pneumothorax inducedfor localized disease in one or two segments oflung has given rise to pleural complications.Again, most of the lung is perfectly healthy, andthere is certainly initially no generalized pul-monary fibrosis which would prevent expansionof the lung. Stenosis of a major bronchus isrelatively rare, but its presence would certainlycontraindicate decortication alone.We consider that, provided disease is not

scattered widely through the lung, failure to ex-pand the lung after decortication is due to in-ability to deal with the air leaks and fluid collect-ing in the pleural space in the early post-operative

on March 30, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.10.4.293 on 1 D

ecember 1955. D

ownloaded from

http://thorax.bmj.com/

T. SAVAGE and H. A. FLEMING

days. Such failure will inevitably lead to thelack of improvement, or even deterioration inlung function frequently reported in tuberculouscases (Wright, Yee, Filley, and Stranahan, 1949;Carroll, McClement, Himmelstein, and Cournand,1951 ; Falk, Pearson, and Martin, 1952), andleading to the conclusion that functional con-siderations should not be taken as a prime indica-tion for operation.While agreeing that these points must be con-

sidered, we feel that the cases reported here showthat such objections are far from absolute, and that,when the post-operative care is carried out asherein described, the end-result is good: the lungis fully expanded, and the patient has not onlylost a potential source of danger to his health buthas gained an improvement in functioning lungtissue.

In this hospital, where the indications for resec-tion in pulmonary tuberculosis are wide and manyoperations are carried out, minor decorticationsare often necessary in the course of resection oftuberculous segments of lung. However, in thispaper it is intended to confine the study to a seriesin which thoracotomy was carried out primarilyfor the thickened pleura, with or without empyema.

INDICATIONS AND CONTRAINDICATIONSFOR OPERATION

Forty-three cases were operated upon, the firstin September, 1950, and the last (of this series) inNovember, 1954.

INDICATIONS.-(I) Tuberculous empyema witheither no detectable disease in the underlying lung,or disease treatable by chemotherapy, partialresection, or thoracoplasty; (2) thickened pleuragiving rise to appreciable loss of respiratory func-tion, the conditions as regards parenchymal diseaseapplying as in indication (1).CONTRAINDICATIONS.-(1) Stenosis of a main

bronchus, though decortication plus a lobectomyin a lobar bronchostenosis is possible ; (2) severeuncontrolled disease in the underlying lung, e.g.,destroyed lung; here, pleuropneumonectomywould be indicated, provided the contralaterallung were satisfactory; (3) severe contralateraldisease making an operation on the side with thethickened pleura functionally impossible. How-ever, in lesser degrees of contralateral diseasedecortication is often indicated in order to allowlater surgical treatment of the contralateral lesion.

It has been suggested that a mixed infectiontype of empyema is a contraindication, but in thisseries there were two such cases (Case 1 and

Case 3) and in both the outcome of operationwas successful. Modern antibiotics undoubtedlyplay an important part in the success of theseoperations. Though there are no such cases inthis series, we would not consider the presenceof a bronchopleural or even a bronchopleuro-cutaneous fistula a contraindication to operation.

DETAILS OF CAUSE OF THE CONDITIONREQUIRING OPERATION

Primary tuberculous effusion and complicatedartificial pneumothorax were the causes in analmost equal number of cases. Spontaneousrupture of a cavity in a free pleura was the causein one case. Table I shows causes of empyemaand thickened pleura.

TABLE ICAUSES OF EMPYEMA AND THICKENED PLEURA

Primary tuberculous pleurisy and effusion .20Pleurisy and effusion complicating artificial pneumothorax 22Spontaneous rupture of cavity in a free pleura I

Total .. 43

Bacteriology of the pleural fluid was as inTable II.

TABLE IIRESULTS OF CULTURE FOR M. TUBERCULOSIS

Fluid positive.26sterile.17

Total .. 43

Of those in which the fluid was positive forM. tuberculosis, a mixed infection was present intwo cases, introduced, presumably, during themultiple aspirations that were carried out.

AGE OF PATIENTSThe age range was from 8 to 53 years. The

oldest was Case 30, a man in whom the result wasclinically, radiologically, and functionally satis-factory. The youngest was a boy of 8 with alarge empyema which developed in a primaryeffusion. The result was clinically and radio-logically excellent; function tests were not done.The majority of patients were in the late teens toearly thirties.

SEX OF PATIENTSTen cases were in women. The high proportion

of males in this series is probably not of anyspecial significance.

SIDE OF OPERATION

On these 43 patients, 44 operations were done,20 on the right side, 22 on the left, and one onboth sides (Case 6).

294



j.com/


ecember 1955. D

ownloaded from


DECORTICATION IN TUBERCULOSIS

DURATION OF THE CONDITION BEFORE OPERATION

In one case due to a complicated artificialpneumothorax, the condition had been presentfor 20 years before operation. In this case (Case22, see Fig. 10) the result was very satisfactory.The shortest duration was three months beforeoperation.

TABLE IIIDURATION OF CONDITION BEFORE DECORTICATION

20 years.1 case .. .. Complicated A.P.15 to20 years .. .. No cases .. --10 ,, 15 years .. .. 4 ,, .. Complicated A.P.5 ,, 10 ,, .. .. 2 ,, .. Complicated A.P.I 5 ,, .. .. 26 ,, .. 13 complicated A.P.

13 complicated primaryLess than 1 year .. .. 10 ,, 6 complicated primary

3 complicated A.P.1 ruptured cavity

Total .. 43 cases

INVESTIGATION OF CASES

Quite apart from the indication of the presence

of an empyema, many of these cases were operatedupon in an effort to improve function alreadynoticeably reduced, or in an effort to reduce lossof function and deformity of the chest wall due toconstrictive pleuritis when the patient grew older.Consequently pre-operative function tests were

carried out in as many patients as possible, andagain post-operatively wherever possible. Thetests used were maximum breathing capacity(M.B.C.) and bronchospirometry, the oxygen

uptake and ventilation of the individual lungsbeing measured as described by Fleming andWest (1954). Pre-operative bronchospirometrywas carried out in 28 patients but post-operativelyin only 16 so far. Five patients have not had itdone for various domestic reasons but havepromised to attend later. The remaining seven

have refused for a variety of reasons, the mainone being that as they are now fit and workingthey do not want to lose time for such an in-vestigation. The M.B.C. was determined in 30cases pre-operatively and in 28 post-operatively.Effort tolerance has been noted in all cases bothpre- and post-operatively, and a note has beenmade of the movement of the chest, air entry, anddiaphragmatic movement on fluoroscopy. Unfor-tunately a definite record of the diaphragmaticmovement was not made pre-operatively in allcases, but the postero-anterior and lateral radio-graphs give a good indication of the state of thediaphragm. Previous therapeutic phrenic crusheswere noted. A clinical and radiographic assess-

ment of the result of operation has been made in

all cases, but M.B.C. and bronchospirometryfigures are available in only a proportion. Thesewill be discussed later.

Postero-anterior and lateral radiographs weretaken pre-operatively in all cases, and tomographywas invariably done before cases were assessedfor operation. Bronchoscopy to rule out tubercu-lous endobronchitis or stenosis of a major bron-chus was carried out in some cases. All casesoperated upon in 1953 and 1954 (22 cases) hadpre-operative bronchography using an aqueoussuspension of " dionosil," this being considered anessential part of the complete investigation of allcases of tuberculosis before surgery. The broncho-grams showed no findings specific to the thickenedpleura except crowding of the segmental bronchiin a lung occupying a reduced space and some-times a " windswept " appearance. However, valu-able evidence was gained of the bronchi relatedto any parenchymal disease, and anatomical detailswere provided which were of use when resectionwas carried out at the time of decortication.

PRE-OPERATIVE PREPARATIONApart from general measures undertaken in the

preparation of any patient who is to undergomajor surgery, the following special procedureswere undertaken:PHYSIOTHERAPY.-The patient is taught breath-

ing exercises, both costal and diaphragmatic.Particular attention is given to the flat and poorlymoving or actually immobile affected side. Littleimprovement in movement was noted beforeoperation, but the exercises are most importantpost-operatively. The patient is also taught how tocough properly. Once again this is most impor-tant in the post-operative phase in preventing ortreating atelectasis.

PLASTER CAST.-A plaster cast which will hyper-extend the affected side is made pre-operatively(see Fig. 2). This is for use post-operatively. Insome earlier cases of this series such a plastercast was used to see how much improvement inmovement could be gained without operation. Theresult was very disappointing and the cast is nolonger employed pre-operatively.CHEMOTHERAPY.-A11 patients coming to the

operation had had some chemotherapy and atthe time of operation were on an appropriatecombination of two of the drugs streptomycin,P.A.S., and isoniazid. The earlier ones had shortcourses, but latterly more prolonged continuoustherapy had been used. No patient was sputum

295



j.com/


ecember 1955. D

ownloaded from



positive at the time of operation, and only a smallproportion had ever been sputum positive. Athree months' operative cover of drugs was usedin all cases, but latterly the drugs have beencontinued for at least six months post-operatively.In some cases where parenchymal disease hasbeen left, it may be advisable to continue muchlonger. While the patient is awaiting operationcavitated disease, whether ipsilateral, contralateral,or bilateral, should be controlled by posturalreduction. Reference to Dillwyn Thomas'smethod of postural reduction has been madepreviously (Thompson, Savage, and Rosser, 1954).

OPERATIONPatients were operated on in the lateral position

under general anaesthesia. The incision was as

for a standard postero-lateral thoracotomy, beingkept well clear of the inferior angle of the scapula.In some cases a rib was resected, usually the sixth,seventh, or eighth, but in many of the more recentcases an intercostal approach at either the upper

or lower border of a rib was used. However,in cases where the chest is " frozen " and the ribsare very close together or even overlapping, it isimpossible to enter the thorax without removinga rib. The rib was selected according to wherethe main thickening of the pleura was judged tobe radiographically. On several occasions twoapproaches were necessary, e.g., at the sixth andeighth ribs. This helped considerably in cases

where the parietal pleura was very tough, and itdid not seem to be detrimental to the end-result.The parietal pleura and lung are mobilized in

the extrapleural plane using the fingers as dis-sectors as much as possible. This is hard manualwork. Occasionally in very tough cases sharpdissection with knife or scissors is needed. It isnecessary to remove this thick parietal pleura(and it is always the parietal pleura which isthickest) entirely in order to get good movementof the chest wall eventually (Fig. 1). The stripover the apex is sometimes difficult, but on themediastinal aspect it is, as a rule, fairly easy; infact there is seldom much thickened pleura heredue to the movement of the mediastinum acrossto the side of a chronic effusion. The dissectionis continued on all sides round to the hilum ofthe lung. Special care is taken to prevent damageto any mediastinal structures. The thoracic ductand vagus nerve are preserved, and the phrenicnerve is treated with great respect, as the end-result of the operation may be markedly affectedby a paralysed diaphragm. An oesophageal tubepassed by the anaesthetist gives valuable evidence

FIG. 1.-Specimen of pleura removed at decortication showing thick-ness (0.5-1.2 cm.) and dense structure of parietal pleura (on left)and less thick visceral pleura which enveloped a very small lung.

as to the site of the oesophagus, which may bedangerously displaced and obscured.Removal of the thickened pleura from the dia-

phragm is often very difficult, as it is denselyadherent and a plane for dissection can seldom befound. Great care is taken here. It is better toleave plaques of thickened pleura on the dia-phragm than to damage its musculature. Smallaccidental incisions through the diaphragm areimmediately sutured and no morbidity has re-sulted. At this stage the diaphragm is well freedfrom the lung and from the chest wall by strippingdown in the costophrenic sinus all round. This isalso an important step.The visceral pleura is never as thick as the

parietal, but the ease with which it strips varies.In some cases, notably A.P. cases, it may berelatively easy, but in primary cases or where thereis much parenchymal disease the pleura is usuallyvery adherent and many alveolar leaks are inevit-ably produced in the dissection. It is advisable toleave the thickened pleura at the site of paren-chymal disease if it is not intended to resect it.When the lung is freed it is not infrequently

collapsed to the volume of a clenched fist, and theintact visceral " peel" has the same size and shape(Fig. 1). However, when the lung is inflated itusually expands well and assumes a size approxi-mately that of the hemithorax. If an empyemais present in the mass of pleural fibrous tissue, it isalmost invariably opened in dissecting the pleuraoff the diaphragm. In any case there is no greatvirtue in removing the empyema intact. In factwhen a large empyema is present it should beopened and evacuated early in case a broncho-pleural fistula should be present, which, by aspira-tion, would cause the bronchial tree to be floodedwith pus. The dangers resulting from spilling pus

296



j.com/


ecember 1955. D

ownloaded from



into the pleural space are notgreat, as chemotherapy andefficient expansion of thelung post-operatively preventthe re-formation of anempyema.Another important step to-

wards ensuring full expansionof the lung is to free the fis- .sures and any infolding of the X?lung caused by thickenedbands of pleura and adher-ence of one part of the lungto another.

After removal of the pleuraand empyema any palpableparenchymal disease is dealtwith on its own merits, thatis, as assessed by pre-opera-tive radiographs and palpa-tion at operation. In this FIG. 2.-Postseries palpable parenchymaldisease was present in 30 cases. In 14 casessurgery was deemed necessary to treat it, asfollows:

Thoracoplasty . .Segmental resectionWedge resection ..

. . 4 cases8 cases

. I case

The resections were carried out at the time ofthe decortication. The thoracoplasties, when indi-cated, were performed some weeks after the de-cortications. In one case (Case 22) a contra-lateral lesion was segmentally resected five monthsafter the decortication.

BLOOD TRANSFUSIONS DURING OPERATION.-Thedissection of the pleura is always a bloody busi-ness and rapid intravenous replacement is essen-tial, when 1,500 to 2,000 ml. of blood is usuallyneeded during the course of the operation. It isnot always possible to stop all bleeding beforeclosing the chest, as there is often an extensivecapillary ooze.

DRAINAGE OF THE CHEST.-In the earlier casesa posterior basal tube and either an anterior orposterior apical intercostal tube were used. Thepresent practice is to use the usual basal tube anda long tube inserted in the mid-axillary line ineither the ninth or tenth intercostal space and heldagainst the inside of the chest wall by a catgutsling with the orifice of the tube near the apex ofthe pleural space (see Thompson and others, 1954).The tubes are of firm rubber, internal diameterL in. with wall 1/10 in. thick, in order to preventcollapse of the tube when strong suction is used

SI

t-operative posture and hyperextension plaster cast in use.

post-operatively. Each tube is atached to the skinby passing it through a rubber cuff which is thenstitched to the skin. This system of fixing thetubes is useful because no stitch hole is made inthe tube itself, and post-operatively the tube iseasily shortened through the cuff.

If there is much air leaking from the damagedlung a third intercostal tube is sometimes placedthrough the seventh or eighth intercostal space inthe anterior axillary line, and also held against thechest wall by a catgut sling.

POST-OPERATIVE CAREIt is felt that the success of the operation, par-

ticularly from a functional point of view, dependslargely upon the efficiency of the post-operativecare. It is most important that the lung should befully expanded at the earliest possible time andkept so until all air leaks have ceased. At thesame time flattening and reduced movement of thechest wall must be corrected and the diaphragmencouraged to resume a normal function.MANAGEMENT OF THE INTERCOSTAL DRAINAGE.-

As soon as the patient returns to the ward, strongcontinuous suction is applied through water-sealbottles to all drainage tubes. The degree of suc-tion is approximately one atmosphere (about 30 in.of mercury) of negative pressure. This preventsfluid collecting in the pleural space, and keeps thelung expanded against the chest wall in the face ofeven the most severe leak of air from the decorti-cated lung. At present in this hospital a large

297

w4a



j.com/


ecember 1955. D

ownloaded from



central suction pump capable of supplying multipleleads with a negative pressure of one atmosphere,and of dealing with large volumes of air, is used.The tubes are not removed till they have ceased

functioning, that is, when the air leak has stoppedand the lung is fully expanded, or when a tubebecomes blocked by clot. In the latter instance, ifthere is still a persisting pleural space it may benecessary to insert another intercostal tube underlocal anaesthesia. Not infrequently shorteningone or other of the tubes through the cuff willevacuate a persisting space. In some cases tubeswere necessary for up to 14 days post-operatively.Three patients had tubes in for periods longer thanthis, and these will be discussed under " Compli-cations."Very occasionally aspiration to evacuate a small

loculated space was needed. Most of these smallloculated spaces resolved without any active treat-ment other than the usual post-operative routineof physiotherapy and posture in a plaster cast.

POSITION OF THE PATIENT.-The patient isnursed lying on the unoperated side, and as soonas possible, usually on the second or third day, heis put into the special hyperextension plaster (Fig.2). In this position the mediastinum tends to fallto the opposite side, the diaphragm is carriedcaudally, and the ribs are stretched apart-the veryreverse of the situation before operation. Theposition is surprisingly comfortable.The patient continues to lie in the plaster cast

except for meals for three to six weeks until thechest expansion is equal on both sides or hasreached its maximum. The cast is then used forsleeping until three to six months have elapsed,the time in each case depending on the duration ofthe condition, the extent of the decortication, andthe rate of improvement.PHYSIOTHERAPY.-Breathing exercises, as men-

tioned in the pre-operative preparation, are startedfrom the first post-operative day and continued forthree to six months, depending on the same factorsas control the use of the plaster casts.

POST-OPERATIVE RADIOGRAPHS.-All post-opera-tive care is guided and controlled by frequent port-able A.P. radiographs in the early stages, thesebeing taken at least 12-hourly.

PENICILLIN COVER.-Penicillin or any other suit-able antibiotic is always used as an operative cover,and is continued so long as there are tubes in thechest.COMPLICATIONS.-There were no deaths in this

series.

TABLE IVCOMPLICATIONS

Drainage-tube-track infection (non-tuberculous) . .Bronc hopleural fistula and empyema (non-tuberculous).'. .Persistent pleural spacePhrenic nerve accidentally injured at operation . .Reactivation of diseaseHaemothorax . .

3

6321

cases

,,,,

,,

Tension pneumothorax and extensive surgicalemphysema as post-operative complications havenot been met with. The use of multiple drainagetubes and suction probably prevents their develop-ment.The use of adequate antituberculous drug

therapy and early obliteration of the pleural spaceby good suction drainage is considered to be themain factor in the prevention of tuberculousempyema after operation.

In three cases (Nos. 3, 8, 18) a tube-track becameinfected with pyogenic organisms. In these casesthe tubes could not be removed early, but wereremoved by gradual shortening, controlled byserial sinograms. The sinuses took some weeks toheal, but remain healed after three years. Thiscomplication does not seem to have affected ad-versely the enid-result.

In Case 16 a small basal non-tuberculousempyema developed with a small fistula to a sub-segmental bronchus (demonstrated by sinogram).This healed after simple intercostal drainage andgradual shortening of the tube controlled by serialsinograms. However, the diaphragm has remainedhitched up and relatively immobile, possiblyaccounting in part for the poor result (Fig. 10).The scar remains healed after two years.

Persistent pleural spaces for more than three orfour weeks post-operatively were seen in six cases.In two of these thoracoplasties were carried outand the spaces successfully closed, but it shouldbe pointed out that in both these cases paren-chymal disease was present and the thoracoplastywas done primarily for this. In three cases thesmall spaces disappeared spontaneously in lessthan six months with only a little thickening of thepleura radiologically. We consider that largepersisting pleural spaces, particularly if theyaccumulate fluid, will only cause thickening of thepleura again and so prejudice the end-result; con-sequently they are treated vigorously by insertionof a fresh intercostal tube and the application ofsuction. If such spaces are multiple with muchaccumulation of fluid, and particularly if thepatient becomes pyrexial, we advocate immediatere-thoracotomy, removal of debris, and re-institu-tion of suction drainage. Early full expansion and

298



j.com/


ecember 1955. D

ownloaded from



obliteration of the pleural space without accumula-tion of blood and fibrin is essential for a goodfunctional result.

Despite care, the phrenic nerve was damaged atoperation in three cases. We feel that this hasreduced the effectiveness of the operation from afunctional point of view only in these cases.So far there have been no cases in which new

parenchymal disease has appeared after operation,but in two cases disease has become reactivated.

In Case 6 contralateral disease with cavitationand a positive sputum occurred two years aftera left decortication and thoracoplasty. The patienthad previously had a right artificial pneumothoraxwhich had caused some degree of thickening ofthe pleura. This right-sided disease was dealt withby decortication and segmental resection. Theresult both for disease and function is satisfactorynine months after the second operation (Fig. 10).

In the other case (Case 7) bilateral breakdownhas occurred. The sputum is positive and thepatient is at present in this hospital under treat-ment. Further surgery may be carried out. Inboth these cases bronchography, after the break-down, has shown bronchiectasis in the segmentsin which the active disease appeared. They areboth cases treated early in the series (1951-2), andwe think that if pre-operative bronchography hadbeen carried out there would have been a betterassessment of the disease, with resection at thetime of decortication.

Excessive post-operative bleeding in the first 48hours with blockage of the drainage tubes occurredin one case (Case 30). This patient required im-mediate re-thoracotomy and removal of the bloodclot with reinstitution of suction drainage. Theend-result did not appear to be adversely affected(Fig. 10).From this presentation of post-operative com-

plications it can be seen that only one patient(Case 7) has suffered any serious trouble followingthe operation. However, it is felt that he may stillbe saved and his disease is such that the operationalone cannot be held responsible for his presentcondition.

PATHOLOGYIn 31 cases a frank empyema with tuberculous

pus and caseous matter was present. In two ofthose complicating primary effusions, a caseoushilar lymph node actually communicating with theempyema was found at operation. In three othercases a turbid fluid but no pus was present, andin the remainder no space was found in thethickened pleura. It is interesting to note the largeproportion of persisting empyemata. In quite a

FIG. 3.-Section of pleara removed at decortication, showing densefibrosis above, and typical tubercles with caseation and Langhan'sgiant cells. Haematoxylin and eosin. x 95.

number of cases aspiration just before operationrevealed no space, but at operation an empyemawas found.The removed thickened pleura was always

thickest in its parietal part (in Case 24 it was2-2.5 cm. thick) with the visceral pleura thickenedonly slightly (for example, see Fig. 1).The histology of the pleura showed dense

collagenous fibrous tissue with occasional tuber-culous and caseous areas. Positive evidence, histo-logically, of tuberculosis was always found in caseswhere an empyema was present, but not always inthe others (Fig. 3).

In cases in which segmental resections werecarried out at the time of decortication, the tissueremoved showed the usual changes of pulmonarytuberculosis. There was little if any lung tissueunaffected by tuberculous disease in thesespecimens.

Biopsies of healthy lung were not taken in thisseries, and we can make no factual contribution tothe histological appearance of the lung under con-strictive pleuritis.

RESULTSA recent follow-up by personal examination or,

in a few cases, by information from the Chest

299



j.com/


ecember 1955. D

ownloaded from


FIG. 5

FIG. 4

Before decortication. M.B.C. 63 I. min.

21 months after decortication. M.B.C. 94 I./min.

..-.. ... . ..

Time in seconds

FIGs. 6a and 6b.

AiI'



j.com/


ecember 1955. D

ownloaded from


Clinic has been made in all cases. The follow-upvaries from four and a half years to six months.

(1) In cases where an emphysema or uninfectedpersistent pleural space was present (34 cases), theobliteration of the pleural space has been achievedin every case. Thus the ever-present danger ofcomplication or toxaemia from an empyema hasbeen removed.

After removal of thickened pleura the space hasbeen filled by expansion of lung tissue. In nocase has there been greater mediastinal shift thanthere was pre-operatively; in fact, where this waspresent pre-operatively it has been reduced, ifnot completely, at least partially. In two casesobliteration was assisted by thoracoplasty, but, asalready pointed out, the thoracoplasty was doneprimarily to control parenchymal disease. Inthree cases thoracoplasty was also carried out forremaining disease, but not until the space had beencompletely obliterated.

It follows from this that in many cases un-desirable overdistension of the contralateral lungwas relieved, thus reducing the chance of re-activation of contralateral disease from this cause.Consequently, it would seem that the contralateralreactivation in Cases 6 and 7 occurred from someother cause.

(2) Disability due to displaced mediastinum wasonly seen in one patient, Case 23, a woman of 47in whom thickened pleura had been present for20 years. In addition to shortness of breath onthe slightest exertion, she had marked oedema ofthe ankles. In this case operation replaced themediastinum almost centrally, and the patient isnow leading a normal life for her age and posi-tion (school teacher) with much improved exer-cise tolerance and no oedema of the ankles (seeFigs. 4 and 5). A functionless lung has beenrestoredl to reasonable function (Fig. 6).

In other cases disability from this cause mayhave become apparent in later years if the media-stinal position had not been corrected.

FIG. 4.-Case No. 23, woman aged 47, with a left constrictive pleuritisexisting for 20 years following artificial pneumothorax. There ismediastinal displacement and the function of the left lung isnegligible (see Figs. 5, 6, and 10).

FIG. 5.-Case No. 23 two years after decortication. The mediastinum.. is now central and function improved (see Figs. 4 and 6).

FIG. 6.-Case No. 23 (see Figs. 4 and 5), example of functionlesslung being restored to reasonable function after 20 years (seealso Fig. 10).

FIG. 7.-Case No. 13 before operation.FIG. 8.-Case No. 13 after decortication, showing almost normal

appearance of the postero-anterior radiograph (see also Figs.7 and 9).

FIG. 9.-Case No. 13. Bronchospirometric traces before and sixmonths after decortication (see also Fig. 10).

FIG. 7

,111111

RIGHT C)2 UPTAKE-LUNG .4

j§|t 260 CCJMhIN.

LEFT70 CCjMlt,

LUNG

IO

,18 CCJFI

140 ~CJMN

Fio. 9

2C



j.com/


ecember 1955. D

ownloaded from


BEFORE OPERiRIGHT LEFT

0

C°, V WbO .,

XId

X :fi>.. C

X k X ,

OD > OD >

AFTER OPERATIONRIGHT LEFT

.oa

0 0

,

X C. X V.

OD > CD >

Case 1, G. R.

Age, 33

Condition Present, 3 yr.

Cause, Primary

Case 6, D. D.

Age, 20

Condition Present, 2 6/12 yr.

Cause, A.P.

Case 10, D. 0.

Age, 20


Cause, Primary

Case 12, W. E. T.

Age, 33


Cause, A.P.

Case 13, G. C.

Age, 27

Condition Present, yr.

Cause, Primary

Case 16, M. H.

Age, 21


Cause, Primary

Case 17, T. G. J.Age, 25


Cause, A.P.

Case 22, W. H. W.

Age, 25

Condition Present, 6/12 yr.

Cause, A.P.

Case 23, E. A.

Age, 47


Cause, A.P.

67 66M.B.C i

not done [: 1 n

i' JAN 1953

1 M.B.C.I1 107 I./min

MAI15

7865

...aSDEC,1954

M.BC.I1\ 127 1/min.

5 -RIBTHORACOPLASTY

60

.34 ...

1H H41.E' I:*

MAR.1953 78 MAR,1955

95/min. 103 I/mi n 4 60

nIAIIAM n | R~~~FIGHTSEGMENTX'N LLLILLLL. ~~~~~~~~~~RESECTION :j

JAN, 1953 98

Zi A M.BC._ 631I/min. I:,.-.j

2 4

< ~~NOV., 19S4- ---

3 69

94 I./min:::

FIG. 10FIGS. 10 and 11.-Schematic representation of radiographic appearances and functional results in those cases in which bronchospirometry

in each lung expressed as a percentage of the total function. The normal

I

I li I 16, I

ATION

..iii::



j.com/


ecember 1955. D

ownloaded from


Case 24, W. K.Age, 18Condition Present, 2 yr.Cause, Primary

Case 30, D. P. J.Age, 18Condition Present, 9/12 yr.Cause, Primary

Case 31, G. H.Age, 53Condition Present, 2 yr.Cause, Primary

Case 34, G. B.Age, 18Condition Present, 3 yr.Cause, Primary

Case 35, B. P.Age, 40

AUG.,1954

MAR 1955

M.B

119 1./min

MAR 1955

M B.C. .56

LEFT8 /rDIAPHRAGM11

PARALYSEDCause, A.P.

Case 36, J. F.Age, 27

Cause, A.P.

RIGHT LEFT

0

C-0.'yoed o yz

X c X 4

O > OD

.30 35[j70

Fio. 11was carried out before and after decortication. The shaded blocks represent oxygen uptake and the plain blocks minute ventilationvalue for each of these is 55% for the right lung and 45% for the left.

c0bi0

=ive

IIIII



j.com/


ecember 1955. D

ownloaded from



(3) Clinically, changes in function were judgedby changes in effort tolerance, chest and dia-phragm movement on the affected side, and breathsounds on that side.

Before operation 24 patients complained ofshortness of breath on mild to moderate exertion,e.g., walking on the flat or climbing a flight ofstairs. In all these there was some improvementin effort tolerance after operation. In a few itwas marked, in one only a little, but in all cases

the patient considered it well worth while. Thesecases also showed an improvement in chest move-

ment and breath sounds.Worthy of special note are Case 13, a tin worker,

22 years of age (see Figs. 7, 8, 9, and 10), and Case28, a steel worker, 21 years of age. Pre-opera-tively neither of these young men had been able todo their normal work because of shortness ofbreath, but both are now doing full work. Theother 19 were all young patients who denied any

reduction of effolrt tolerance. Post-operatively,hovAever, the chest movement and breath soundsimproved in all cases except one (Case 34, see

Fig. 11), a girl of 16. There was radiographicimprovement, but the chest movement remainedpoor six months after operation, and the M.B.C.and bronichospirometrical results were correspond-ingly disappointing. A possible reason for thispoor result was that the plaster cast was discardedafter three months. An effort is now being madeto inmprove chest movement by physiotherapy anda plaster cast.

Improvement in diaphragmatic movement hasbeen seen in all cases, except those in which thephrenic neirve had been crushed or avulsed previ-ously (six cases), or had been damaged at theoperation (three cases). The functional results in

these cases might have been better if the diaphragmhad been functioning.

(4) Radiological improvement of varying degreewas seen in all cases. In some it was remarkable,in others it was less striking, and in a few poor.Average examples are shown in the various figures.

(5) The M.B.C. was estimated both pre- andpost-operatively in 25 cases, and the results are

presented in Table V.

It can be seen that there is some improvementin ventilatory capacity in all except two cases. InCase 14 an eight-rib thoracoplasty was necessaryafter decortication to control right upper lobe dis-ease. Case 42 had extensive parenchymal diseasewith bronchiectasis of both lobes, and the,diaphragm was paralysed.

TABLE VPRE- AND POST-OPERATIVE M.B.C. READINGS

CaeAge Duration Pre-Casein of operative Follow- op

No. ~~~M.B.C. up N5No. Years Condition (1 mi.) (1

2i 18 7 monthsl 85 4) yr.7 21 14 yr. 97 4 ,

10 20 2 ,. 105 31,It 36 1,. 7 1 32

13 22 1, 96 314 25 1 ,. 80 2'

17 25 1,, 107 2-),

20 39 2 ,, 100 2).

22 25 6 months, 95 2

23 47 20

24 18 2 ,26 20 1

27 19 8 mon28 21 I yr.29 32 4 mon30 18 931 53 2 yr.32 32 134 16 3 ..

35 40 13 ..

36 27 9..

37

40

41

421

35

21

32

37

nths

nths

4

8

1 5

63 2

133 2139 3 mth.132 yr.131 6 mth.66 8109 1 . yr.70 1

113 2-141 6 mth.

67 8

99 6

94 6

100 6

95 6

65 6

Post-)erat i veA.B.C.I./min.)

143

Remarks

121 Phrenic nerveinjured atoperation

155

105 Phrenic crushbeforeoperation

14680 8-rib tbhoraco-

plastv127 5-rib thoraco-

plasty114 4-rib thoraco-

plasty103 L. decortication.

R. 2 segmentsremoved

94 Phrenic nerseinjured atoperation

1621621Ar1'+D16218417786127119 Chest pove-

ment poorpost-operatively

84 Phrenic crushbeforeoperation

102 Much paren-chymal diseasein lung

117 Two segmentsremoved

111 One segmentremoved

97 One segmentremoved

48 Much paren-chymal disease

in lungBronchiectasisof both lobes,and paralyseddiaphragm

All other cases showed an improvement; asmuch as 50 litres per minute in Case 10. Thiscase is of further interest in that the function, par-ticularly the oxygen uptake, improved some timeafter the operation.CASE 10.-D. O., aged 20 years (Figs. 14 and 15),

had had a right thickened pleura duLe to complicatedprimary effusion for two years three months at thetime of operation.

Pre-operative function tests gave the followingresults:

M.B.C. 105 l. min.

Bronchospirometry Right Lung Left Lung

Oxygen uptake 15% 85%Vent ilat ion .. . 31% 69%

304



j.com/


ecember 1955. D

ownloaded from



FIG. 13.-Case 3, three and a half years after decortication, showingfully expanded left lung.

FIG. 12.-Case 3, a man aged 27 before operation. Duration ofeffusion nine months. This was a primary tuberculous effusionwith positive bacteriology.

FIG. 14.-Case 1U betore decortication (see Fig. 1).

Function tests five weeks after operation:M.B.C. = 131 1. !min.


Oxygen uptake 23% 77%/Ventilation . .. 45°/ 550°

FIG. 15.-Case 10, three and a half years after decortication (seealso Figs. 14 and 10).

Function tests two and a half years after operationgave the following results:

M.B.C.= 155 1. Tmin.


Oxygen uptake 39%/. 61%Ventilation .. . 44% 56%

305



j.com/


ecember 1955. D

ownloaded from



Further point to this observation is given by asmall number of pulmonary angiograms whichdemonstrate that in the early period after decorti-cation there is a poor flow through the branchesof the pulmonary artery, while some months laterthese have opened up considerably. This wouldaccount for the lack of improvement in oxygenuptake parallel with ventilation shortly afteroperation. For this reason in most cases broncho-spirometry was not repeated till about a year afteroperation.On the whole it would seem that those cases in

which the condition had been present a long time(Cases 23, 35, and 42) the ventilatory improvement,if any, is less striking than in cases where it hadbeen present a relatively short time.

(6) In those cases in which both pre- and post-operative differential lung function tests werecarried out the findings are shown graphically inFigs. 10 and 11, and Figs. 6 and 9 depict examplesof the change in function.

Significant improvement in function of the de-corticated lung is seen in all except four cases,namely, Nos. 16, 34, 36, and 42. In the first anon-tuberculous empyema which developed afteroperation may in part account for the poor result;the second has been discussed already; in the thirdthe pleura was not grossly thickened, but scatteredcalcified parenchymal disease was present in bothlobes; in the last the lung had been reduced to avery small volume for 15 years, and the extensivedisease has already been described.

In two cases in which the diaphragm is para-lysed (Cases 12 and 13), the improvement is prob-ably impaired by this.

In most cases the alteration in ventilation andoxygen uptake is either roughly parallel or theventilation has improved rather more than theoxygen uptake, but in five cases (Nos. 17, 30, 31,40, and 41) the improvement in oxygen uptakeis greater than that in ventilation. This supportsthe suggestion that any vascular changes in thelung are not wholly irreversible. Further, Case 10suggests, as already noted, that improvement inoxygen uptake may be slower than in ventilation.An early poor result may prove satisfactory afteran interval of several years.

DISCUSSIONLittle further need be said about the indications

for operation and the technique used. The absentmortality and the low morbidity rates also speakfor themselves. The importance of the post-opera-tive regime described must be emphasized, andthis is judged to be imperative for the avoidance

of fistulae, unexpanded lungs, and space infections.It is further discussed under function. The find-ings of frank tuberculous empyema in 31 of the43 cases is a strong argument for a radical surgicalapproach to the condition. The gross functionalloss and the chest deformities produced are furtherstrong points in favour of surgical treatment.FUNCTION.-Previous reports on changes in pul-

monary function following decortication in tuber-culous cases have largely involved a detailedanalysis of a small number of cases (Gordon andWelles, 1949; Wright and others, 1949; Falk andothers, 1952). These workers find little or no in-crease in the function of the decorticated lungfollowing operation, and are agreed that this is incontrast to the improvement seen after decorti-cation for haemothorax and non-tuberculousempyema. Carroll and others (1951) do reportimprovement in function and find that the bestresults are obtained in the cases with the shortesthistories. Our Cases 35 (duration 13 years) and42 (duration 15 years) with their disappointingfunctional results (see Fig. 11) would support thisthesis, but Case 23 (Figs. 4, 5, 6, and 10), showingmarked improvement in function after 20 years,demonstrates that this rule is by no means abso-lute, and useful function may be restored afterlong periods.The functional results presented here are better

than any previously reported except those ofPatton, Watson, and Gaensler (1952), and it is feltthat there may be two reasons for this. First, themeticulous attention to the post-operative period,with emphasis on early and full re-expansion ofthe lung, avoidance of accumulation of pleuralfluid, and the use of the hyperextension plastercast; and secondly, the fact that we wait nearly ayear before reassessing the function. Many ofthe reported cases have been investigated at amuch shorter period after operation, and forreasons already stated we think that this is unwise,as full restoration of function may not yet be com-plete. There is no doubt that functional improve-ment of significant degree can be achieved in mostcases, and we are now prepared to recommend op-eration on that ground alone. This is particularlyimportant in the younger age groups where theloss of function is often associated with a severeand progressive contraction of the hemithorax withresulting deformity. The importance of such con-ditions later in life is not known, but they aresurely better avoided.

Patton and others (1952) have studied the func-tion in a series of patients up to three years afterdecortication, and also find that function tends to

30)6



j.com/


ecember 1955. D

ownloaded from



improve steadily with the passage of time, andthat the final result is not closely related to theduration of the condition before operation. Insome of their cases only visceral decortication was

carried out. We feel that this would unfavour-ably influence the functional result, as the chestwall movement must be severely limited by thepresence of the thick parietal layer and deformitywould be liable to be progressive.Any discussion of functional results must take

into consideration the pathology of the underlyinglung. In our cases where there was pre-existingextensive parenchymal disease it was only to beexpected that final function would be poor. How-ever, in many of these cases the major portion ofthe lung is free from tuberculous disease, and theactual pathological changes caused by the collapseare the important features. There is remarkablylittle accurate information on the histologicalappearance of this lung, and we are unable tomake any addition to it. However, in a case ofpleuropneumonectomy carried out two and a

quarter years after collapse of the lung, it was

reported that "the areas of macroscopicallynormal lung tissue showed on microscopy thatthey were partially collapsed, and there was

thickening of the alveolar septa which appearedto be partly due to increased cellularity and partlydue to increase of connective tissue."Gordon and Welles (1949) quote Vorweld as

finding similar appearances together with tortuosityof the blood vessels and dilatation of capillaryspaces. Gardner (1925) demonstrated, in pneumo-thorax cases, that there was interlobular fibrosisand considerable endarteritis often even of thelarger vessels near the hilum. It is inevitable, ifarterial hypoxia is to be avoided when a lung iscollapsed, that there must be a reduction in theblood flow to that lung. If this has been in exist-ence for a long period, it cannot be expected toreturn to normal immediately, but we feel thatthe evidence presented shows that considerableimprovement can take place even after a longperiod, though naturally this will occur more

rapidly and completely in cases operated upon

early. Opening of the larger vascular channelsbefore perfusion of the alveoli is restored wouldaccount for the arterial hypoxia reported afterdecortication (Gordon and Welles, 1949; Wrightand others, 1949) and for the delayed improve-ment in the oxygen uptake we have already noted.

All reports are agreed on the pathology of thepleura and the fact that fibrosis may extend intothe interlobular septa and limit expansion andfunction. However, the striking features are the

relative ease with which the visceral "peel" isachieved, and the great toughness of the materialremoved. To one who has handled these " peels,"often up to 2 cm. in thickness and with the -con-sistency of boot leather, it must be obvious thatthe pleura has finished demonstrating its astonish-ing powers of healing referred to in the annota-tion in the Lancet (1955), and that the only hopeof improving function and avoiding furtherdeformity to the chest wall is by removing suchtissue.For the future, prevention of such long-term

effects may well be the answer. Attention toaspiration, chemotherapy, posture, and physio-therapy as described by Birath (1952) when fluiddevelops should limit the development of empyemaand of constrictive pleuritis with its resultingeffects. However, where the condition is estab-lished, it is felt that the present report demon-strates that decortication can be a safe and satis-factory method of treatment.

SUMMARY

A study of 43 tuberculous patients in whomdecortication of the lung was carried out is pre-sented. The condition of constrictive pleuritisoccurred in about equal numbers in primarypleural effusion and as a complication of pneumo-thorax treatment. The cases have been followedfrom between six months and four and a halfyears from operation. The indications for opera-tion are discussed and include functional con-siderations as well as the frequent finding of atuberculous empyema (31 cases).

Pre-operative investigations and treatment andoperative and post-operative methods are dis-cussed. Of great importance are the use of ade-quate post-operative suction to secure early andcomplete expansion of the lung with space oblitera-tion, and the use of hyperextension plaster caststo expand the hemithorax and the lung.There has been no mortality and little morbidity.Pre- and post-operative bronchospirometry has

been carried out in 18 cases, and maximum breath-ing capacity measurements and clinical assessmentin most of the others. In the majority of thesecases lung function has been considerably im-proved. The functional assessment is not usuallymade until one year after operation.

It is concluded that in selected cases of tuber-culous pleural disease decortication of the lung isa safe and satisfactory procedure, removing infec-tion, obliterating space, re-expanding the lung, andimproving function.

307



j.com/


ecember 1955. D

ownloaded from



We are especially grateful to Mr. Dillwyn M. E.Thomas for suggesting this study and for his helpand encouragement, and to Dr. L. R. West for carry-

ing out many of the functional studies. Thanks are

due to Dr. H. M. Foreman. M.B.E., Dr. L. R. West.Mr. H. R. S. Harley, and Mr. Dillwyn M. E. Thomas,under whose care the patients were treated at SullyHospital, to the chest physicians of South Wales fortheir assistance in following up patients, to Dr.R. M. E. Seal for pathological studies, to Mr. W. R.Probert for helpful criticism, to Miss V. Davies andher staff for providing Figs. 10 and 11, and for makingthe plaster casts and the estimation of the maximumbreathing capacities, to Mrs. B. Marshal for the photo-graphs, and to Miss J. Hilary-Jones for the typing andsecretarial assistance.

REFERENCESAnnotation (1954). Lancet, 1, 34.--(1955). Ibid., 1, 1162.

Birath, G. (1952). Amer. Rev. Tubere., 66, 134.Carroll, D., McClement, J., Himmelstein, A., and Cournand, A.

(1951). Ibid., 63, 231.Ellis, F. H., Hedberg, G. A., and Krueger, V. R. (1953). Proc. Afa.vo

Clin., 28, 537.Falk, A., Pearson, R. T., and Martin, F. E. (1952). Amer. Rev.

Tuberc., 66, 509.Fleming, H. A., and West, L. R. (1954). Thotax, 9, 273.Gardner, L. U. (1925). Anwer. Rev. Tuberc., 10, 501.Gordon, J., and Welles E. S. (1949). J. thorac. Surg., 18. 337.Gurd, F. B. (1947). Ibid., 16, 587.Lam, C. R. (1948). Arch. Surg., Chincago, 56, 1.

O'Rourke, P. V., O'Brien, E. J., and Tuttle, W. L. (1949). Ainer. Rev'.Tuberc., 59, 30.

Patton, W. E., Watson, T. R., and Gaensler, E. A. (1952). Surg.Gynec. Obstet., 95, 477.

Quinlan, J. J., Schaffner, V. D., and Hiltz, J. E. (1952). J. thorac.Surg., 23, 125.

Riley, A. O., and Kaunitz, V. H. (1951). Ibid., 22, 341.Stead, W. W., Eichenholz, A., and Stauss, H. K. (1955). Amer. Rem'.

Tubere., 71, 473.Thompson, H. T., Savage, T., and Rosser, T. H. L. (1954). Thorax,

9, 1.Waterman, D. H., and Domm, S. E. (1951). Dis. Chest, 19. 1.

Weinberg, J., and Davis, J. D. (1949a). Amer. Rev. Tubere., 60, 288.(1949b). J. thorac. Surg., 18, 363.

Wright, G. W., Yee, L. B., Filley, G. F., and Stranahan, A. (1949).Ibid 18 372.

308



j.com/


ecember 1955. D

ownloaded from


decortication of the in tuberculous disease · type of empyema is a contraindication, but in this...

Documents