Part II
COMPLICATIONS OF WOUNDS OF THE CHEST
CHAPTER IV
Complications and Sequelae
Thomas H. Burford, M.D.
Except for two or three immediate complications, such as tension pneumothorax,hemorrhage, and acute gastric dilatation, almost any of the complications ofwounds of the chest might occur at any time after wounding. It was thereforenecessary to be on the watch for them, and to try to forestall their occurrence,at all times.
While the attempt to separate them chronologically is somewhat academic, thecomplications of chest wounds may be divided, generally speaking, into (1) earlyand (2) late or delayed complications as follows:
Early complications included tension pneumothorax,hemorrhage, acute gastric dilatation, emphysema, pulmonary edema, atelectasis,hematoma of the lung, bronchopleural fistula, (which sometimes was persistent),wound disruption, clostridial myositis, and chylothorax.
Delayed complications included abscess of the lung,pneumonitis and pneumonia, chronic parietal sinuses, defects of the chest wall,adhesive pleuritis (infolded lung), traumatic osteomyelitis, and abscess of themediastinum.
The important sequelae of chest injuries in World War IIwere diaphragmatic hernia, hernia of the lung, and a group of residual symptoms,of which pain and dyspnea were the most prominent.
TENSION PNEUMOTHORAX
General Considerations
Pathogenesis-Tension or pressure pneumothorax might be of fourorigins (figs. 42 and 43):
1. It might result from a pulmonary laceration in which a tangential openingin the lung acted as a check valve, so that, during inspirations, graduallyincreasing amounts of air were trapped in the pleural cavity. The flap of lungprevented the escape of air into the bronchial tree on expiration. Most tensionpneumothoraces were of this origin.
2. It might result from small perforating associated woundsof the subglottic larynx or the cervical trachea. In the latter type of wound,air dissected along the fascial planes to the mediastinum and might break intoone or both subcavities. Tension pneumothorax of this origin was infrequent.
146
147
148
3. It might result from the ingress of air through the thoracicwall in a sucking wound.
4. It might follow the emergency closure of a sucking woundin forward areas, without provision for release of trapped air or the continuedescape of air into the pleural cavity.
An occasional tension pneumothorax occurred during operation forthoracoabdominal wounds (p. 103).
Incidence-Tension pneumothorax as the result oftrauma was encountered only infrequently in forward hospitals, though it wassomewhat more common as a postoperative or posttherapeutic complication. Itsinfrequency is evident in the statistics of the 2d Auxiliary Surgical Group. Itwas encountered by this Group only 11 times in its 2,267 thoracic andthoracoabdominal injuries. It is doubtful, furthermore, that all of these 11cases were examples of true tension pneumothoraces, since the valvelikemechanism that produces it and that permits easy ingress, but difficult egress,of air was not always found. In occasional cases in which air was found undermore than normal pressure, or even above atmospheric pressure, the explanationwas usually either compression of air by intrapleural bleeding or, most often,the irregular, splinting type of respiration that was the result of pain in thechest wall.
149
The infrequency of pressure pneumothorax probably has several explanations:
1. The preponderance of shell-fragment wounds, which resulted in apreponderance of large defects of the chest wall. For physical reasons, aircould not be entrapped in the chest in such wounds.
2. The routine application of occlusive dressings tightly enough to end thesucking or blowing effect of the open chest wound but not so tightly that atension pneumothorax was built up.
3. The almost routine use of a flapper-valve decompressive catheter incollecting stations or battalion aid stations.
4. The excellent screening of inductees in World War II, as aresult of which preexistent restricting pleural adhesions, which were ratherfrequent in World War I soldiers, were extremely uncommon.
The majority of tension pneumothoraces observed in World War II resulted fromcontinued leakage of air from the lacerated pulmonary parenchyma. If such alaceration existed and was leaking air, closure of the external chest woundconverted the open pneumothorax into a progressive tension pneumothorax. Thistype of tension pneumothorax was also uncommon, possibly because bleeding fromthe pulmonary laceration usually stopped promptly and the hematoma thatsurrounded the torn parenchyma produced a sufficient area of nonaeration tolimit the amount of air that reached the pleural cavity until the laceration wassealed over by natural processes.
Continued leakage of air occasionally had unusual explanations. A retainedmissile might offer a physical obstacle to natural closure of the air leak inthe lung (fig. 44), or there might be a leak through a laceration in one of thelarger bronchi. Two unusual cases of tension pneumothorax were observed bysurgeons of the 2d Auxiliary Surgical Group:
In one case, the tension pneumothorax developed from a perforating wound ofthe subglottic larynx: the air dissected the fascial planes of the mediastinumand perforated into the pleura. Tracheotomy was necessary to short circuit thecolumn of air.
In the other case, an instance of crushing injury, the patient sustainedwounds of the lower esophagus, mediastinum, diaphragm, and stomach, and the airin the pleural cavity was swallowed air.
An occasional tension pneumothorax was of extreme degree (fig. 45). That itcould be survived was striking proof of the tremendous respiratory reserve ofthe young, well-conditioned fighting men who made up the battle casualties.
Clinical picture-Extreme dyspnea was the mostprominent symptom of tension pneumothorax. The patient preferred to sit uprightor semierect. His worried, anxious expression indicated that he was fighting forbreath.
150
151
Cyanosis, which might not be present originally, becameincreasingly prominent as mediastinal shift caused interference with cardiacaction. The veins of the neck also became prominent. Physical examinationrevealed hyperresonance of the involved hemithorax, with absence of breathsounds (the so-called silent resonant chest) and a shift of the heart andmediastinum to the opposite side.
Management
The emergency management of tension pneumothorax has been discussed underresuscitation (vol. I).
If emergency measures were not required (that is, if the compression of thelung did not exceed 25 or 30 percent and if dyspnea were not a feature), thepractice was merely to observe the patient carefully for 10 to 14 days. In mostinstances, pulmonary reexpansion had occurred by this time, and no treatment wasnecessary.
In cases of any consequence, pneumothorax was best handled by the promptinsertion of a water-seal intercostal catheter of sufficient size (30-36 F.) toproduce immediate decompression of the chest. As a rule, this procedure wasfollowed by prompt reexpansion of the lung (fig. 46), without which neitherresuscitation nor recovery could insure.
152
153
HEMORRHAGE
Incidence
There is surprisingly little discussion in the military literature ofhemorrhage caused by wounds of the chest. The explanation is probably threefold:
1. Casualties with bleeding from one of the vessels in the pulmonary hilus orfrom the other great vessels of the thorax or from the heart usually died on thebattlefield or in the battalion aid station. They seldom survived to reach afield hospital.
2. Bleeding from other vessels of the chest was not usually a problem.
3. Another explanation of the infrequency of severe bleeding in chest woundswas that the first effect of parenchymal bleeding into the pleural cavity wascompression of the lung. In many instances, hemothorax had the mechanical effectof ending the hemorrhage by compressing the blood vessels.
There were, of course, exceptions to the generalization that all casualtieswith bleeding from the major thoracic vessels died promptly. In theMediterranean theater, survivals were recorded after lacerations of the azygosand hemiazygos, and even after a laceration of the superior vena cava, but theywere most unusual.
Two casualties with vascular injuries of magnitude were observed by Maj.(later Lt. Col.) Paul C. Samson, MC, in a field hospital. In the first case, alaceration of the right innominate vein was controlled by packing, but deathfollowed a massive recurrence of the hemorrhage 24 hours later. The secondpatient presented massive bleeding at the left apex, presumably from thesubclavian vein. His condition was critical, but rapid reflection of theclavicle permitted the blind placement of mass ligatures and control of thehemorrhage before he became exsanguinated. Recovery ensued but was complicatedby a serious pleural infection.
Severe bleeding in chest wounds observed at field hospitalsmost often arose from one of the systemic vessels, such as the intercostal, thepericardiophrenic, or the internal mammary vessels. The most serious hemorrhagesoccurred from the internal mammary and the intercostal vessels, from both ofwhich it was likely to be brisk. On the other hand, while complete division ofthese or any other thoracic vessels might cause severe intrapleural hemorrhage,it was unusual to find active bleeding at operation, even if from 1,000 to 2,000cc. of blood had previously been aspirated from the chest. The explanation wasphysiologic: The reduction in blood pressure caused by the blood loss and thesubsequent reduction in blood volume had allowed the vessels to retract andbecome sealed off. The most frequent source of bleeding at operation was anintercostal artery that had been incompletely divided and therefore was unableto retract.
The actual infrequency of serious bleeding in thoracic wounds is evident inthe analysis of 200 penetrating and perforating wounds of the chest, including
154
55 thoracoabdominal wounds, observed in forward hospitals byMaj. (later Lt. Col.) Lawrence M. Shefts, MC, and Capt. (later Maj.) Ernest A.Doud, MC. Active bleeding from the intercostal arteries was found at operation10 times and from the internal mammary arteries, from the mediastinal vessels,and from the heart in 2 cases each. This is a combined frequency of 8 percent.
It had previously been the practice of these observers toexpose the intercostal vessels routinely at operation, for fear of hemorrhage.They abandoned the practice after finding in these 200 cases, 33 lacerations ortransections of these arteries in all of which bleeding had ceased by the timeoperation was done. In this series, all of the bleeding vessels were found inone of two areas, (1) the area extending from the origin of the vessel to theangle of the ribs or (2) the area extending approximately 6 cm. from the lateralborder of the sternum.
In 4 of the 200 cases observed by Major Shefts and CaptainDoud, bleeding of considerable magnitude occurred from lacerations in the upperhalf of the upper pulmonary lobes. It was readily controlled by suture.Hemorrhage of similar degree was not observed from lacerations of the sameextent in other pulmonary lobes, probably because the injuries in the otherlobes involved vessels of smaller size than the comparable lacerations in theupper lobes.
Diagnosis and Indications for Thoracotomy
Continued hemorrhage into the tracheobranchial tree wasreadily recognized by the reappearance of fresh blood after repeated attempts toclear the airway. Otherwise, external evidences of bleeding were few; in chestwounds, the most serious bleeding was usually internal.
It was not safe to make the diagnosis of intrathoracicbleeding solely on the basis of the pulse rate, blood pressure, respiration, andgeneral status. Some patients with actively bleeding intercostal vesselsremained in surprisingly good condition.
The most reliable guides to the existence of continued serious hemorrhagewere:
1. Failure of the blood pressure to rise after apparently adequatetransfusion.
2. A fall in a blood pressure that had risen to relatively normal levels.
3. Reaccumulation of 1,200 to 2,000 cc. of blood in the pleural cavity within24 hours after the initial aspiration of similarly large amounts.
4. Persisting severe anemia, in spite of blood replacement, as determined byserial hematocrit determinations.
5. Plotting of the probable course of the missile with reference to thevascular supply.
It proved difficult to generalize on how much bleedingconstituted life-endangering thoracic hemorrhage. The fact that from 1,000 to1,500 cc. of blood could be aspirated from the pleural cavity within 12 to 18hours after
155
injury was not, in itself, an absolute index of lethalbleeding. Nearly all the patients who had lost blood in these and even largeramounts responded promptly to massive blood replacement, and subsequentaspirations produced progressively smaller amounts of bloody fluid.
Decision for or against exploratory thoracotomy on the basis of the criteriajust listed proved entirely satisfactory during periods of relative quiet, whencasualties were admitted in small numbers and surgeons had time for deliberateconsideration and watchful delay. During rush periods, when they were confrontedwith large numbers of casualties, their judgment was often sorely tried. Theirdecisions became more accurate in proportion to the number of casualties theyencountered.
The general policy was followed of exploring any injury inwhich there was a reasonable doubt concerning the possibility of activeintrathoracic hemorrhage. In the really urgent cases, in which procrastinationwould have been disastrous, there was seldom any doubt of the need for surgery.
Management
The techniques employed at operation for intrapleural hemorrhagewere those ordinarily employed in vascular surgery. In hemorrhage from theinternal mammary artery, it was imperative that both ends of the damaged vesselbe ligated, since hemorrhage could occur from either end. If emergency access tothe great vessels proved necessary, it could be obtained at once by reflectingthe clavicle and cutting through the manubrium sterni obliquely with a Gigli'ssaw. This technique, devised by Col. Johan Holtz, chief consultant in surgery tothe Norwegian Forces in the United Kingdom, was preferable to section of theclavicle or to sternoclavicular disarticulation.
The prewar expectation that lobectomy would frequently be necessary inparenchymal lacerations, especially those extending into the hilus and involvinglarge vessels, was not fulfilled, nor was the expectation that totalpneumonectomy might be necessary in some cases (p. 18).
The possibility of the postoperative recurrence of apreoperative intercostal hemorrhage had always to be kept in mind when theintercostal vessels had not been exposed at operation. The experience of MajorShefts and Captain Doud (p. 153) showed that the policy of nonexposure was safe.There were no deaths from any cause in the first 35 patients in whom, on thebasis of their previous experience, they did not expose the intercostal vesselsat operation.
Case Histories
Case 1-Acasualty wounded by a high explosive shell fragment that perforated the rightupper chest had a sudden massive hemorrhage shortly after his admission to afield hospital. Blood filled the entire trachea and the primary branches of thebronchus.
Bronchoscopy cleared the trachea only temporarily. Hemorrhagecontinued from the orifice of the right upper lobe bronchus. The patient wastaken to the operating room
156
immediately. Hemorrhage could be controlled only by packingthis orifice and holding the packing in place by packing the entire right mainstem bronchus. It was then possible to perform a thoracotomy and suture theinjuries in the bronchus and lung. The patient made a good recovery and had afully expanded lung 1 month later.
Comment-As thiscase illustrates, massive hemorrhage was always a possibility in chest injuries,infrequent though it might be. This patient's life was saved by his immediatetransfer to the operating room and the bold and imaginative measures by whichbleeding was controlled.
Case 2-This patient wasadmitted to a field hospital 5 hours after he had sustained a penetrating woundof the left chest and penetrating wounds of the right upper arm and left neck.The wound of entrance in the chest injury was in the left part of the neck.Several attempts at aspiration of a large hemothorax were unsuccessful, and itwas concluded that the blood had clotted. The wound of the neck was debrided,and the track was explored in the area in which it ran superior to the firstrib.
Aspiration was not attempted for the first 3 days afteroperation. On each of the next 5 days, from 400 to 500 cc. of blood was removed.On the 11th day, the yield was 900 cc. of thin, bloody fluid. During this time,the patient received numerous transfusions.
On the 12th day after wounding, the patient suddenly went intosevere shock. Neither the blood pressure nor the radial pulse could be obtained.The respiration was slightly labored, and the skin was cold and clammy. Breathsounds on the left side of the chest were distant. The patient rallied afterreceiving 1,000 cc. of blood. Attempts at aspiration of the left chest were notsuccessful.
The following (13th) day, 400 cc. of blood-tinged fluid wasaspirated. The pulse ranged to 148. After the aspiration, under endotrachealanesthesia, a sickle-shaped incision was made, extending from the mid third ofthe clavicle to the fifth costosternal junction. The first rib was excised atthe sternocartilaginous junction. As the pleural cavity was entered, copiousamounts of blood poured out. The second, third, and fourth ribs were severed atthe sternocartilaginous junction, and through the exposure thus secured, itcould be seen that the bleeding was in the region of the subclavian artery andinnominate vein. The bleeding point was compressed by the finger until it couldbe controlled by gross clamping and mass suture of the vessels at the apex ofthe chest. The chest wall was closed with catgut sutures for the muscle andfascia and silk for the skin. A catheter was inserted into the eighthintercostal space.
Although blood had been started in both ankles and the rightarm as the operation was begun, the blood pressure was imperceptible during mostof the procedure. During the operation and the immediate preoperative period, atotal of 3,750 cc. of blood was given. An hour after operation, the bloodpressure was 100/60 mm. Hg, and oxygen was discontinued. Four hours later, theblood pressure was 102/80 mm. Hg.
The postoperative course was without incident for the next 13days. Then the hospital blew down in a severe snowstorm, and this patient, alongwith 700 others, was evacuated to another forward hospital, and 4 days later toa general hospital. When he was received in the latter, the intercostal catheterwas not draining, and he was dyspneic and distressed. Several days later, alarge drainage tube was inserted in the bed of the eighth rib, to drain awell-established empyema. Recovery thereafter was uneventful.
Comment-Inthis case, the direction taken by the missile and the early clotting of thehemothorax pointed to the possible laceration of a large blood vessel. In theabsence of shock and other evidence of gross bleeding, thoracotomy seemed to becontraindicated. Subsequent events showed that it would have been the betterprocedure. The subclavian artery and innominate vein had apparently beenpartially lacerated, and local infection completed the severance 12 days afterwounding. This patient's life was unquestionably saved by the introduction ofblood into three veins during the operation.
157
ACUTE GASTRIC DILATATION
Acute gastric dilatation was a frequent occurrence in both thoracic andthoracoabdominal wounds and was often sufficient to cause respiratoryembarrassment. Its recognition was important before operation. Unlessdecompression was carried out, regurgitation would occur, and gastric contentsmight be aspirated, with the usual serious consequence.
Routine decompression by the Miller-Abbott or Wangensteentechniques was carried out after operation in all thoracoabdominal wounds untilall risk of acute dilatation was past. The same method was used if thiscomplication developed, as it sometimes did, after surgery limited to thethorax.
EMPHYSEMA
Surgical emphysema was not frequent in World War II, probablybecause of the low incidence of preexisting pleuropulmonary disease in U.S.soldiers. It was of two types, superficial (subcutaneous) and mediastinal.
Subcutaneous Emphysema
The superficial variety of surgical emphysema resulted frompenetration of the lung by a missile or by indriven fragments of fractured ribs.In either type of wounding, air escaped into the subcutaneous tissue (1)directly from the lung if adhesions were present, or (2) indirectly, through themedium of an interposed pneumothorax. The trapped air then spread upward toinvolve the tissues of the neck and face or downward to involve the abdominalwall. Occasionally, the extremities were involved.
Diagnosis was readily established by the crackling sensation felt onpalpation of the involved area. Roentgenograms confirmed the presence of air inthe tissues. Recognition was important because of the possible association oftension pneumothorax. The only condition which required differentiation wasclostridial myositis, which is also characterized by crepitation in the tissues.
Superficial emphysema seldom required any special treatment. Once the openingin the pleura became sealed, which usually happened promptly, the air in thetissues was readily absorbed. If the pleural opening did not close and the airin the tissues increased progressively, a needle was introduced into the pleuralcavity to permit the air to escape.
Mediastinal Emphysema
Mediastinal emphysema (fig. 47) was a much more serious condition thansubcutaneous emphysema. Air escaped into the mediastinum because of dam-
158
159
age to the extrapleural, intramediastinal segments of the tracheobronchialtree. If it could escape into the pleural cavity, a tension pneumothoraxdeveloped. If it had no route of escape, there was serious danger of compressionof the great vessels and even of the esophagus.
Diagnosis was easily made by palpation of the soft swelling in thesuprasternal notch. It was confirmed by roentgenograms.
As a rule, decompressive intubation of the pleural cavity was sufficient forrelief of the condition (figs. 48 and 49). Tracheotomy was sometimes necessary.Radical treatment was indicated only when the symptoms of compression weresevere enough to endanger life. In such cases, the swelling in the neck could beincised, or, if this measure was not effective, cervical mediastinotomy could beemployed. So far as is known, this operation was not required in any patient inthe Mediterranean theater.
160
161
PULMONARY EDEMA
Pulmonary edema was a possibility in all thoracic wounds andfrequently occurred with wounds in other parts of the body. It was most oftenassociated with, or caused by, one of the following conditions:
1. Severe pulmonary contusions, in which it tended to appear early.
2. Dyspnea and anoxia, in both of which, as Drinker and Warren (1)showed, pulmonary transudation is frequent.
3. Too rabid administration of intravenous fluids, or the administration offluids in too large quantities.
4. Fat embolism, which most often was a complication of an associatedfracture.
5. Lower nephron nephrosis, a condition in with pulmonary edema seldomappeared before the fifth day after wounding.
The clinical manifestations of pulmonary edema chiefly took the form offrothy pulmonary secretions in excessive amounts. If right heart failure
162
FIGURE 50.-Technique of intermittent positivepressure oxygen administration in field and evacuation hospitals. A. Anestheticmachine with to-and-fro filter. B. Anesthetic face mask, held over nose andmouth by rubber sling. C. Manual compression of bag synchronous with eachrespiration (pressure 15-20 cm. H2O).
supervened, the symptoms and signs included: Hyperpnea, distention of theveins of the neck, and falling pulse pressure and systolic pressure.
Since pulmonary edema was usually secondary to some other pathologic process,which often was obscure itself, specific treatment was generally unsatisfactoryand often failed entirely. The basis of therapy was an attempt to increaseoxygenation, and oxygen given through a mask, under positive pressure (fig. 50),was sometimes distinctly beneficial. Repeated catheter suction was employed toremove the pulmonary secretions. If necessary, a small catheter was leftindwelling in the trachea, and oxygen was administered between aspirations. Afew patients were benefited by atropine sulfate (gr. 1/100) by vein.
Right heart failure was treated by prompt venesection of from 500 to 750 cc.of blood or by the application of tourniquets (the so-called bloodlessphlebotomy). It was difficult to distinguish from shock, but the differentiationhad to be made before either measure was instituted.
163
ATELECTASIS
General Considerations
Massive pulmonary collapse as the result of trauma to the chest, as well asof trauma to the abdomen, might involve all the lobes of one lung but more ofteninvolved only one or two. The precise amount of trauma seemed relativelyinsignificant. Collapse sometimes occurred from the pressure effects of bombs orshells. In such cases, as well as in instances of direct trauma, contrecoupcollapse might occur, with only minor involvement of the homolateral side.
The frequent existence, before wounding, of bronchitis and other upperrespiratory infections favored the development of atelectasis. Prolonged etheranesthesia also encouraged its development.
Postoperative atelectasis (fig. 51) was infrequent. In this variety, as wellas in the traumatic variety, the basic pathologic process was excessive fluidaccumulations that completely obstructed a bronchus. The explanation for itsvery low incidence was the routine and effective treatment of wet lung beforeand after operation.
164
Clinically, the patient with atelectasis presented a fargreater degree of dyspnea than might be expected from the collapse of a singlelobe. Cyanosis, though also more than might be expected, was not so prominent asdyspnea. The trachea was deviated to the side of the collapse, and cardiacdislocation toward the affected side was usual. A sudden rise of temperature,from normal to 103? or 104? F., was typical. Physical examination of the chestrevealed the usual signs of a nonaerating lobe or lobes.
Management in Forward Hospitals
It was imperative that atelectasis be treated promptly and vigorously.Persistence of the collapse was likely to result in pulmonary consolidation.Many chest surgeons, in fact, believed that the majority of postoperativepneumonias were preceded by some degree of atelectatic collapse. In this belief,a number of surgeons prescribed sulfonamide therapy as a prophylactic measureagainst postatelectatic pneumonia.
The treatment of established atelectasis was directed to thesame end as the treatment of wet lung; that is, aiding the patient to expelobstructing fluid from the tracheobronchial tree. He was positioned with theinvolved side up and was made to cough at intervals of 5 to 10 minutes. Ifcoughing was not effective, a curved, semirigid rubber catheter of the Magilltype was inserted into the trachea, and sudden suction, of considerableintensity, was instituted. Often, during intubation, the patient coughedstrenuously enough to clear out and aerate the involved lobe or lobes. Carbondioxide inhalations were sometimes used to increase the depth of respiration,and back slapping and blow bottles were also sometimes used.
If these measures were not promptly effective, bronchoscopy was employed,under local anesthesia. No analgesic was applied to the vocal cords or trachea.In a surprising number of cases, aspiration of only small amounts of mucus fromthe orifices of the bronchi of the involved lobes corrected the atelectasisimmediately. As experience increased, and especially as the efficacy of themethod came to be appreciated, it became more and more the practice to resort tobronchoscopy without delay when atelectasis of any serious degree wasencountered.
Typically, after spontaneous aeration or aeration following active treatment,rales appeared and persisted for several days, in spite of clinical androentgenologic evidence of satisfactory aeration and the disappearance ofdyspnea and cyanosis.
Management in Base Hospitals
Atelectasis encountered in base hospitals was also infrequent. The lobar typewas occasionally encountered in the first year of the experience in North Africaand Italy but was seldom observed later. The atelectasis observed in basehospitals was frequently complicated by infection and was practically alwaysthe result of inadequate bronchial drainage. This was in contrast to
165
the atelectasis seen in forward hospitals, in which the important problem waswet lung secondary to trauma and anoxia.
In the base hospital, atelectasis, like tracheobronchitis andpneumonitis, was likely to be present in casualties complaining of pain in thechest wall and the abdomen and in casualties with spinal damage. Under thesecircumstances, respiration was shallow and cough ineffective. When there wasincreased secretion of mucus or of mucopurulent sputum in bronchitis, varyingdegrees of tracheobronchial obstruction occurred, and atelectasis was always apossibility. Catheter suction or bronchoscopy was the correct treatment,supplemented by intercostal nerve block if pain was a feature.
In 338 wounds of the chest treated at the 21st General Hospital,Mediterranean theater, there were 2 deaths directly due to lobular pneumonia,and the same complication played a possible part in a third fatality. Noinstance of lobar atelectasis was observed after operation, however, because ofthe vigorous measures routinely instituted to prevent its development.
In the single case of total pulmonary atelectasis observed inthis series, a nonspecific type of bronchitis was found by bronchoscopy on theinvolved side. No real improvement followed standard therapy. Pulmonarytuberculosis was suspected, but it required numerous sputum examinations toestablish the diagnosis. In retrospect, it was concluded that in this case, asubclinical pulmonary tuberculosis was activated by the pyogenic infection thatwas responsible for the pulmonary collapse. The patient had had a normal chestroentgenogram 2 months before he sustained the severe contusion of the chestthat led to atelectasis.l
HEMATOMA OF THE LUNG
General Considerations
Hematomas of the lung (figs. 52, 53, and 54), which werecommon, were found in association with all types of chest injuries. The degreeof pathologic involvement ranged from simple contusions with smallextravasations of blood into the pulmonary interstitium and, to some extent,into the alveoli, to massive interstitial hemorrhage. The latter variety ofhematoma was associated with intra-alveolar hemorrhage which involved wholelobes and even the whole lung.
In general, the clinical significance of a hematoma was in direct proportionto the amount of pulmonary parenchymal tissue involved. In the majority ofcases, the entity represented nothing more than a roentgenologic finding. In themore severe cases, the process involved sufficient pulmonary tissue to give riseto dyspnea and even to cyanosis.
1So far as is known, active tuberculosis did not occurafter any battle injury in the Mediterranean theater, and there is no knowninstance in which reactivation of arrested disease could be attributed to battletrauma. These circumstances are testimony to the effectiveness of the screeningmethods used in the selection of combat personnel.
166
From a diagnostic point of view, hematomas were occasionally confused withintrapleural collections of fluid or with atelectasis. Thoracentesis andbronchoscopy were frequently required to exclude these diagnostic possibilities.An occasional hematoma was associated with a febrile response and hemoptysis,which might lead the unwary to suspect pulmonary suppuration. The diagnosticconfusion might lead, in turn, to the performance of incorrect drainageprocedures.
Management
Once the diagnosis of a hematoma was established, the accepted therapy wasentirely nonsurgical. No lesion encountered in all the broad spectrum ofthoracic trauma demonstrated more dramatically the remarkable recuperativepowers of the lung than did massive pulmonary hematomas. As a rule, serialroentgenologic examinations showed slow but steady resolution of the process andcomplete clearing of the lung.
Hematomas associated with retained foreign bodies of considerable size ofcourse represented a different therapeutic problem, as did those associated withlong missile tracks (fig. 52) and with peripheral lacerations. In such cases,operative interference was frequently necessary for removal of the foreign bodyor closure of the laceration. If a foreign body was present, it was wise to waitfor complete resolution of the pathologic process and for stabilization of thepatient before an attempt at removal was made.
167
FIGURE 54.-Posteroanterior roentgenogramshowing sharply defined intrapleural hematoma.
168
169
Case History
The following case history is typical of the progress in most hematomas ofthe lung:
Case 3-This patient was wounded on 24 May 1944. A large shell fragmentpenetrated the left chest and caused a laceration of the lung and ahemopneumothorax on that side. The wound was debrided, the sucking wound closed,and the shell fragment removed from the chest wall. Repeated thoracenteses wereperformed.
Roentgenologic examination 6 days after wounding showed a large hematoma ofthe left lung (fig. 55A). Clinically the patient's course was smooth anduneventful. Four weeks after wounding, the hematoma was very much smaller (fig.55B), and 6 weeks later, roentgenologic examination showed almost no evidence ofit (fig. 55C).
BRONCHOPLEURAL FISTULA
General Considerations
Bronchopleural fistulas of traumatic origin were of two types.
In the first group, the fistula occurred without seriousparenchymal damage and without pleural infection or significant pulmonaryinfection. It represented a disruption of the visceral pleura and underlyingperipheral lung,
170
with the establishment of pleuroalveolar or pleurobronchiolar and bronchialcontinuity, or with both.
These fistulas, which were usually small, might be caused in several ways:
1. By erosion or laceration of the lung from sharp, fixed rib fragments.
2. By the reopening of peripheral lacerations of the pulmonary parenchyma.
3. By the avulsion of the line of repair of a pulmonary laceration that hadpreviously been sutured.
4. By necrosis around metallic foreign bodies or indriven costal spicules.
5. By denudation of lung tissue as the result of its being stripped from anadherent area, either by collapse of the lung because of the accumulation ofexudates or transudates or by expansion of the lung after thoracentesis.
In the second group of bronchopleural fistulas, the fistulawas part of a serious total situation, and the surgical measures necessary tocorrect it were part of the major surgery required to cane for whatever pleuralor pulmonary damage might be present. The procedure most often included drainageof an empyema by rib resection, decortication of the lung (occasionally withsegmental pulmonary resection), and, in very occasional instances, drainage of alung abscess.
Although bronchopleural fistulas were always a potentialcause of tension pneumothorax, they were seldom associated with it. A fewpatients complained of a tight feeling in the chest, but most of them wereunaware of what had happened. In a small number of cases observed in theMediterranean theater, in which the fistulas developed during evacuation fromforward to base hospitals, the tension pneumothorax was of considerableproportions, and the patients arrived dyspneic and in generally poor condition.Removal of the air by needle resulted in only temporary relief, but permanentrelief of symptoms and prompt healing followed the use of water-seal intercostaldrainage.
Diagnosis was made by the discovery of a pneumothorax, eitherby physical examination or roentgenologic examination, or by both, in a patientwhose chest was previously free of pleural air and who had had no breakdown of achest wall wound to cause an open pneumothorax.
Management
Treatment of bronchopleural fistulas had three purposes: (1) To bring aboutprompt removal of the air in the chest, (2) to obtain complete reexpansion ofthe lung, and (3) to close the fistula so that firm and permanent healing wouldfollow. Treatment was instituted as soon as the diagnosis was made.
After manometric determinations of the intrapleural pressure, a smallintercostal catheter was inserted under local analgesia and via trocar into thechest at the site of the pneumothorax. Since most bronchopleural fistulas gaverise to apical air pockets, the most usual site of insertion was the secondinter-
171
space in the midclavicular line. The site, however, corresponded with thelocation of the air pocket, which, because of adhesions, might be restricted tosome other portion of the chest. The catheter was connected to a long rubbertube, the end of which was kept under water, to provide closed drainage. Thisarrangement permitted air to escape from the pleural cavity only when thepressure in it rose above atmospheric pressure. The lung was thus allowed toexpand slowly. At the end of 48 hours, and sometimes by the end of 24 hours, thecatheter could be withdrawn. If the fistula reopened, as occasionally happened,the catheter was reinserted.
The conservative method of treatment just described was always given a fairtrial, even though catheter drainage had to be instituted more than once. Itgave very satisfactory results. In 870 penetrating wounds of the chest analyzedby Maj. Thomas H. Burford, MC, from the 2d Auxiliary Surgical Group experience,there were 22 bronchopleural fistulas of the type just described. Twenty-onewere closed permanently by the conservative method outlined, though one patientexperienced three recurrences of the fistula before solid sealing was obtained.In the remaining case, operation was resorted to after repeated attempts atcatheter suction had proved unsuccessful. No patient in the group developed apleural infection, which is highly significant, since in several instances, thefistula persisted for 2 weeks or more. It was the general experience thatfistulas not associated with significant parenchymal damage were not a source ofpleural infection if the pleural cavity did not contain blood.
If intrapleural infection did occur, operation was performedat once. Otherwise, thoracotomy for repair of the fistula was not instituteduntil the full possibilities of thoracentesis and intercostal water-sealdrainage had been exhausted.
In some cases, operation was necessary to smooth dangerously sharp rib endsor remove indriven rib spicules or metallic foreign bodies. Only occasionallywas it necessary to operate for the fistula alone. This was fortunate, for thedifficulties of finding a small fistula at an open operation were oftenconsiderable.
The management of bronchopleural fistulas that were a part of other majorpathologic processes did not usually furnish problems referable to them alone.Occasionally, however, pressure pneumothorax appeared suddenly in patients withempyema, before dependent drainage had been instituted. Intercostal catheterdrainage was instituted at once, by the technique just described.
Case History
A typical instance of bronchopleural fistula treated by the method justdescribed follows.
Case 4-This soldier, who was wounded in action on 17September 1944, sustained a perforating wound of the right thorax, withlacerations of the right lung, right diaphragm, and liver. A hemothoraxdeveloped promptly. He was treated by thoracolaparotomy,
172
173
with suture of the diaphragm and subcostal drainage of the hepaticlaceration. The pulmonary laceration was not leaking and was therefore notsutured.
The immediate postoperative course was uneventful. Biliarydrainage ceased, and the drains were removed. The chest also cleared, but 2weeks after wounding, the patient began to complain of tightness in the righthemithorax, in which physical examination revealed the presence of air.Roentgenologic examination (fig. 56A) confirmed this finding. Manometric studiesalso furnished evidence of a small bronchopleural fistula of the right lung.
A small catheter was inserted into the pneumothoracic space,through the second intercostal space in the midclavicular line, and was attachedto a water-seal bottle. Twenty-four hours later, the tube had ceased to bubbleair. At the end of 48 hours, the water column had ceased to oscillate, and thetube was removed. Examination at this time revealed total disappearance of thepneumothorax and complete reexpansion of the right pulmonary apex. Thesefindings were corroborated by roentgenograms (fig. 56B). The fistula did notrecur.
WOUND DISRUPTION
General Considerations
The causes of wound disruption in thoracic injuries and theconsequences of the accident are best presented in an analysis of 10 casesobserved at the 21st General Hospital in the Continental Advance Section afterthe invasion of southern France.
The 10 disruptions occurred in 74 sucking wounds, whichoccurred, in turn, in 338 battle casualties admitted to the thoracic service ofthis hospital. All but one of the patients had been treated in U.S. Army fieldhospitals. This man had received his initial treatment in a German hospital.
Analysis of these 10 cases of wound disruption revealed the following data:
1. In eight cases, the damage to the chest wall was severe, and there hadbeen considerable loss of soft tissue.
2. Rib fractures were present in all cases, the number of fractures rangingfrom one to five.
3. Pulmonary damage was severe in all cases.
4. In two cases, the wound of the chest wall had been contaminated byintestinal contents from a thoracoabdominal wound.
5. Six wounds were in the anterior chest wall, in whichmuscles are thin and difficult to draw together. Three were located laterally,and the remaining injury was close to the spine.
6. Debridement had obviously been inadequate in the two small wounds in theseries and in another large wound, the latter treated at the German hospital.
7. The wound had been left open in two cases. One of these patients had beentreated at the German hospital.
8. Thoracentesis had been performed after operation in only three cases.
9. Eight patients had bronchopleural fistulas, bronchitis, and severe wetcoughs. In none of these cases had any measures been instituted to improvebronchial drainage.
174
10. Empyema was present in every patient, and in only threeinstances had there been an attempt to drain the pleural cavity. Infection inthe pleural cavity and lung, as well as in the chest wall, was thus an importantfactor in all of these disruptions.
11. In six instances, respiratory embarrassment was the result of mobility ofthe mediastinum.
12. All but one patient had had penicillin systemically, and at least threehad had it intrapleurally also.
Management
When these patients were received in the base hospital,disruption of the chest wound had, in every instance, produced, or reproduced, asucking wound. Many patients were admitted in severe respiratory distress as aresult of the anoxia which had developed during evacuation. Closure of the woundwas a secondary consideration in every instance. All the patients needed othertreatment far more urgently.
The following treatment was instituted:
1. The wound was occluded with an airtight, petrolatum-impregnated gauzedressing, kept in place by strips of adhesive tape or by an Ace type of bandage.Because of its elasticity, this type of bandage permitted expansion of the lungwhile at the same time it furnished necessary support to the paradoxicallymoving chest wall.
2. Intercostal nerve block was carried out if respiration was painful orparadoxical.
3. Oxygen was administered to all patients.
4. Tracheobronchial suction was used if the cough was ineffective.
5. Adequate pleural drainage, which was the keystone of treatment, wasinstituted as soon as these emergency measures had been carried out.
In six cases, the pleura sealed over without other treatment.In three cases, redebridement was necessary because the wounds containednecrotic muscle tissue. Muscle flaps were developed and the wounds closed inlayers with interrupted silk sutures. The remaining patient, whose wound hadbeen secondarily closed at an evacuation hospital, had a second rupture, whichwas satisfactorily closed by the same measures.
In other words, when the multiple etiologic factors of wounddisruption in these 10 cases were under control, wound closure was possible,either by surgical closure or by spontaneous closure of the pleura aftersatisfactory drainage of the intrapleural infection. The chest wall defect inthe latter cases was allowed to heal by granulation, after which the skin edgeswere drawn together by adhesive tape. With proper initial management, it isdoubtful that any of these wound ruptures would have occurred.
175
CLOSTRIDIAL MYOSITIS
Clostridial myositis was extremely infrequent in wounds ofthe chest. When it did occur, the etiologic factors and clinical manifestationswere essentially the same as in clostridial infections in civil life. Theprinciples of management were also the same.
A point of great importance was that subcutaneous emphysema be clearlydistinguished from clostridial infections. The chief point of differentiationwas that in emphysema, except in the more extreme cases, air was confined tosubcutaneous tissues, while in clostridial infections, it was within the muscleplane.
CHYLOTHORAX
There was no instance of chylothorax in any of the thoracicor thoracoabdominal wounds cared for by surgeons of the 2d Auxiliary SurgicalGroup. The assumption is that most patients with these injuries died on thebattlefield from their severe wounds.
LUNG ABSCESS
Incidence
The incidence of lung abscess in World War II was extremelylow. Preliminary figures compiled by the Medical Statistics Division, Office ofThe Surgeon General, Army, based on sample tabulations of individual medicalrecords (table 10), show only 541 cases for the entire war, 382 in thecontinental United States and 159 outside of the United States. There were 16deaths, 10 in the continental United States. Of 255 cases of lung abscess (basedon 20 percent sample tabulations of individual medical records) classified assecondary (table 11), 130 occurred outside the continental United States. Thefigures include both battle and nonbattle cases, and not all of the abscessessecondary to battle wounds were related to wounds of the chest.
The statistical evidence of the infrequency of lung abscessas a complication of battle wounds of the chest supports its infrequency inindividual studies and also supports the clinical impression of the World War IIchest surgeons. In one analysis of 870 battle wounds of the chest, by MajorBurford and eventually included in the total figures of the 2d AuxiliarySurgical Group, there were only 5 instances of lung abscess. Maj. (later Lt.Col.) John Burke, MC, and Capt. (later Maj.) Theodore T. Jacobs, MC, in ananalysis of 122 penetrating wounds of the chest at the 23d General Hospital inthe European theater, encountered only 2 lung abscesses. Other observers hadsimilar experiences.
The explanation of the infrequency of this complication inbattle-incurred wounds of the chest is twofold, (1) penicillin therapy and (2)the intensive efforts immediately after wounding to restore the normalrespiratory physi-
176
[Preliminary data based on sample tabulations of individualmedical records]3
| Area | 1942-45 | 1942 | 1943 | 1944 | 1945 |
| Continental United States | ADMISSIONS | ||||
| 382 | 81 | 132 | 89 | 80 | |
Overseas: |
|
|
|
|
|
| Europe | 60 | 2 | 8 | 15 | 35 |
| Mediterranean4 | 25 | 1 | 13 | 6 | 5 |
| Middle East | 5 | --- | 5 | --- | --- |
| China-Burma-India | 14 | 1 | 2 | 1 | 10 |
| Southwest Pacific | 24 | 2 | 4 | 13 | 5 |
| Central and South Pacific | 22 | 1 | 6 | 5 | 10 |
| North America5 | 4 | 1 | 2 | 1 | --- |
| Latin America | 5 | 2 | 3 | --- | --- |
| 159 | 10 | 43 | 41 | 65 |
| 541 | 91 | 175 | 130 | 145 |
| Continental United States | DEATHS | ||||
| 10 | 1 | 7 | 1 | 1 | |
| Overseas: |
|
|
|
|
|
| Europe | 4 | 1 | 1 | 1 | 1 |
| Mediterranean4 | 1 | --- | 1 | --- | --- |
| Middle East | 1 | --- | 1 | --- | --- |
| China-Burma-India | --- | --- | --- | --- | --- |
| Southwest Pacific | --- | --- | --- | --- | --- |
| Central and South Pacific | --- | --- | --- | --- | --- |
| North America5 | --- | --- | --- | --- | --- |
| Latin America | --- | --- | --- | --- | --- |
| 6 | 1 | 3 | 1 | 1 |
| 16 | 2 | 10 | 2 | 2 |
1It is estimated that virtually none were CRO (cardedfor record only) cases. For the 2 years, 1943 and 1945, in which the number ofCRO cases was known, the 320 admissions included but 1 CRO case.
2Underlying cause of death; year of death.
3Complete files of records used for deaths, 1942admissions, and oversea admissions in 1943. Samples of admissions were: 20percent for 1945, U.S. 1943, and Europe 1944; 80 percent for 1944, excludingEurope.
4Includes North Africa.
5Includes Alaska and Iceland.
177
TABLE 11.-Number of secondary cases1 of longabscess in the U.S. Army, by area and year,1944-45
[Preliminary data based on 20 percent sample tabulations of individualmedical records]
| Area | 1944-45 | 1944 | 1945 |
|
| Number | Number | Number |
| Continental United States | 125 | 75 | 50 |
| Overseas: |
|
|
|
| Europe | 60 | 15 | 45 |
| Mediterranean2 | 20 | 15 | 5 |
| Middle East | --- | --- | --- |
| China-Burma-India | --- | --- | --- |
| Southwest Pacific | 45 | 20 | 25 |
| Central and South Pacific | --- | --- | --- |
| North America3 | --- | --- | --- |
| Latin America | 5 | --- | 5 |
| Total overseas | 130 | 50 | 80 |
| Total Army | 255 | 125 | 130 |
1Battle and nonbattle cases.2
Includes North Africa.3
Includes Alaska and Iceland.
ology. Since lung abscesses were infrequent even beforepenicillin became available in the spring of 1944, the second of theseexplanations must be considered the more important.
The low incidence of lung abscess can also be interpreted asfurther evidence of the remarkable recuperative capacity of the lung,particularly when, as just mentioned, its inherent capacity was supported byprompt measures to assure a patent airway and total pulmonary reexpansion.Consistent with the infrequency of localized intrapulmonary suppuration was theobservation that bacteria were seldom cultured from missile cavities whenforeign bodies were removed in uncomplicated cases (p. 327).
There was no explanation, however, of why pulmonary abscessesdeveloped about an occasional foreign body and not about others. One explanationadvanced was that while the heat of shell fragments might be sufficient tosterilize most pathogenic bacteria, bits of clothing driven into the lung alongwith the missile might carry infecting micro-organisms in with them. This is notan acceptable theory, partly because this occurrence was so frequent and lungabscesses so infrequent, and partly because the infecting micro-organisms foundin lung abscesses were usually resident respiratory flora. The number of casesis too small to permit anything beyond the mere statement of facts, but it mightbe mentioned that the abscess formed about a foreign body with recognizable bitsof cloth attached to it in three of the five cases reported by Major Burford.Other, isolated observations were to the same effect.
178
Management
The type of lung abscess seen in battle-incurred injuries differed from thetype seen in civilian practice. In civilian practice, the abscess begins with adiffuse area of pneumonitis, and, if progress is favorable, as it usually is,the process proceeds to localization. In the traumatic type of abscess, thelocalized phase was the first phase. This was a distinction of fundamentalimportance from the standpoint of treatment: In the traumatic type of abscess,the earlier the attack, the simpler the procedure, and the better the results.
In four of the five cases studied by Major Burford, the abscess was excised,the excision including the foreign body in the three cases in which the missilewas retained. Drainage was instituted in the remaining case. All five patientseventually recovered, though in two instances, the abscess recurred and requiredsecondary drainage. Drainage of an associated empyema was necessary in both ofthese cases.
Penicillin therapy and the intelligent use of whole blood were importantparts of the routine of management, but good surgery was the basis of treatment.
Massive gangrene of the lung was not observed by surgeons of the 2d AuxiliarySurgical Group, but in two instances, extensive suppuration developed,apparently on the basis of extensive vascular damage. The process in both caseswent on to multilobar areas of pneumomalacia with multiple bronchopleuralfistulas and widespread pleuropulmonary sepsis. One patient recovered aftermultiple drainage operations. The other died after total pneumonectomy.
MEDIASTINAL ABSCESS
Surgeons of the 2d Auxiliary Surgical Group encountered abscess formation ofthe anterior mediastinum in only one instance. The infection developed about aretained foreign body, and the organism recovered was a pure strain of hemolyticstreptococcus. Complete and prompt recovery, without complications, occurredafter thoracotomy, removal of the foreign body, evacuation of the abscess, andintrapleural instillation of penicillin.
An occasional patient was observed in forward hospitals with localizedabscesses of the posterior mediastinum or diffuse posterior mediastinitis. Theseinfections were considered the only indication for posterior extrapleuralmediastinotomy in war wounds of the thorax. None of this small group of patientssurvived to reach base hospitals.
PNEUMONITIS AND PNEUMONIA
Infectious pulmonary complications, chiefly pneumonitis andthe pneumonias, were considerably more infrequent in chest injuries than hadbeen expected, in view of the World War I experience. The low incidence, whichwas
179
practically the same in thoracic and thoracoabdominal wounds, could beexplained in several ways:
l. The attention devoted to the correction of wet lung both before and afteroperation. Measures that drastically reduced the incidence of atelectasis, asthe correction of wet lung and maintenance of a patent airway reduced it, alsodrastically reduced the incidence of pneumonitis and subsequent pneumonia.
2. The postoperative routine which emphasized coughing, frequent turning inbed immediately after operation, and early ambulation.
3. Perhaps the unwitting prophylaxis instituted by theadministration of sulfonamides and later of penicillin as part of the routine ofmanagement of all wounds.
Pneumonitis almost always developed on the basis of atelectasis or ofstagnation of excessive bronchial fluids. Pneumonia was more frequent in thewinter months, when it was often superimposed on a preexisting purulentbronchitis. An occasional lobar pneumonia was a primary pathologic process.
Diagnosis of pneumonitis and pneumonia was by the usual symptoms and signs(toxicity, increased fever, pulmonary consolidation) supplemented byroentgenologic examination.
Sulfadiazine was the drug of choice. It was frequentlyobserved that a pneumonia that had developed under penicillin therapy respondedpromptly to sulfonamide therapy. If stagnation of bronchial secretions was apersistent feature of the disease, and if the patient could not or would notcough effectively, either catheter suction or bronchoscopy was employed, evenwhen pulmonary infection was clearly established. No undesirable consequencesfollowed this practice.
Atelectasis and pneumonia caused by inadequate bronchial drainage were fairlycommon on neurosurgical and general surgical wards. The staffs of the thoracicsurgical wards saw many of these patients in consultation and assisted in theircare, particularly in the performance of catheter suction and bronchoscopy.
TRAUMATIC OSTEOMYELITIS
Neither osteomyelitis nor osteochondritis was frequent after trauma in WorldWar II, in contrast to their relative frequency in World War I. Both weremanaged by removal of the involved segments under systemic penicillinprotection. The wound was left widely open for 4 to 10 days. Then secondarysuture or staged closure was carried out, depending upon the rapidity with whichthe wound cleared up. To await sequestration of the bone was not in harmony withmodern surgical principles.
Osteomyelitis of the scapula was seen at least 10 times more frequently inthe early fighting in North Africa than later. The explanation was that in
180
the early months of the war, debridement about the scapulawas not sufficiently radical. Later, there was a better realization of theimportance of adequate initial wound surgery. Free incision with removal ofsequestra usually resulted in prompt cure.
CHRONIC SINUSES OF THE CHEST WALL
General Considerations
Chronic sinuses of the parietes were encountered in a small number ofpatients in the base area and in a larger number in Zone of Interior hospitals.The causes were chiefly as follows:
1. Injury to the costal cartilage, caused by the so-called gutter type ofwound, with destruction of the cartilage as a result. Sinuses of this origincontinued to drain until all diseased cartilage had sloughed off, or, better,had been excised.
2. Retained foreign bodies. Sinuses were particularlyfrequent when bits of clothing or other foreign organic material had collectedaround retained metallic fragments, but metallic fragments alone could beresponsible for the sinuses.
3. An unwise selection of suture material. In closing particularly largedefects of the chest wall, it was sometimes necessary to use pericostal orperichondral sutures to pull the ribs and cartilages together. Experience provedthat the use of heavy silk or wire or other nonabsorbable suture material wasunwise; healing was sometimes satisfactory, but many times these materials actedas retained foreign bodies. As the war progressed, pericostal and perichondralsutures were not used if they could be avoided. If they could not be avoided,absorbable suture material was used.
4. Un-united fractures or fractures that had healed with excessive scartissue and callus formation. These were infrequent causes of parietal sinusesand were seldom an indication for surgery.
5. Osteomyelitis of the ribs or other bones (fig. 57).
Management
The basis of management of these sinuses of the chest wallwas the removal of the offending retained material, whatever it might be. Theremoval of retained foreign bodies and unabsorbed suture material offered nodifficulties. If bone or cartilage was responsible for the sinus, the infectedarea was excised. Numerous exceptions were noted to the opinion that infectioninvolving cartilages in the fused portion of the thoracic cage requires completeexcision of the involved fused cartilage for permanent cure. As experienceaccrued, it was found that a less radical resection, involving only the infectedsegment,
181
was all that was necessary in most cases. If there was a recrudescence of theinfection, the more radical operation could be done as a secondary procedure.
At operation, the wound was opened widely and the offending material removed.The extent of the debridement depended upon the circumstances in the individualcase. Since all of these wounds were infected, the skin and subcutaneous tissueswere left open, as at initial wound surgery. Secondary closure of the incisionwas usually possible under penicillin protection from 5 to 7 days later.
DEFECTS OF THE CHEST WALL
Wounds causing large defects of the chest wall were frequent. They reflected,in a dramatic fashion, the effectiveness of modern weapons. They were alsoobserved frequently because adequate debridement of an initially large woundresulted in a sizable opening in the chest wall. This was particularly true atAnzio, where many patients with large traumatic thoracotomies would
182
183
not have survived except for immediate surgery in hospitals close to thefrontlines.
These large defects presented special problems of management.The necessary intrathoracic procedures were carried out with dispatch,particular attention being paid to removal of fragments of ribs and clothing,which were found more often, and in larger amounts, than in any other type ofchest wound. Debridement had to be even more thorough than usual, because it wasoften advisable to close the skin primarily, to enhance the security of repair.
Satisfactory restoration of the chest wall often required patience andingenuity. The procedures utilized to accomplish closure without dangeroustension varied with the number and location of the defects. Pericostal sutures,though generally undesirable, had to be used in some cases to approximate theribs and thus reduce the size of the defect. In flail segments of the ribs, itwas occasionally feasible to use wire to fix the ends. Muscle flaps were veryuseful, particularly in the anterior chest (fig. 58), in which tissues forrepair were sparse. The paraspinalis group of muscles was used for flaps low inthe posterior chest, where there was a similar paucity of tissue. In anoccasional case, the subscapularis was used, as suggested by Maj. (later Lt.Col.) Reeve H. Betts, MC, and Capt. (later Maj.) William M. Lees, MC, with thearm positioned to bring the scapula over the opening. If nothing else waspossible, closure was effected by reattachment of the hemidiaphragm.
ADHESIVE PLEURITIS
Pathologic Process
Deforming adhesive pleuritis, also known as infolded lung,was a fairly common complication of chest injuries. Its mechanism was notclearly understood but could be postulated as follows:
1. The blood present in the pleural cavity after wounding resulted incompression of the lung.
2. The lung was additionally compressed by an associated pneumothorax.
3. If the pleural blood clotted, it occupied a considerable portion of thepleural space.
4. In the simplest type of infolded lung (fig. 59), as aresult of the thoracic wound, the associated hemothorax, and the resultingdeflation of the lung, the diaphragm was elevated and splinted. The costophrenicfringe of the lung was dragged along the dome of the diaphragm and presumablybecame retracted. Since blood was invariably present in some amount in allthoracic wounds, symphysis occurred between the costophrenic fringe of the lungand the diaphragmatic pleura at what might be termed the high watermark.
5. Eventually, as the result of absorption of air, or ofaspiration of blood or air, or of both processes, the lung began to reexpand,filling the pleural space that remained by rolling out over the area of thepleural symphysis and expand-
184
ing those portions of its contour that could most conveniently occupy theavailable space. The upper portions of the lung were free to expand and fill theapex of the chest. The peripheral portion of the base could be filled only byadaptation of the lung to the available space.
The process just postulated furnished a satisfactory explanation for slightdegrees of pulmonary infolding. They were of no great clinical significance, andthey probably occurred much more often than they were recognized. On the otherhand, extensive degrees of infolding and distortion were serious. They sometimesoccurred as the result of adhesions between the visceral and parietal pleuralcoats or between opposed portions of the visceral pleura (fig. 59). Bizarredistortions were occasionally observed, as the result of a symphysis between thelateral costophrenic fringe and the mediastinum, along its diaphragmatic border.
Though the mechanism of infolded lung was not clearly understood, theconsequences were evident. Because the hemithorax was not completely filled bythe lung, areas within it that were not adequately filled by lung tissue werelikely to be filled by recurrent accumulations of pleural fluid. This fluid,after the blood in the chest had been removed, was uniformly serous. It formedonly in small amounts, but it had a tendency to reaccumulate after it wasremoved by aspiration.
If the adhesive process involved large areas of lung, pulmonary expansion wasconsiderably slower than when an uncomplicated hemothorax was present, andfilling of the hemothorax was considerably delayed. These observations
185
explained the contracture of the chest wall observed in such cases; it wasmerely Nature's attempt to reduce the size of the hemithorax.
Some degree of infolding was uniformly seen in associationwith massive intrapleural clotting. Most observers thought that ifroentgenologic examination showed obscuration shortly after wounding and showedclearing within a period of weeks of expectant treatment and vigorous breathingexercises, the patient had had an infolded lung and not an organizing hemothorax.An organizing hemothorax was likely to be irreversible without more activetherapy.
Diagnosis
The only condition easily confused with deforming adhesive pleuritis wasorganizing hemothorax. The differentiation was extremely important: Anorganizing hemothorax of any extent constituted a definite indication forthoracotomy and decortication; an infolded lung did not.
The differentiation was usually readily accomplished by roentgenologicexamination, but certain clinical features were also helpful. In organizinghemothorax, but not in infolded lung, the physical findings usually includeddullness on percussion over the affected area; decreased tactile fremitus;suppression of breath sounds; narrowing of the interspaces; and possiblyretraction of the mediastinum toward the affected side, with elevation andrelative immobility of the corresponding hemidiaphragm. Blood was usuallyobtained by thoracentesis.
Roentgenograms taken in infolded lung showed a generalized obscuration of thepleural space in the posteroanterior projection. In lateral films, there was atriangular posterior obscuration generally regarded as typical of the process.The opacity was produced by the infolding and malaeration of the affectedpulmonary area, as well as by the small, irregular, and usuallyintercommunicating collections of pleural fluid. It was the findings in thelateral roentgenograms which differentiated adhesive pleuritis from organizinghemothorax. In the former, the obscuration was patchy and diffuse because theprocess was not localized. In the latter, the opacity was posterior and basal.
Management
If within 3 or 4 weeks after wounding, an infolded lung had expandedsufficiently to fill the hemithorax, expectant treatment was continued. Eventhough the chest was somewhat contracted, it was thought that return of functionto the chest wall by the practice of breathing exercises would eventuallycorrect the deformity. Conservative management was indicated whenever serialroentgenograms showed that pulmonary reexpansion had been shared proportionatelyby all the lobes, so that the patient would not be left, for example, with afully expanded upper lobe and a seriously contracted lower lobe.
On the other hand, release of limiting adhesions by surgery was consideredjustified if, within 3 or 4 weeks after wounding, the lung had not expanded
186
sufficiently to fill most of the hemithorax or if serial roentgenogramsshowed that the filling had been accomplished predominantly by one portion ofthe lung at the expense of other portions. In most cases, though not in all,when filling of the hemithorax was long delayed, it was much more likely thatfailure of pulmonary expansion was the result not so much of infolding of thelung, even if it was present to some degree, as of compression of the lung bythe peel or rind of an organizing hemothorax. In that type of case,decortication of the lung was clearly indicated.
DIAPHRAGMATIC HERNIA
Up to 1944, the Medical Statistics Division, Office of The Surgeon General,Army, coded diaphragmatic hernia with a residual group of hernia conditions. Thedata for 1944 and 1945, when diaphragmatic hernia was coded separately, show atotal of 449 cases and 9 deaths (table 12). In 230 cases, diaphragmatic herniawas the secondary diagnosis.
Data on deaths were obtained from a complete file of records.Admissions in 1945 and for the European theater in 1944 were derived from a20-percent sample of records. Admissions in 1944 for areas other than theEuropean theater were derived from an 80-percent sample. Data on secondarydiagnoses were derived from a 20-percent sample. Not all of the herniasassociated with battle-incurred wounds were related to wounds of the chest.
Oversea Experience
The small samples from which the statistics just cited arederived do not permit generalizations. They are in correspondence, however, withthe impressions of surgeons both overseas and in the Zone of Interior thattraumatic diaphragmatic hernias were decidedly infrequent. Only 2, for instance,were observed in 1,028 intrathoracic wounds analyzed by Major Burford from the2d Auxiliary Surgical Group experience.
In the first of these cases, the hernia was not suspectedbefore thoracotomy, which was undertaken for removal of a missile from the leftlower pulmonary lobe. In the second case (fig. 60), which was not a combatinjury, the casualty sustained a crushing injury to the upper abdomen when ajeep was blown off the road, by the explosion of a mortar shell, and wasoverturned. Early thoracotomy revealed that the stomach and colon had herniatedthrough a large tear in the left leaf of the diaphragm. The tear extended to,but not through, the left crus of the hiatus. Convalescence was uncomplicated,and postoperative roentgenologic examination showed normal configuration of thediaphragm and stomach. Both of these patients were returned to duty in thetheater.
So far as is known, no instance of diaphragmatic herniationwas observed in the Mediterranean theater in a battle wound in which thediaphragm had been sutured. The early experience had demonstrated the risk ofsuturing the
187
TABLE 12.-Number of total cases (incidence), admissions,secondary diagnoses, and deathsfrom diaphragmatic hernia1 by area of admission andyear, 1944-45
[Preliminary data based on sample tabulations of individualmedical records]
| Area | 1944 | |||
| Incidence2 | Admissions3 | Secondary diagnoses4 | Deaths5 | |
| Continental United States |
146 |
86 |
60 |
4 |
| Overseas: |
|
|
|
|
| Europe | 40 | 10 | 30 | --- |
| Mediterranean6 | 18 | 8 | 10 | --- |
| Middle East | 9 | 4 | 5 | 1 |
| China-Burma-India | --- | --- | --- | --- |
| Southwest Pacific | 18 | 3 | 15 | --- |
| Central and South Pacific | 3 | 3 | --- | --- |
| North America7 | --- | --- | --- | 1 |
| Latin America | 5 | --- | 5 | --- |
| 93 | 28 | 65 | 2 |
| 239 | 114 | 125 | 6 |
| 1945 | ||||
| Continental United States |
115 |
70 |
45 |
2 |
| Overseas: |
|
|
|
|
| Europe | 40 | 5 | 35 | --- |
| Mediterranean6 | 15 | 5 | 10 | --- |
| Middle East | --- | --- | --- | --- |
| China-Burma-India | 5 | --- | 5 | --- |
| Southwest Pacific | 20 | 10 | 10 | --- |
| Central and South Pacific | 10 | 10 | --- | 1 |
| North America7 | 5 | 5 | --- | --- |
| Latin America | --- | --- | --- | --- |
| 95 | 35 | 60 | 1 |
| 210 | 105 | 105 | 3 |
1Battle and nonbattle cases.
2Sum of admissions and secondary diagnoses.
3Cases in which diaphragmatic hernia was the primarycause of admission to a medical treatment facility.
4Cases of diaphragmatic hernia secondary to, orconcurrent with, some other admission diagnosis.
5Underlying cause of death, year of death, and theaterof admission.
6Includes North Africa.
7Includes Alaska and Iceland.
188
189
diaphragm with catgut whenever there was any risk of bile leakage. Suturesof silk, cotton, or nylon held consistently. Bile, free of its normal confines,always makes trouble, and the success of diaphragmatic suture was in large partdue to recognition of this fact and provision of adequate drainage for asufficiently long period (p. 118).
Zone of Interior Experience
Diaphragmatic hernia of traumatic origin was seen only slightly snorefrequently in the Zone of the Interior than overseas. The 28 patients observedat the Kennedy General Hospital chest center, Memphis, Tenn., during the 3-year period of its operation may be taken as typical, and the analysis ofthese cases is the simplest method of presenting the main points in the clinicalpicture and management of this condition.2
2This analysis was made by Maj. Felix A. Hughes, Jr., MC; Maj. EarleB. Kay, MC; Lt. Col. Richard H. Meade, Jr., MC; Maj. Theodore R. Hudson, MC; andMaj. (later Lt. Col.) Julian Johnson, MC. All of these officers served atvarious times in the thoracic surgery center, Kennedy General Hospital.
190
191
Etiology and pathogenesis-In 16 of the 28 cases, thewounds were caused by shell fragments, or machinegun, rifle, or pistol bullets.In the remaining 12 cases, the hernias, which were not combat-incurred, werevariously caused or precipitated by traffic accidents (9), a stab wound,a fall, and jumping from a truck. In five of the combat-incurred hernias, theinjured diaphragm had been repaired overseas at the time of wounding.
In the 11 combat-incurred wounds in which the diaphragm hadnot been repaired overseas, the defect was so small that no herniation occurreduntil sudden strain or prolonged strain caused it to enlarge. One soldier, forinstance, had sustained a penetrating wound 13 months before, for whichthoracotomy was not considered necessary. Suddenly, while carrying heavy sacks,he was seized with severe upper abdominal pain. At operation, the entire stomachwas found in the chest (fig. 61). Another patient had sudden, acute symptomsafter he jumped from a truck. At operation, the colon was found herniatedthrough a defect only 1.5 cm. in diameter, and gangrene had already set in.Three other patients with known hernias of several months' duration alsodeveloped acute intestinal obstruction while under observation in the hospital.Emergency operation was successful in all three cases.
The first patient with a diaphragmatic hernia observed at the chest center atKennedy General Hospital furnished a useful lesson. He had sustained a bulletwound of the right chest in May 1943. A few weeks later, he showed noabnormalities of any consequence on roentgenologic examination (fig. 62A) andhad no symptoms of any kind. He continued well for a year. Then he began tocomplain of pain in the right lower chest, and roentgenograms showed a roundedmass overlying the diaphragm (fig. 62B). Exploratory thoracotomy was performed,and the mass was found to be a herniated portion of the right lobe of the liver.Undoubtedly, many cases of this kind have been observed in veterans' hospitalsand in private practice since the war. If there is no hemorrhage, gangrene, orperforation of the gastrointestinal tract, patients may accustom themselves toliving with a surprising degree of diaphragmatic herniation.
In no instance was it possible to correlate the hernia, either in size orlocation (fig. 63), with the anatomic areas of weakness in which such herniasare prone to occur. Obviously, however, increased abdominal pressure, producedby strain, as a number of cases in this series show, may be the immediate causeof herniations through previously created defects, even when the defect issmall.
Pathologic process-A definite sac was demonstrable in only two of thehernias in the series. A membrane frequently covered a part of the herniatedstructure, but it was not complete. The adhesions present in practically allcases between the affected structures varied in extent and character. At times,it was impossible to separate the herniated structures from the lung withoutsome pulmonary damage. The size of the opening in the diaphragm varied from adefect 1.5 cm. in diameter to one that involved the entire dome and extendedinto the pericardium.
192
193
The dome of the left diaphragm was the site of the hernia in 12 of the 28cases (fig. 63). Contusion defects were located more peripherally andpenetrating defects more centrally.
At operation, practically all the organs in the peritoneal cavity were foundherniated through the diaphragm, either singly or in various combinations. Atone extreme were asymptomatic herniations of the omentum, discoveredaccidentally at operation for other conditions. At the other extreme were twoherniations, each consisting of the kidney, spleen, stomach, small bowel, colon,and omentum. In one of these cases, the structures were densely adherent andcompletely covered the lateral portion of the diaphragm. In the other, there wasan 8-inch tear in the pericardium.
In one case, there was a constricting ring defect around theherniated stomach (fig. 64); if diagnosis had been delayed, ischemic changes,followed by perforation, would undoubtedly have ensued. In one instance, inwhich previous repair had been done elsewhere, the diaphragm had been severedfrom its attachments to the posterior chest wall, and a hydronephrotic,congenital bifid kidney had herniated through the defect. It was necessary tomobilize and remove the kidney before repair could be accomplished.
In one case, incarceration of the stomach resulted in progressive gastricdistention, and eventually the entire left pleural space was filled with theherniated organ, the lung on that side being completely atelectatic. After thedefect in the diaphragm had been enlarged, it was possible to express the airfrom the stomach through the nasal tube inserted before operation. When this hadbeen accomplished, the adhesions could be separated and the hernia reduced.
Clinical picture and diagnosis-Herniation of solidviscera was associated with vague pain in the lower chest but usually with noother symptoms. Herniation of hollow viscera gave rise to more varied symptoms.In 11 cases, all of combat origin, the initial symptoms were epigastric pain,especially after meals and during recumbency; flatulence; indigestion; anddyspnea, which was practically always present when there was considerablecollapse of the lung. Several patients stated that they could hear and feelgurgles in the chest.
If a hollow viscus suddenly herniated through the diaphragmatic defect, thesymptoms were more dramatic. In two instances, unconsciousness occurred, not tobe accounted for by associated injuries. The degree of pain and the presence ofnausea and vomiting depended upon the extent of the herniation and the amount ofobstruction it produced.
In some cases, as already mentioned, herniation of the omentum through thediaphragm was discovered unexpectedly, at operation for other conditions. In thefour cases in which the liver was the herniating organ, only a presumptivepreoperative diagnosis was possible. In one of these cases, the herniation,which was recurrent, was found in the course of exploration of a draining sinusof the chest wall. In two cases, the chest was aspirated, in the belief that thefluid in the herniated stomach and intestines was pleural fluid. No evidence ofdamage or infection was found at operation in either case.
194
195
196
Surgical repair-All 28 hernias in this series were successfully repairedthrough a transthoracic approach, usually under endotracheal anesthesia. Theexcellent exposure so easily obtained by this approach made it preferable to theabdominal approach. In a number of instances, the herniated structures extendedup to the dome of the thorax, and the adhesions present could have been handledonly with great difficulty through an abdominal incision.
Closure was eventually accomplished in every case in the series, though insome, at first glance, it was doubtful that enough diaphragmatic tissue remainedto be useful. After careful dissection, however, enough tissue was developed topermit satisfactory closure in every such case, though it was sometimesnecessary to dissect the extension of the diaphragm into the transversalismuscle before closure could be accomplished without tension. Two rows of silksutures were used, the second burying the first. When the diaphragm was detachedfrom the thoracic wall, secure closure was achieved by using mattress suturesthrough the chest wall. The sutures were tied over tubes, and another row ofinterrupted intrapleural sutures was placed.
In the first cases in this series, the phrenic nerve was notcrushed unless the line of the defect was such that contraction of thediaphragmatic fibers would exert a pull on the suture line. This technique waslater made routine because, in a case in which it had been omitted, a herniarecurred 5 hours after operation, presumably because of excessive movements ofthe diaphragm during tracheal aspiration. Successful secondary repair wascarried out several months later. There was no evidence that paralysis of thediaphragm retarded healing, and earlier experimental evidence by Meade (2)had shown that it did not decrease the tensile strength of the diaphragm.
Drainage was employed only three times, once when a gangrenous colon had tobe resected and twice when empyema had preceded the hernia. Water-seal drainagewas used in these three cases until the lung had fully reexpanded andobliterated the pleural space. There were varying degrees of pulmonary collapsein all cases in this series, but the lung reexpanded in all without difficulty,no matter how long the collapse had been present.
Postoperative management-The usual routine of postoperative managementwas employed in these cases. Gastric suction was always instituted, usually for72 hours. If the patient could not keep the airway clear by his own efforts,endotracheal aspiration was used.
Postoperative complications-There were only threepostoperative complications of any consequence. In one case, as alreadymentioned, there was an immediate recurrence of the hernia. In one case, inwhich the lung was damaged while it was being separated from the herniatedliver, a temporary pneumothorax developed. In one case, there was apostoperative recurrence of the empyema which had complicated the originalinjury and which had to be treated before repair of the hernia could beundertaken. In three other cases transient pleural effusions appeared, but theydisappeared promptly and did not delay recovery.
197
HERNIA OF THE LUNG
General Considerations
Hernia of the lung was an extremely infrequent complication ofcombat-incurred wounds, probably because of the inherent elasticity of pulmonarytissue. This property causes the tissue to contract and retract from the chestwall when intrapleural and intrapulmonary pressures become equal in the presenceof the opening into the pleural cavity caused by a chest wound. The concomitantpneumothorax and hemothorax that occur after wounding also act to preventpulmonary hernias. On the other hand, in view of the considerable number oflarge chest wall defects observed in forward hospitals, the rarity of pulmonaryherniations is remarkable.
According to Funk (3), who made a comprehensive review of theliterature in 1918, only three hernias of the lung were reported in the CivilWar, and none were recorded in the Crimean War or in World War I. In the SouthAfrican War (1899-1900), Makins (4), in an extensive experience, observedonly one case. Adams (5) found only 5 pulmonary hernias in 20,000 woundsof the chest in the Russo-Japanese War. Up to 1945, only 185 cases had beenrecorded, most of which were combat-incurred.
Occasional instances of hernias involving single pulmonary lobes wereobserved, but there was only one instance of true pulmonary hernia (fig. 65) inthe 2,267 thoracic and thoracoabdominal wounds in the 2d Auxiliary SurgicalGroup experience. Of the seven cases recorded from the section of thoracicsurgery, Walter Reed General Hospital, Washington, D.C., by Capt. (later Maj.)Elmer P. R. Maurer, MC, and Lt. Col. (later Col.) Brian Blades, MC, six werecombat-incurred, and the other was the result of trauma.
Traumatic hernias represent the largest group of pulmonary hernias. They havebeen reported after crushing wounds, stab wounds, and perforating wounds causedby rifle bullets and shell fragments. They occasionally appear after surgicalprocedures. The congenital and spontaneous varieties are decidedly infrequent.
In traumatic hernias, the first manifestation is the appearance, usually fromseveral weeks to several months after injury, of a tender swelling in the chestwall, which increases in size on forced expiration and disappears oninspiration. In acute traumatic hernias, the symptoms referable to the herniaare obscured by the symptoms arising from the injury itself.
The diagnosis is usually made without difficulty because of thecharacteristic behavior of the air-containing tissue. In four cases observed atWalter Reed General Hospital, however, the classical signs and symptoms ofpulmonary hernia were present, but at operation no lung tissue was found in thedefect in the chest wall. In three of these cases, the liver had herniated intothe defect. The clostridial infections which occur in crushing injuries of thechest wall may produce crepitant bulging in soft tissues, but the circumstancesmake it unlikely that the infection will be confused with a true hernia of thelung.
198
Management
Because of the infrequency of hernias of the lung, it is not possible to makemany definitive statements about their management. Emergency measures were notnecessary, and there was no indication for operation overseas. The symptomscould be easily controlled by pressure dressings over the defect, andspontaneous healing was always a possibility.
199
Except for such complications as might be introduced by thewound, there was nothing in the management of pulmonary hernias of traumaticorigin encountered in wartime that differed from the management of similarhernias in civilian practice. Operation was always performed under endotrachealanesthesia, to avoid risk when the pleura was opened widely, as well as to leavethe lung in the desired state of inflation after operation.
At Walter Reed General Hospital, tantalum plates wereemployed tentatively, but the method had to be abandoned because of theimpossibility of maintaining immobility of the plate on the constantly movingthorax. In the five true hernias in which operation was performed at thishospital, plastic repair was accomplished with structures of the chest wall only(rib, periosteum, and muscle.) The most important feature of the repair wascoverage of the defect in the chest wall with sturdy bone or with periosteumthat would produce bone. After appropriately located ribs had been divided andreleased from their periosteal envelopes, they could be shifted until the gapwas bridged. They were then anchored in the desired position by pericostalfixation, by suture to the fascia or adjoining periosteum, or by fastening themobilized rib to an adjacent rib with absorbable sutures threaded through drillholes.
An essential point of repair was the creation of a localpneumothorax, so to speak, by freeing of adhesions and of lung tissue about themargins of the hernia orifice. The pneumothorax thus produced acted as a cushionbetween the lung and the freshly repaired chest wall and reduced to a minimumthe chances that the lung would be forced into the chest wall if the patientcoughed or strained immediately after operation, when the repair was weakest.
Case History
Case 5-This 25-year-old soldier sustained a penetratingshell-fragment wound of the right chest on 15 January 1945. Treatment consistedof closure of the wound by suture, intercostal water-seal drainage for 3 or 4days, and three aspirations of the chest.
When he was admitted to Walter Reed General Hospital about 6 weeks later, hisonly complaint was mild pain in the right chest, not associated with hemoptysisor cough. A well-healed linear scar, 11 cm. long, was present in the rightanterolateral chest. A second, well-healed scar, 9 cm. in length, was presentalong the vertebral border of the right scapula. Percussion and auscultation ofthe chest revealed no abnormalities. Roentgenologic examination revealed ametallic foreign body, 7 by 7 mm., and several smaller metallic foreign bodies,in the right lung. A defect in the right sixth rib laterally measured 8centimeters. It was possible to insert three fingers into this defect for about1 centimeter. When the patient coughed, as well as on forced expiration, withthe glottis closed, there was a definite bulging through the defect.
Operation was performed on 8 June 1945, under endotracheal ether anesthesia,by the following technique (fig. 66):
An incision was made over the defect in the right anterolateral chest wall atthe level of the sixth rib. The skin and muscles were divided. When the muscleswere dissected free, the lung was exposed. It was found adherent to the chestwall. The adhesions were divided and the lung was separated from the chest wallover an area 10 cm. in diameter.
200
201
202
The periosteum of the fifth and seventh ribs was developed into flaps, whichwere sutured in place to bridge the defect. The first layer of closure consistedof intercostal muscles brought together with interrupted sutures of medium silk.The ribs were then approximated with retractors, and the two flaps of periosteumwere joined with interrupted sutures of medium silk. No effort was made toreinflate the lung, because, as just pointed out, it was desirable to maintain aresidual localized pneumothorax for a time. The muscles of the chest wall werebrought together with interrupted sutures of fine silk, and the skin wassimilarly closed, without drainage.
Convalescence was entirely uneventful. The wound healed perprimam. The soldier was sent on a convalescent furlough 4 weeks after operation,and when he was reexamined on 14 August 1945, the area of the defect was firm,and there was no evidence of recurrence of the hernia.
RESIDUAL SYMPTOMS
General Considerations
A large number of patients who had sustained chest injuries presented a triadof residual symptoms, consisting of shortness of breath, soreness of the chest,and paresthesias or pains in the affected side. These symptoms were observed atvarious times after wounding, in both oversea and Zone of Interior hospitals. Atthe Walter Reed General Hospital, they constituted the most serious obstacle toreturning men with chest injuries to full duty within a short period of time.
Dyspnea was usually mild, but it appeared on very littlephysical exertion, and most patients complained of reduced exercise tolerance.It was often necessary for them to fall out of ranks.
203
Pain was usually vague and usually paresthetic in nature. Thepatient distinguished it from soreness, which most of them complained of atnight, when they tried to sleep on the affected side.
Except for variations in degree, the symptoms were consistent. They bore nodiscernible relation to the type of injury sustained. They were as frequent insimple penetrating wounds of the chest with no consequences beyond hemothorax asthey were in more severe injuries. They occurred whether or not thoracotomy hadbeen performed. They also occurred whether the soldiers were returned to combatduty or were given limited-duty assignments.
Special Studies
When this problem became of importance in thoracic surgerycenters in the Zone of Interior in chest casualties who otherwise could havebeen returned to duty, it was found that very little was known about it. Therewere no subjective studies of dyspnea on exertion in such cases, and noobjective studies by determinations of the vital capacity to investigatepossible permanent damage.
At the Walter Reed General Hospital, a study of theintercostal pain and numbness which followed thoracic injury was undertaken byMaj. Donald L. Rose, MC. Three methods were used:
1. The patient was asked to outline the painful area on his chest wall.
2. A complete neurologic examination was made.
3. Skin resistance tests were carried out by the Richter technique, to checkthe impressions gained by other examinations.
With few exceptions, the patient's own zoning of pain pointson his chest wall corresponded with the results of the skin resistance tests. Inthe few cases in which the results of the examinations did not coincide, it wasthought safe to discount the symptoms.
These studies established three practical points:
1. There was no satisfactory treatment for damagedintercostal nerves once they had been injured. Intercostal nerve block providedtemporary relief, but, once the immediate effects had disappeared, the pain wassometimes more severe than it had been originally.
2. Injury to a single intercostal nerve did not cause serious residuals.
3. If three consecutive nerves were crushed or cut, prolonged discomfortfollowed, and it was serious from the standpoint of future vigorous activity.
Major Rose's study showed that residuals from the bruising of a single nervewere of little consequence, whether it was damaged during rib resection or inthe course of an intercostal incision. Injury of at least two nerves wasnecessary to produce symptoms. Pressure transmitted from the rib-spreadingretractor was apparently of little consequence.
204
Management
This problem was of extreme importance in military medicine.In civilian practice, it would have been considerably less serious. A civilianwith discomfort after a chest injury could generally avoid activity which mightaggravate it. The situation was quite different in military service, in whichthe soldier who was ordered to drill, carry a pack, or perform some otheractivity had no recourse but to obey orders as long as he was on active duty.
In military practice, furthermore, the medical officer wasobliged to make a decision as to the validity of the patient's complaint. He hadto consider seriously whether the pain was genuine and severe, or whether thesoldier was motivated by a desire to escape active duty or to be separated fromservice. If the surgeon discounted the symptoms and sent the soldier off to dutywhen he was really unfit for it, he might do him a serious injustice.
Since the etiologic factors in postwounding discomfort were not clarifiedduring the war, treatment was unsatisfactory. The variable results in nerveblock have already been mentioned. Psychotherapy was useful in some cases.
As a result of the study conducted at Walter Reed General Hospital, it wasconcluded that the solution of the problem was preventive, to exercise extremecare in handling intercostal nerves at operation and to avoid, as far aspossible, any surgical technique that might traumatize them unnecessarily. Thecorollary of this reasoning was that pericostal and perichondral sutures shouldnot be used. Encircling sutures of this kind had been used in a number of thepatients observed, with resultant temporary destruction of two or more nerves.
Surgeons at Walter Reed General Hospital believed that it was seldomnecessary to employ pericostal and perichondral sutures to accomplish asatisfactory closure of the chest wall. If the wound was posterior, particularlyif a segment of rib had been resected, it was not necessary to hold the ribstogether by sutures. Even when an intercostal incision was employed, there wasusually sufficient tissue to close the defect without fixing the exposed portionof the thoracic cage with encircling ties.
The following technique was found satisfactory for the closure of practicallyall surgical wounds of the chest:
Interrupted sutures were placed in the intercostal muscles while the ribs orcartilages were brought together with retractors. The placing of the instrumentswas important: The lower retractor must be placed not in the interspaceadjacent to the wound but at least one interspace lower. The pressure from thisretractor would be transmitted directly against the nerve on the inferior edgeof the rib if it were placed in the adjacent interspace. By placing it one ortwo interspaces lower, injury to two or three consecutive intercostal nerves,which this study had shown to be responsible for painful sequelae, could beavoided.3
3The reader is referred to chapter XI (p. 441) for long-term followupstudies on casualties with a number of the complications of thoracic woundsdescribed in this chapter.
205
References
1. Drinker, C. K., and Warren, M. F.: The Genesis and Resolution of PulmonaryTransudates and Exudates. J.A.M.A. 122: 269-273, 29 May 1943.
2. Meade, R. H., Jr.: The Tensile Strength of the Paralyzed Diaphragm.Preliminary Report. J. Thoracic Surg. 2: 503-516, June 1933.
3. Funk, E. H.: Hernia of the Lung, With Report of Case of Spontaneous Herniaof the Right Lung. Internat. Clin. series 28, 1: 102-107, 1918.
4. Makins, George Henry: Surgical Experiences in South Africa, 1899-1900.London: Smith, Elder, & Co., 1901.
5. Adams, E.: Hernia of the Lung, With Report of a Case of Spontaneous Herniaof the Right Lung. Am. Med. 6 (n.s.): 41-43, January 1911.
