Part I
SPECIAL TYPES OF WOUNDS OF THE CHEST
CHAPTER I
Special Types of Thoracic Wounds
Lyman A. Brewer III, M.D., and Thomas H. Burford, M.D.
WOUNDS OF THE CHEST WALL
The simplest form of chest wound, the type which involvedonly the thoracic wall and produced no bony or visceral injuries (fig. 1),always had one serious possibility: The pain which followed might causevoluntary restriction of respiration and of the cough reflex, which, in turn,could cause retention of secretions and introduce the problems of wet lung.Injuries which involved only the soft parts, such as abrasions and lacerations,were usually not serious. Those associated with fractures of the ribs could beserious. Contusions of the chest wall could also be serious.
An occasional patient who seemed to have only a simplecontusion of the chest wall developed pleural effusion, apparently because thepleura had shared in the trauma. In these cases, which were usually diagnosed astraumatic
FIGURE 1.-Superficial bullet wounds of chest wall notinvolving bony structures.
4
pleuritis with effusion, the fluid was straw colored or amberand had the characteristics of a transudate. Repeated aspiration was usually theonly treatment necessary, but in an occasional case, in which the pleural fluidhad a high fibrin content, clotting occurred, and the total picture wassuggestive of clotted hemothorax. If the condition remained static,decortication might be necessary.
Contusions of the chest wall were also responsible for some intrathoracichematomas (p.165).
Case History
The following case history illustrates the intrathoracic complications thatsometimes followed a wound of the chest wall not, in itself, of majorimportance:
Case 1.-This casualty was picked up by aidmen between1 and 2 hours after he was wounded by shell fragments in the right shoulder andleft knee. When he was received in an evacuation hospital 11 hours afterwounding, he was found to have a perforating wound of the right upper chest,associated with a moderate hemothorax. Wet lung was demonstrated clinically bymany loud rales and rhonchi and was confirmed by roentgenograms. Roentgenologicexamination also showed comminuted fractures of the posterior portion of thesecond and third ribs, and possibly of the fourth rib, on the right side. Noforeign bodies were visible on this roentgenogram nor on the roentgenogram ofthe left knee, which showed an epiphyseal fracture of the tibial tubercle.
Fourteen hours after the patient's admissionto the hospital the wound of entrance just below the outer third of the rightclavicle was widely debrided under local analgesia, and a large amount ofclotted and liquid blood was evacuated. A small fragment of the fractured rightclavicle was removed. The wound was sprinkled with powdered sulfanilamide, and apetrolatum-impregnated gauze dressing was applied. Paravertebral nerve blockwith procaine hydrochloride was carried out at the level of the seventh cervicalvertebra and the first through the fourth thoracic vertebrae. The hemothorax wasaspirated through the seventh intercostal space in the posterior axillary line,and the 500 cc. of blood thus obtained was used as an autotransfusion, with 500cc. of physiologic salt solution. Bronchoscopy was employed to clear profusesecretions from the tracheobronchial tree. The wound of the left knee wasdebrided.
The postoperative course was satisfactory for the first 4days. Then paravertebral block had to be repeated because of pain. No fluid wasfound in aspiration at this time. On the eighth day, 1,350 cc. of bloody fluidwas aspirated from the right chest. Repeated aspiration on the 10th day produced150 cc. of thick, pinkish fluid, and on the 12th day, 725 cc. of thin fluid ofthe same color was removed. Thereafter the chest was dry.
When the patient was evacuated on the 22d postoperative day,he had entirely recovered from his chest injury, but he still had some loss offunction of the upper right arm.
Comment-Contusion of the chestwall was present to some degree in all injuries of the chest in which directviolence was a factor. It undoubtedly played a part in the case just recorded,helping to produce both wet lung and hemothorax. There was no directcommunication between the pleural cavity and the wound of the chest wall.
Although this patient had an extensive wound of the chestwall, with hemothorax and wet lung, there was no indication for formalthoracotomy. The infection in the chest wall was controlled by adequatedebridement. The lung was expanded by needle aspiration of the blood in thepleural cavity, thus increasing the vital capacity, and intercostal nerve blockallowed the patient to breathe deeply and cough effectively with relief of thewet lung syndrome.
5
SUCKING WOUNDS
General Consideration
A sucking wound (fig. 2) is defined as a wound in which thereis more or less free interchange of air through the wound between the pleuralcavity and the outside atmosphere. The term, as pointed out elsewhere (vol. I),is both unsuitable and misleading, and blowing wound is preferable nomenclature.Sucking wound, however, has the sanction of usage, and for that reason, its useis continued in this volume.
Potentially, if not actually, all penetrating and perforating wounds of thechest observed in World War II were sucking wounds. A sucking wound sometimessucked (blew) constantly and sometimes only when the skin and muscle planes werein a certain alinement. The regional anatomy frequently determined whether ornot a wound sucked. Generally speaking, a large wound was more likely to suckthan a small wound, but size was not the most important determining factor. Arelatively small wound in the anterior chest wall, 2 cm. or less in diameter,might suck constantly because in this area the muscles are scanty and theinterspaces wide. A larger wound located posteriorly, in the heavy muscles ofthe back, might be less likely to suck.
Other considerations also determined whether or not a woundsucked. If the missile traversed the chest at such an angle that the externalwound of entry was at a considerable distance from the point of its entranceinto the pleural cavity, the movement of the intervening muscles of the chestwall might prevent the wound from sucking at one moment and permit it to suck atanother. If the wound involved the muscles of the pectoral girdle, movement ofthe upper extremity on the injured side might so realine the openings in theskin and pleura that sucking occurred constantly.
Casualties with chest injuries were usually transportedsupine or in the sitting or semisitting position, but not many were wounded inthese positions. Change of position after wounding therefore influenced theblowing characteristics of the wound.
Pathologic Physiology
Studies by Graham (1) (pp. 285-319 of reference cited)with Bell in World War I showed that a wound that exposes the lung and pleuralcavity is not necessarily fatal. It can be tolerated if the amount of airentering the wound from the exterior is not greater than the difference betweenthe tidal air and the original vital capacity. In the absence of treatment, theoutcome is determined by the size of the wound and the original vital capacity.
The lung on the injured side does not collapse to the point of nonfunctionwhen it is exposed to atmospheric pressure after wounding, even when the pleuralcavity is free of adhesions (2). Furthermore, the pressures exerted onthe exposed lung during the respiratory cycle are not always atmospheric.
6
7
8
The thorax is a flexible container, which moves during therespiratory cycle, and there may even be negative pressure in the exposed sideduring inspiration. The response to negative intrapleural pressure occasioned byenlargement of the chest on inspiration is a rush of air through the glottis. Ifthe size of the traumatic opening is less than the size of the glottis, asufficient amount of air enters through the glottis to answer the tidal airrequirements and for functional necessities, and there is no danger ofasphyxiation. When, however, the size of the wound approaches or exceeds thesize of the glottis, the amount of air which enters the chest through the openwound during each inspiration is large enough to interfere with the tidal airrequirement intake, and asphyxiation is a real danger.
Another factor must also be taken into consideration in the pathologicphysiology of chest wounds, interference with the function of the contralateralside. In the healthy person, the mediastinum is not held down by adhesions orstiffened by previous inflammatory disease. It is therefore capable oftransmitting pressure, and as a result, the intrathoracic pressure is likely tobe almost identical in both cavities. In other words, the decrease in vitalcapacity caused by an open wound of the chest is bilateral in the normal person,whereas a casualty with a stiffened mediastinum can sometimes tolerate even verylarge wounds.
When the amount of air entering an open chest wound duringeach inspiration is large enough to interfere with the tidal air requirementintake, some degree of asphyxia is inevitable. A vicious circle then ensues:
1. There is an initial and increasing reduction in the vital capacity.
2. The mediastinum shifts with each inspiration and expiration, since it isno longer opposed by the lung on the injured side.
3. The mediastinal shift (flutter) increases in rapidity as the reduction invital capacity increases.
4. The mediastinal shift has a direct and deleterious effecton the right heart, for anatomic reasons. The base of the heart is fixed, butthe apex is relatively free. The apex therefore is capable of movement and isrelatively unaffected by mediastinal flutter. The base is incapable of movement,and kinking of the large vessels that enter the right side of the heart is thenext consequence of this phenomenon.
5. Pendulum respiration ensues and causes a further decreasein the vital capacity. On each expiration, air from the contralateral lung, withits predominant carbon dioxide content, is only partly exhaled. The residualamount enters the main and secondary bronchi on the injured side. On the nextinspiration, the air taken into the uninjured lung is thus composed of bothatmospheric air from the trachea and air from the injured lung, with a highcarbon dioxide content.
9
Immediate Management
Casualties with large sucking wounds were usually in shock andnontransportable. All casualties with actual or potential sucking wounds werealways in jeopardy, whether or not the wound was blowing at the particularmoment of examination.
Surgical closure was necessary to bring about permanent restoration of normalcardiorespiratory physiology and to prevent intrapleural infection. This did notmean, however, that immediate surgical closure was necessary.
Early in the war, in obedience to the dictum that all sucking wounds shouldbe closed promptly, attempts were often made to close these wounds hastily bysuture as soon as the casualty reached the most forward medical installation.U.S. Army medical officers promptly learned, just as British medical officershad learned (vol. I), that this was a disastrous policy unless ample facilitieswere at hand for debridement and careful surgical closure. These facilities werenot available ahead of a field hospital. When closure was undertaken withoutcareful preliminary debridement, one or all of several complications mightoccur, including tension pneumothorax; extensive subcutaneous emphysema; orinfection of the wound, with disruption, which produced a recurrence of theoriginal sucking wound, with the difference that it was now surrounded bytissues that were infected and no longer suitable for approximation. Theincidence of empyema was high in wounds that broke down after repair. It reached50 percent in some series.
That treatment was a matter of extreme urgency was not open to debate. It wasalso quite simple. As already described (vol. I), medical aidmen were originallyinstructed to cover sucking wounds with an occlusive dressing (fig. 2C). Later,all chest wounds were treated in this way, on the ground that a wound that didnot suck at one time or in one position might suck in another. On thebattlefield, any type of dressing that was available could be used, even a pieceof clothing. As soon as the casualty was brought to the battalion aid station,the temporary dressing was replaced by a larger dressing, preferably at leasttwice as large as the wound. It was thickly impregnated with petrolatum and wascovered with a still larger gauze dressing held firmly in place with strips ofadhesive. If the wound was very large, coarse sutures were placed through theskin and tied over the dressings.
This sort of dressing closed the chest efficiently for 5 or 6hours, or longer. During this period, the patient was transported to a clearingstation in which it was determined, by triage, whether he should be transferredto the adjacent field hospital for emergency surgery or could safely betransported to an evacuation hospital farther to the rear, though it wasfrequently necessary first to change the dressing. When a petrolatum-impregnatedgauze dressing had become caked with blood, it had stiffened and was no longerpliable enough to act as a one-way valve. A needle with a flutter valveattachment was usually placed in the second interspace parasternally to providefor the escape of air and prevent tension pneumothorax (fig. 2D).
10
As the war progressed, emergency management of sucking wounds became sosatisfactory that their potential seriousness was considerably minimized.
Surgical Closure
Small wounds-Small sucking wounds, 2 cm. or less in diameter, althoughthey were likely to blow continuously if they were located in portions of thethoracic cage where muscle was scanty, could usually be closed withoutdifficulty. Lacerated skin and muscle were trimmed away, and accessible ribfragments were removed. Blood and clots were aspirated from the pleural cavitywhenever they were present. In the absence of other indications, however, thewound was not enlarged into a formal thoracotomy for exploration or for removalof small metallic foreign bodies.
The pleural defect was usually closed by a few sutures in the intercostalmuscle or, occasionally, by a small muscle plug in lieu of a pedicled graft. Themore superficial muscles and the subcutaneous tissues were closed in layers,with sutures of fine interrupted silk. The skin was left open, to be closed atthe base.
Intercostal water-seal drainage of the pleural cavity was frequentlyinstituted, even in small wounds, both because the lung had often been injuredand because air leakage was a possibility.
Large wounds-Sucking wounds associated with large defects in thethoracic wall and with multiple rib fractures were usually grossly contaminated.Thorough excision of devitalized tissue was therefore necessary, together withexcision of all loose rib fragments and of 3 to 6 cm. of the damaged costalends. Excision was thorough, but as much periosteum and intercostal muscle andfascia was preserved as seemed safe. When a debridement had been carried outaccording to these principles, a traumatic thoracotomy had often beenaccomplished, through which intrapleural exploration could be conducted. Foreignbodies were removed if they were readily accessible. The lung was sutured ifthere was leakage of air or oozing of any consequence from the parenchyma.Otherwise, no lung surgery was done (p. 17). The pleural cavity was thoroughlycleansed.
The adequate exposure necessary for the debridement of large sucking woundsoften made for difficulties in closure, particularly if the wound was in theanterior chest, where the muscle layer was scanty and the interspaces wider.Suture under tension had to be avoided. It was always undesirable, and it wasparticularly undesirable in contaminated areas. Even when the wound was locatedbasally, the diaphragm was never used to close the pleural defect if othertissues could be utilized.
Closure of the deepest layer of the wound could often be effected by usingthe intercostal structures, periosteum, and deep fascia. If large segments ofribs had been removed, these tissues were relaxed and could be manipulated morereadily. Relaxing incisions in the long axis of the intercostal bundles werehelpful.
11
When necessary, the first muscle layer could be reinforcedwith pedicle flaps or muscles (vol. I), or the inferior surfaces of the deepmuscles could be approximated to the first layer of the closure by interruptedsutures. When muscle tissue had been lost, subcutaneous dissection of the musclesometimes simplified closure, or relaxing incisions might be employed at somedistance from the defect. The subcutaneous tissues were usually closed, but skinclosure was deferred. The dressings were applied with a moderate amount ofpressure.
Tension pneumothorax was always a risk after a sucking wound had been closed.Some surgeons considered careful postoperative observation the only precautionnecessary. Others preferred to use a safety valve in the form of a catheter witha flutter valve or a large-bore needle in the second anterior interspace.
Since the empyema rate was greater in sucking woundsassociated with large defects of the thoracic wall, posterolateral closedintercostal drainage was sometimes employed. A second intercostal tube in one ofthe upper anterior intercostal spaces was useful in encouraging promptreexpansion of the upper lobe. If empyema occurred, it was then limited to thebase of the lung. If penicillin or a sulfonamide had been instilled into thechest and the posterior tube had been temporarily clamped off, the anterior tubecould be relied upon to facilitate prompt expansion of the lung.
FRACTURES OF THE RIBS AND ADJACENT STRUCTURES
Simple Fractures
Simple fractures of the ribs were not common incombat-incurred injuries. When they were encountered in combat zones, they wereusually the result of traffic or other civilian-type accidents. Relief of painwas secured by intercostal nerve block. Strapping was not practiced.
Solid, painless healing of simple rib fractures requiredabout 6 weeks. If several ribs were fractured, another 2 or 3 weeks had toelapse before the patient could resume his full activities.
Simple fractures represented no threat to life, but they accounted for aconsiderable loss of manpower because of the time required for totalconvalescence.
Costochondral separations were remarkably infrequent. They were managed, likesimple fractures, by intercostal nerve block. Recovery was usually prompt, buteven if pain at the costochondral joint was persistent, joint resection wasnever practiced.
Simple fractures of the clavicle were managed by standardmeasures. Wiring was unnecessarily radical; it enhanced the chances of infectionand could give rise to suppuration, which often was prolonged. Such heroicsurgery was occasionally encountered early in the war, but it was neverperformed by physicians aware of the essential function of the clavicle and ofits excellent natural capacity for recovery from trauma.
12
Simple fractures of the scapula were also infrequent in war wounds. Theyrequired only immobilization of the shoulder by a sling.
Acromioclavicular and sternoclavicular injuries as well asfractures of the glenoid fossa required special orthopedic care. Since they werenot managed primarily by thoracic surgeons, they are discussed in the volumes onorthopedic surgery in this series.
Compound Fractures of the Ribs
About 75 percent of all combat-incurred penetrating and perforating wounds ofthe chest were associated with compound fractures of one or, more often, ofseveral ribs. The scapula and clavicle were often similarly injured, andcomminution of the bony structures might be extreme.
Pathologic process-Occasionally, a tangential injury to the chestresulted in the fracture of one or more ribs in such a manner that the sharpedges of the fractured ribs penetrated the pleura even though the missile itselfdid not. Bone fragments from fractured ribs, because of their irregular, jaggededges and the respiratory movements of the chest, sometimes caused long, raggedpulmonary lacerations and other damage more serious than that caused by theoriginal missiles. Even if this did not happen, penetration of the intact pleuraby bone fragments could cause a pneumothorax, and continued trauma to the lungsby embedded bone fragments resulted, in a few cases, in both empyema and lungabscess. Both infection and sequestration were, however, remarkably infrequent,because of the almost universal practice of adequate debridement in forwardhospitals.
Flail chest-Flail chest, with paradoxical respiration, resulted frommultiple fractures of several ribs. It was not often seen, but it was extremelyserious because of the consequent reduction in vital capacity (fig. 3). The morethe chest wall was sucked in toward the contralateral side during inspirationand the less it returned to the normal position on expiration, the greater wasthe decrease in vital capacity and the more the patient compensated by breathingfaster.
Immediate management-The immediate treatment offractured ribs (fig. 4) was the relief of pain by intercostal nerve block, whichusually lasted for 24 hours or more. When the casualty was able to breathedeeply, cough, and raise sputum, he became transportable, if associated woundsdid not require forward surgery, and definitive treatment could be deferreduntil he reached an evacuation hospital.
Adhesive strapping was practically never used. It was often unsuccessful inaccomplishing immobilization and relieving pain, and the adhesive often causedadditional discomfort. Strapping was also unphysiologic. Many patients withfractures of the ribs showed varying degrees of wet lung with increasedbronchial secretions and intrapulmonary bleeding. Raising the fluid would havebeen difficult or impossible because of pulmonary compression and therestriction of expansion of the chest caused by strapping.
13
14
15
16
In civilian practice, a patient with flail chest would beideally treated in a respirator. This was, of course, completely impractical inmilitary surgery. Nerve block often resulted in a considerable improvement insymptoms and a considerable reduction in the paradoxical movement of the chestwall. If it was not successful, adhesive strapping, undesirable as it was,sometimes had to be employed, or an elastic binder was used. Positioning withsandbags was occasionally useful, but traction was seldom employed. If simplermeasures were not successful, operation had to be resorted to. Costal fixationwas secured by placing stainless steel wire or heavy silk through the ends ofthe shattered ribs. Tracheotomy was sometimes necessary also.
Surgical management-The surgical management of fractures of the ribsrequired careful individualization of patients. In most instances, debridementcould be limited to the soft tissues. If the wound was tangential, there mightbe extensive comminution of from three to five ribs without penetration of thepleura. Often the greater portion of the damage was subscapular. In a few cases,roentgenograms showed what looked like a picket fence of bony spicules along theparietal pleura.
Whenever debridement included the removal of irregular,subpleural rib fragments, the pleural cavity was necessarily entered, and thesurgeon had, in effect, performed a thoracotomy. He had to have sound reasonsfor extending the wound and increasing the magnitude of the operation. The merepresence of bone fragments within the pulmonary parenchyma was not an indicationunless they seemed responsible for continued bleeding or for an air leak.
One exception to this generalization was the presence of fragments partlywithin the pulmonary parenchyma and partly within the pleural space. Fragmentsin this location could produce a good deal of trauma when the lung reexpandedand they came into contact with the chest wall. Both air leaks and infectionswere possible sequelae. A number of cases were observed in which the lung, thepleural cavity, and the chest wall were all involved in the infection thatfollowed debridement carried down only to the muscles, the bone fragments beingleft in situ.
WOUNDS OF THE LUNGS
General Considerations
Lacerations of the pulmonary parenchyma were encountered inpractically all perforating chest wounds. They varied from small puncture woundsto very extensive lacerations.
When surgeons, early in the war, first encountered badly contused, lacerated,boggy, hemorrhagic lungs, they found it hard to refrain from immediate lobectomy.In fact, this operation would probably have been done rather often if thecasualties had not been in such poor general condition or had not had associatedmajor wounds. For these reasons, pulmonary surgery
17
was, of necessity, limited to simple suture. The sutures, also of necessity,were superficial, since the friable, hemorrhagic tissues were incapable ofholding deep sutures. The practice was to perform one-layer closure, withinterrupted sutures of fine, nonabsorbable material on an atraumatic needle. Itproved surprisingly easy to secure an airtight closure in this manner, onereason being that the tendency to leakage was diminished by the fact that thedamaged lung was somewhat slow to aerate completely. Roentgenograms examined 2or 3 weeks after surgery showed remarkable clearing of the lung fields, andclearing was usually complete in 6 to 8 weeks.
One reason for these good results can be explained in theform of an analogy: Engineers found that the best protection for wiring inairplanes was simply to fasten the wires loosely along the side of the plane.Missiles that passed through wires thus placed seldom damaged all of them,whereas if the wiring had been placed in a conduit, a single bullet that struckthe container could divide all of the wires in it. The same phenomenon wasobserved in the lung. The blood vessels are surrounded by an elastic medium thatpermits them to be easily displaced in any direction when a foreign bodystrikes them. As a result of this physiologic capability, damage to the lungswas frequently more limited than it might otherwise have been.
The real explanation of the tremendous recuperative power ofthe lung is its dual blood supply, through the pulmonary and bronchial arterialsystems. The blood supply, being in the form of a tree, was irreparably damagedonly when the trunk was severed. Since all the major blood vessels branch outradially from the hilus, they had to be damaged near the point of origin tocause lethal damage to the pulmonary parenchyma.1Undoubtedly there were many such injuries in World War II, but the casualtieswho sustained them did not live long enough to reach field hospitals.
Management in Forward Hospitals
As the war progressed, the policy of leaving smalllacerations of the lung unsutured became general. No untoward effects werenoted. Untreated lungs were found to expand as well, and about as rapidly, aslungs which had been sutured. These observations bore out the theory that thesurrounding airless lung, when it was infiltrated by blood, preventedpostoperative parenchymal pleural fistulas.
Eventually it became the custom to suture lacerations of lung tissue only onthe following indications:
1. Obvious parenchymal pleural fistulas.
2. Large lacerations, chiefly to reduce the amount of postoperative drainage.
1Wartime observations are borne out by a series of experiments on the pathologic effects of foreign bodies in the pulmonary artery reported in 1959 (3). These studies showed that the lung can recuperate when a branch of the artery or the lobar pulmonary artery is completely blocked. Only when the main artery is blocked will the pathologic changes be irreversible. Severe pathologic changes result, however, when the bronchial artery and the pulmonary artery to the involved lobe are blocked.
18
3. Lacerations bleeding into the pleural cavity.
4. Lacerations bleeding into the tracheobronchial tree.
It was logical to repair pulmonary lacerations that were oozing actively orthat presented serious air leaks. It was particularly important to suturelacerations that were bleeding into the tracheobronchial tree. In thesecircumstances, even with repeated endotracheal suction, it would have beendifficult to keep the airway free of blood.
If the chest was to be drained, closure of a small air leak was notconsidered important. Air leaks were identified by filling the hemithorax withphysiologic salt solution under pressure of 10 to 15 mm. H2O;it was necessary that the lung be completely covered with the solution.
There was no indication for pulmonary repair in the contusiontype of injury, since it caused either scattered areas of hemorrhage in bothlungs, as in blast injuries, or massive bleeding that involved all the lobes.This therapeutic point of view is in interesting contrast to the World War Iconcept that all hemorrhagic infiltrations ("splenizations") had to beresected if a fatal issue were to be prevented (4).
Occasional small peripheral wedge resections were performed for localizedareas of necrosis around foreign bodies. Otherwise, there were few validindications for resection in forward hospitals. They were limited to:
1. Irreparable damage to a major bronchus.
2. Partial traumatic amputation of a portion of a lobe in which the bronchihad been transected and little or no blood supply remained.
3. Damage to major pulmonary vessels of such a character as to requirecomplete ligation. Very few patients in this category lived long enough to reacha forward hospital.
The records of the 2d Auxiliary Surgical Group, covering2,267 thoracic and thoracoabdominal wounds, show only a single lobectomy in aforward hospital, on a patient who died in the course of the operation.Pneumonectomy was not attempted in any instance.
Management in Base Hospitals
Even in base hospitals, indications for pulmonary resection seldom arose. Anoccasional peripheral wedge resection was performed, as in forward hospitals, toremove a localized area of necrosis around a foreign body or on the indicationof a small peripheral posttraumatic abscess or multiple bronchopleural fistulas.The tissue to be excised was held between clamps. After hemostasis had beensecured, the raw edges of the lung were sutured together, and the edges of thepleura were inverted with two rows of interrupted sutures of fine silk. Airleakage did not occur with this type of suture.
The experience of the 2d Auxiliary Surgical Group shows how infrequent wasthe need in World War II for the radical excision of lung tissue. Maj. Thomas H.Burford, MC, and Maj. (later Lt. Col.) Paul C. Samson, MC, found it necessary toperform only one lobectomy and only two total pneumonectomies
19
in the base hospitals in which they worked. In retrospect, they questionedthe wisdom of all three operations:
The lobectomy was performed for a massive liquefied hematoma of the upperlobe, which had shown no evidence of regression over a 4-week period. Thepatient died on the operating table. In this case, simple external drainagewould probably have been wiser.
The first pneumonectomy was performed on the indication of a chronicfistulous tract resulting from a penetrating shell-fragment wound. The tract wasmore than 1 cm. in diameter and extended through both upper and lower lobes atthe hilus. Surgical obliteration was impractical. The patient survived thepneumonectomy but eventually died from empyema and purulent pericarditis.
The second pneumonectomy, in which death occurred from the same causes, wasperformed on the indication of multiple recurrent fistulas, a collapsed lung,and an infected hemothorax. The original injury was a perforating wound from arifle bullet. Pathologic examination of the excised specimen showed multipleareas in which alveoli were ruptured, and widespread vascular damage wasindicated by thrombosis of numerous small vessels. In this case, it wouldprobably have been wiser merely to drain the empyema and accept the collapse ofthe injured lung as permanent.
The results in these cases show that the mortality for pulmonary resectionperformed for thoracic injuries or their sequelae is so high that operations ofthis sort are seldom if ever indicated. They should be undertaken only onclear-cut indications and only after the most careful consideration.
In the Mediterranean theater, an increasingly conservativeattitude toward lobectomy and pneumonectomy was gradually crystallized asexperience in the management of combat-incurred casualties increased. In thelast year of the war, in treating a series of 338 chest injuries at the 21stGeneral Hospital, Maj. Lyman A. Brewer III, MC, found no indication forlobectomy in any of the patients. In one case, it was thought that lobectomymight be necessary later, but the decision was deferred until the lung had had achance for maximum recovery and the patient could be studied more deliberatelyin the Zone of Interior.
WOUNDS OF THE MEDIASTINAL STRUCTURES
General Considerations
Wounds of the mediastinal structures were uncommon in forwardhospitals, chiefly because injuries to adjacent vital structures were usuallylethal. Mediastinal wounds included injuries of the larger hilar blood vessels,which were usually promptly fatal; injuries of the intrathoracic trachea andmajor bronchi; and injuries of the esophagus. Diagnosis, which was difficult andconfusing, was based upon the following considerations:
20
1. The projected course of the missile was the most useful index of possibledamage.
2. Signs of continuing intrapleural hemorrhage were frequently, but notnecessarily, present when a large blood vessel had been injured.
3. Injuries to the trachea and major bronchi caused a rapidaccumulation of air in the pleural cavity, frequently under increased pressure.Whenever leakage of air continued, it could be assumed that a bronchus or branchbronchus had been injured.
4. Mediastinal emphysema of varying degrees was frequentlypresent, as the result of wounds of the hilar bronchi, trachea, or subglotticlarynx, though in the experience of the 2d Auxiliary Surgical Group, the air wasnever under sufficient pressure to obstruct the venous return to the heart andgive rise to symptoms. It was the opinion of these surgeons that most of thesymptoms ascribed in the literature to mediastinal emphysema were more probablycaused by an unrecognized associated pressure pneumothorax. The ease with which,in many cases, air in the mediastinum dissected into the pleural cavity wasexplained by the absence, in most World War II soldiers, of old pleural diseaseand subsequent pleural thickening and adhesions. The most frequent clinicalevidence of mediastinal emphysema was a precordial crunch or click synchronouswith the heartbeat.
Management
Correction of the pressure pneumothorax usually stabilizedthe patient sufficiently to permit whatever surgery was necessary. Onlyoccasionally were suprasternal incisions into the deep fascial planes necessaryas an emergency procedure to relieve emphysema.
Exploration in injuries of the mediastinum was usually carried out for one oftwo reasons, suspected injury to the esophagus or continued bleeding from largevessels or from the heart. The mere demonstration of shell fragments within themediastinum was not considered a sufficient indication for surgery in theabsence of one or the other of these indications.
Enlargement of the mediastinum by extravasation of blood, asdemonstrated by roentgenograms, was not an indication for exploration unlessthere were also signs of obstruction of the superior vena cava, as shown byprogressive swelling of the cervical veins or unless progressive enlargement ofthe mediastinal shadow suggested continued hemorrhage. Considerable enlargementof the superior mediastinum sometimes accompanied wounds of the neck as theresult of dissection into it of blood from a cervical vessel. It was the rule toinvestigate wounds in the neck thoroughly, before proceeding with such extensiveoperations as removal of a part of the clavicle and the sternum, in the hopethat, even if the missile was in the superior mediastinum, the source of thehemorrhage might be reached from above or might become accessible by exposure atthe base of the neck. When a missile in this region could not be readilyremoved, the area could be drained through the cervical wound.
21
As a matter of convenience, wounds of the trachea and bronchi and of theesophagus are discussed separately.
WOUNDS OF THE TRACHEA AND BRONCHI
General Considerations
Injuries of the intrathoracic trachea and main bronchi wereextremely uncommon in forward hospitals, chiefly because, if they were ofconsequence, death promptly occurred on the battlefield. The few patients whosurvived to reach the hospital-there were only 4 in the 2,267 thoracic andthoracoabdominal injuries encountered by the thoracic surgical teams of the 2dAuxiliary Surgical Group-were in severe cardiorespiratory imbalance. None ofthem presented a sizable intrathoracic laceration of the trachea.
Unless casualties with this kind of injury had also sustained a traumaticthoracotomy, wounds of major bronchi were characterized by a rapid, completecollapse of the affected lung, associated with an extreme degree of pressurepneumothorax (fig. 5). Subcutaneous emphysema might be extreme, extend-
22
ing from the scalp down to, and into, the testicles. When thetrachea was the site of a major wound, both lungs were collapsed and the patientusually died very quickly.
These findings, combined with the location of the missile,pointed to the diagnosis, particularly in injuries of the trachea. The onlypossible source of confusion was a sucking wound at the apex of the lunganteriorly, which might be mistaken for a perforation of the trachea.
Management
In wounds of the trachea and bronchi, immediate, adequatedecompression was imperative. It was best accomplished by the introduction intothe second interspace anteriorly of a large catheter, which was connected with awater trap. Large amounts of blood were frequently present in the pleura, andunless thoracentesis was also carried out promptly, the patient could drown fromthe sudden entrance of this blood into the tracheobronchial tree. The diagnosiswas established if the lung did not reexpand completely when these procedureswere instituted; the bubbles of air expelled during quiet expiration could beobserved in the water-seal drainage.
If bleeding was not severe, it was best to reestablish the cardiorespiratorybalance before operation was undertaken. If bleeding was continuous, operationhad to be performed promptly. Even if the diagnosis was only suspected,thoracotomy was warranted, since both the bronchi and the trachea lendthemselves quite well to suture.
A high posterolateral approach, preferably on the right side, provided thebest exposure for wounds of the trachea. Repair was accomplished by simpleclosure with interrupted sutures of fine silk. If there was loss of substance inthe tracheal wall, mediastinal tissue, pleural tissue, or a long-pedicled flapfrom the latissimus dorsi could be used to aid in the repair. Occasionally, inhigh tracheal wounds, repair was accomplished through an anterior approach, withdivision of the clavicle.
The diagnosis, or even the suspicion, of injury to a majorbronchus was an indication for exploration. The mere existence of damage to abronchus was not, however, an indication for lobectomy, though if the vein orartery to a lobe of a lung was lacerated, resection might conceivably benecessary. Not many such injuries were seen because they led to rapidexsanguination, and the casualties did not usually survive long enough to reacheven forward hospitals. In contrast to the general experience, Maj. Robert H.Wylie, MC, had two recoveries after operation for injuries of major bronchi.
In one case in which serious bleeding from an upper lobe bronchus madeanesthesia hazardous, Maj. (later Lt. Col.) Lawrence M. Shefts, MC, packed themain stem bronchus under endoscopic vision. It was then possible to occlude theblood supply temporarily and proceed with the repair of the damaged bronchus.The procedure in this case was lifesaving.
23
Case History
Col. Edward D. Churchill, MC, Consultant in Surgery to theSurgeon, Mediterranean Theater of Operations, U.S. Army, early predicted that nopatient with an injury to a left main stem bronchus would survive to be treatedin a forward hospital, as the anatomy of the left hilar region made itinevitable that a missile could not pass through it without inflicting fataldamage. No such injuries were observed. Lt. Col. W. Paul Sanger, MC, observedtwo casualties with damage to the right main steam bronchus. Both survived to behospitalized because of the unusual tactical situation: Both were wounded on theAnzio beachhead, in close proximity to the hospitals to which they wereadmitted. One of them died early in the postoperative period because ofoverwhelming pleuropulmonary sepsis. The history of the other follows.
Case 2-This casualty washospitalized within an hour after he had sustained a shell-fragment wound of theeighth posterior interspace on the right side, presumably while prone. Heimmediately became dyspneic and was unable to move the lower extremities. Whenhe was admitted, he was in shock and was completely paralyzed below the eighthdorsal segment. Respiration was profoundly embarrassed, obviously because of alarge wound, 5-6 cm. diameter, in the eighth right interspace posteriorly, whichwas sucking air and draining blood. It was filled with bits of clothing and bonefragments. The sucking was controlled at once with petrolatum-impregnated gauzepacks. Paravertebral block of the sixth to the tenth intercostal nerves greatlyimproved the respirations and permitted the patient to cough up large amounts ofblood. The blood pressure was elevated to 86/52 mm. Hg by the infusion of 25 gm.of serum albumin. Aspiration of the chest produced 800 cc. of blood, which wasimmediately returned to the patient by autotransfusion. Insertion of anintercostal catheter relieved the developing pressure pneumothorax.
By 9 hours after injury, the patient hadbecome fairly well stabilized. Then there was a sudden increase in thepneumothorax. Emergency surgery was performed, under endotracheal anesthesia.The wound in the region of the eighth right rib was excised, and aposterolateral thoracotomy incision was made in the eighth right interspace. Alarge, ragged, bullet-shaped shell fragment, 12 cm. long, was found penetratingthe posterolateral wall of the right main stem bronchus, just distal to thebifurcation of the trachea. Air and blood were pouring freely from the wound.The azygos vein was lacerated; it was sutured, without complete occlusion. Then,with considerable difficulty, the defect in the bronchus was approximated withinterrupted silk sutures; the closure was reinforced with muscle. When theclosure was water-tested, there was no evidence of leakage.
A transverse laceration of the upper lobe of the right lung,between 12 and 14 cm. long, was closed with interrupted silk sutures. The lungwas then reexpanded except for the portion of the upper lobe in the area oflaceration. The incision was closed in anatomic layers.
The patient received 1,000 cc. of blood and 250 cc. of plasmaduring the operation and was in fair condition at the end.
The immediate postoperative course was stormy, not because ofthe chest wound but because of the distention and other complications thataccompany cord transection. Suprapubic cystostomy was done on the seventhpostoperative day, and a hip spica was applied, to facilitate evacuation.
The right lung remained expanded, but on the 12th day, theincision became infected in the area of the original wound. Empyema followed;such an experience shows the unwisdom of including the wounded area in thethoracotomy incision. Treatment consisted of the intramuscular injection of25,000 units of penicillin every 3 hours for 10 days; no sulfonamides weregiven.
24
Reports from the base hospital a month after wounding statedthat the empyema cavity was still from 200 to 300 cc. in volume. There was somereturn of sensation in the extremities. The last report, 4 months later, statedthat progress was continuing to be good.
WOUNDS OF THE ESOPHAGUS
Incidence
Wounds of the thoracic esophagus were uncommon. In the 1,364 thoracic woundsand 903 thoracoabdominal wounds cared for by the surgeons of the 2d AuxiliarySurgical Group, there were records of only 6 cases. In one of these, thediagnosis was doubtful; the surgeon who removed the missile from the wall of theesophagus noted that he did not think that the lumen had been penetrated. Thetwo other cases in which surgery was done ended fatally, and the three remainingcases were diagnosed only at autopsy. This surgical group, therefore, with itswide experience of thoracic injuries, had no record of certain recovery from aconfirmed penetrating wound of the intrathoracic esophagus.
Diagnosis
Diagnosis of wounds of the esophagus (figs. 6 and 7) wasdifficult, but it was imperative that it be made because continued spillage inan unrecognized wound was likely to be followed by mediastinal infection (fig.8), which was
25
often fatal, or massive fulminating empyema. The chiefdiagnostic difficulty was that neither symptoms nor signs were consistent andthat suspicion was frequently not aroused until serious complications hadensued. Another difficulty was that many of the symptoms and signs present wererelated to wounds of associated structures, particularly large sucking woundsand massive hemothoraces, and tended to overshadow whatever clinical picturemight be produced by lacerations of the esophagus.
Perhaps the most common symptom of esophageal injury wascontinued substernal discomfort, with acute substernal pain on swallowing. Truedysphagia was not the rule. Severe pain in the back, radiating down into thelumbar region, was pathognomonic when it was present. Typically, it was notalleviated by large doses of morphine. The probable origin of the pain was theinflammatory reaction in the posterior mediastinum caused by leakage from theesophagus.
Mediastinal emphysema was often present and was sometimesassociated with cervical emphysema. Equally often, it was absent, for theopening in the mediastinal pleura that usually was a part of such injuriespermitted the escape of air into the pleura (fig. 6). The presence of apneumothorax was not always helpful, for the missile which had penetrated theesophagus might very well have penetrated or perforated the lung also.
26
Projection of the track of the missile was not helpful in many cases becausea hemothorax or hemopneumothorax might obscure the path which the missile hadtaken in traversing the lung. If, however, one kept in mind the exact positionof the esophagus in various parts of the thorax in relation to the location ofthe missile, as demonstrated by anteroposterior and lateral roentgenograms, itwas often possible to make diagnostic assumptions. This help was not availablewhen a missile of moderate size had passed completely through the mediastinum.
If an injury of the stomach could be ruled out, blood obtained on the passageof a Levin tube was an indication of possible esophageal injury. Ifroentgenograms showed a widening of the mediastinal shadow, a swallow or two ofthin barium or Lipiodol was given and additional roentgenograms taken.Mediastinal widening was, however, a confusing finding, for it might also becaused by injuries at the base of the neck. The aspiration from the pleuralcavity of gastric juice or of fluid containing food particles was pathognomonicof a gastrointestinal perforation. Food particles were identified by the nakedeye. The presence of gastric juice was readily determined by the very acid
27
reaction of litmus paper to it. If the diaphragm was not injured, thesefindings were evidence of a perforation of the esophagus.
As a practical matter, the diagnosis of injury to thethoracic esophagus was chiefly made on suspicion, aroused by whichever of thesigns, symptoms, and roentgenologic findings just listed might be present in theparticular case. It almost always had to be confirmed by exploration, which waswarranted even if the suspicion was based on nothing more concrete thanknowledge of the course of the missile in a perforating wound or of its positionin a penetrating wound. Any rent in the posterior mediastinum or any hematoma inthis region found in the course of exploration of the chest on other indicationswas always investigated, with the possibility of injury to the esophagus inmind.
Management in Forward Hospitals
The extremely high mortality of wounds of the esophagus, as just noted, madeit mandatory to perform an exploratory operation on the slightest suspicion thatsuch an injury existed. The limited World War II experience, however, makes itimpossible to make definitive statements about management.
If the patient's condition did not permit immediate operation, the best planwas to intubate the stomach, withhold all food by mouth, and keep the patientunder penicillin protection.
When surgery was indicated, the esophagus, if the wound was in the upperthird, was best exposed from the right side, by a posterolateral transpleuralapproach. If both wound and missile were on the left side, a left-sided approachwas indicated.
The margins of the esophageal wound were excised back to healthy tissue. Thena two-layer closure was carried out, with the second layer of sutures in themusculature. Interrupted sutures of fine silk were used. The mediastinal pleurawas left open. The pleural space was drained with two intercostal water-sealdrains.
The single patient observed by the 2d Auxiliary Surgical Group on whom thistotal technique was used did not survive, because of other wounds, but for the 5days he lived after operation, there was no clinical evidence of leakage fromthe esophagus or of pleural involvement. Post mortem examination, unfortunately,was not carried out.
Most surgeons of experience considered extrapleural posterior mediastinotomyboth useless and harmful because it jeopardized the first objective of theoperation, complete and permanent reestablishment of esophageal continuity. Oncethis was accomplished, drainage, in their opinion, would add little to thesafety of the patient and might encourage breakdown of the wound. Posteriormediastinotomy was, in fact, an operation of limited usefulness incombat-incurred chest wounds.
Postoperative management-Gastrostomy for nutritional purposes was notconsidered necessary in a forward hospital. If decompression was required,
28
a Levin tube was used, either passed into the stomach, or, as advocated bySweet (5) in resections of the esophagus, passed down to the site ofrepair.
Management in Base Hospitals
While few patients with undiagnosed wounds of the esophagus were encounteredin base hospitals, it was necessary to be on the alert for such injuries. Rapiddevelopment of a fulminating empyema and aspiration of a thin pleural exudatewith a musty odor characteristic of gastric juice were almost pathognomonic ofan esophagopleural fistula. If methylene blue taken by mouth appeared in thepleural space, the diagnosis was confirmed. The site of the injury was localizedby fluoroscopy after the patient had been given a swallow of Lipiodol or thinbarium solution. If a tracheoesophageal fistula was suspected, diagnosis wasmade by bronchoscopy.
Once the diagnosis was made, immediate rib resection was carried out, and alarge rubber tube was inserted into the empyema cavity. For the first few days,the tube was connected to a water-seal bottle. Gastrostomy was performed at thesame operation even if the wound in the esophagus was very small. In this sortof case, intubation was often tried, with nothing by mouth. Full doses ofpenicillin were given intramuscularly.
The patient was nursed in the sitting position, to reduce the tendency toregurgitation, which was present in practically every case. Gastric contractionto a degree incompatible with nutritional needs was common, and the associatedpylorospasm, which was probably the result of vagal irritation resulting, inturn, from mediastinal involvement, made continuous regurgitation almostinevitable. Unless it could be controlled, healing of the esophagus was delayed.If regurgitation was troublesome after an otherwise satisfactory gastrostomy,jejunostomy was resorted to. In addition to preventing regurgitation, thejejunostomy served admirably as a feeding route. The gastrostomy stoma thenserved as a decompressive vent.
The criterion of satisfactory progress was closure of the esophageal fistula,reexpansion of the lung, and obliteration of the empyema cavity. If theseresults were not accomplished by this regimen, a direct surgical attack wasmade, consisting of thoracotomy, decortication, and closure of the esophagealopening.
Case Histories
The following case histories illustrate the difficulties and complications ofwounds of the esophagus:
Case 3-A paratrooper who was wounded in action on 15 September 1944sustained a perforating bullet wound of the right chest. The wound of exit wasto the left of the midline, just to the left of the xiphoid process. Debridementwas carried out after resuscitation. There were no symptoms or signs to suggesta wound of the esophagus.
When the patient was received in a base hospital 5 days after wounding, hewas critically ill. Investigation revealed a large total empyema on the rightside, with a
29
large esophagopleural fistula (fig. 9A). Forty-eight hours later, the empyemawas drained under local analgesia by rib resection, and gastrostomy wasperformed.
The patient improved, but leakage of regurgitated material through thefistula remained profuse, and the lung showed no tendency toward reexpansion.Four weeks after wounding, thoracotomy was performed, with decortication andclosure of the 2-cm. longitudinal defect of the esophagus just above the cardia.The esophagus had been pulled well to the right by the changes consequent toinjury and to subsequent infection. Operation was performed without difficulty,and pulmonary expansion was immediate and complete. Closure was effected in twolayers, by interrupted sutures of fine silk, with inversion of the knots. Fibrinfoam was sutured over the site of the repair. The mediastinal pleura was closed,and the usual water-seal intercostal catheter drainage was instituted.
The immediate postoperative course was uneventful, but regurgitation wastroublesome, and recurrence of a small basal empyema required secondary drainageby rib resection. Methylene blue given by mouth appeared in the pleural exudateat the end of 2 hours, and fluoroscopic examination after a swallow of Lipiodolshowed a very small leak at the site of repair. After jejunostomy was performed,there was prompt relief of regurgitation as well as rapid obliteration of thebasal empyema cavity (fig. 9B). When the patient left the hospital, hiscondition was excellent in all respects.
Case 4-This patient received his initial surgery in a British mobilefield hospital. A tommygun bullet had penetrated the neck at the base on theright side and traversed it obliquely, emerging at the posterior margin of theleft sternocleidomastoid muscle. The right thyroid lobe was damaged, the rightlateral wall of the trachea opened, and the esophagus almost completelytransected.
At operation, a few hours after injury, the trachea was repaired and theesophagus was closed by end-to-end suture, presumably with catgut. The wound wasclosed with drainage, and gastrostomy was done.
At the end of 5 days, the neck began to swell, and on theseventh day, the anterior wound broke down and began to discharge both purulentexudate and esophageal secretions. When the patient reached the 21st GeneralHospital, he had a large esophageal fistula, with
30
considerable infection. Vomitus was aspirated into the lungwhen he vomited soon after admission, and a typical aspiration pneumonia furthercomplicated the clinical picture. After a stormy period, the pneumonia cleared,and the cervical infection subsided. The infection had not extended to theposterior mediastinum. When the patient was last observed, he still had a largefistula of the esophagus, and his condition still did not warrant barium studiesto determine what type of reparative procedure would be required.
Case 5-This patient was struck in the right shoulder and received apenetrating wound of the neck from an explosion of an antitank shell within anM-4 tank. Fluoroscopic examination revealed multiple foreign bodies in the baseof the neck, one of them in the region of the right lateral wall of theesophagus. Extreme dysphagia was noted soon after the injury. The wound ofentrance was debrided. No attempt was made to remove any foreign bodies.
The diet was limited to fluids, but intubation was not done.A week after injury, there was leakage from the wound of material taken bymouth, and when the patient was admitted to the 21st General Hospital, he had anesophageal fistula.
Intubation was immediately instituted, with high-caloric, high-vitaminfeedings by this route. Nothing was permitted by mouth. The infection clearedrapidly, and the fistula closed with equal rapidity. Two weeks after admission,the patient's barium studies showed a normal esophagus.
Comment-This case reemphasizes the importance ofrecognizing dysphagia as a possible symptom of a wound of the esophagus and ofresorting to intubation at once in any instance of possible perforation.
CRUSHING INJURIES
Compression or crushing injuries of the chest observed in World War II werechiefly of noncombat origin. They varied considerably in extent, but they wereusually relatively mild and required only conservative management.
The pathologic lesion varied according to the degree of trauma. It includedsimple contusion injuries of the pleura and lung; intrathoracic hemorrhage,subpleural in slight injuries and more central and more extensive in severeinjuries; hemothorax; pneumothorax; fractures of the ribs and the sternum;rupture of viscera; and massive collapse of the lung. The so-called contusionpneumonia which followed severer types of compression injury and which could beextremely serious was in reality not a specific entity but the pneumonitisresulting from factors common in thoracic trauma. These factors included thelimited respiration and cough caused by chest pain, retained bronchialsecretions, and hemorrhage and edema of the lung. Together they made up theclinical picture of wet lung (p. 207).
In simple crushing injuries, the only treatment necessary was rest and reliefof pain. In more severe injuries, the management of shock and the correction ofdisturbances of cardiorespiratory physiology resulting from hemothorax,pneumothorax, and emphysema were of primary importance. Surgery was necessarywhen lacerations of the diaphragm were present (p. 103). Penicillin was given,to prevent infection. Late complications included pneumonia, empyema, and lungabscess.
31
Ecchymotic Mask
Ecchymotic mask is the term used for the bluish-violetdiscoloration of the face, neck, and upper portion of the chest that sometimesresults from sudden, direct compression of the chest or abdomen. In civilianpractice, it is most often seen following compression of the chest between therear end of a truck and a loading platform. It may also occur in sudden, directcompression of the extremities as the result of a fall or some other accident inwhich the thigh is violently flexed on the abdomen. In wartime, the casualty maybe actually crushed by falling debris from bombs.
The discoloration may also be found in the pharyngeal and retrobulbar areas,though in these regions, true hemorrhage may have occurred. It is ordinarilymore marked at areas of pressure, such as in the neck from collarbands, aroundthe head from hatbands, and across the shoulders from suspenders.
The explanation of ecchymotic mask is extreme stasis in the peripheral venouscapillaries, which remain intact but become dilated. There is no evidence ofdiapedesis. The proof of the theory of stasis is threefold, (1) the intact stateof the vessels; (2) the absence of extravasation of blood on microscopicexamination; and (3) the absence of the usual shades of color observed as anordinary ecchymotic spot gradually fades.
Clinically, the discoloration begins to fade about 48 hours after the injuryand usually disappears in 2 weeks. The patients are seldom uncomfortable, andsequelae are unusual. Ocular symptoms, which are sometimes present, range frommildly blurred vision to total blindness. Fortunately, ocular involvement isuncommon, for retrobulbar or fundal hemorrhages may produce permanent atrophy ofthe optic nerve.
When Heuer (6), as the result of a personal experience in World War I,reported on ecchymotic mask, he was able to find 127 cases recorded in theliterature up to 1923. There were 27 deaths immediately after injury and 8deferred deaths. Autopsies were performed in 23 of the fatal cases. Symptoms andsigns of intrathoracic injury were common, but they were generally slight anddisappeared promptly. They included hemoptysis, pulmonary edema, hemothorax,pleural effusion with empyema, pneumonia, and subcutaneous emphysema.
Visual disturbances were reported in only 16 of the recorded cases, includingHeuer's own case, but in 7 of these, there was permanent impairment or completeloss of vision, in all instances associated with progressive optic atrophy. Inmost of the recorded cases, there was no statement about the eyes, but in 23reports, it vas stated specifically that the eyes were not affected, and in 10other cases, it can be assumed that they were not affected.
Ecchymotic mask was not often observed in World War II. It did not occur inany of the forward cases treated by the 2d Auxiliary Surgical Group,
32
which suggests that the causative factor is a product oflocalized thoracic crushing rather than the result of a penetrating or a blastinjury. Major Burford, working in base hospitals, observed it only five times in1,200 casualties. All of these were service troops, all were injured in jeep ortruck accidents, and all recovered uneventfully.
Another textbook condition, commotio thoracis, was neverobserved by surgeons of the 2d Auxiliary Surgical Group, whose opinion was thatit should be dropped from the list of traumatic entities.
BLAST INJURIES
Incidence
The British experience with blast injury was entirely gainedthrough German bombings of the British Isles, a fact which, for a time, misledmedical officers in the U.S. Army, who believed that they would encounter thesyndrome frequently in combat. They did not. Blunt injuries of the intact chest,with resulting contusion, were fairly frequent, but blast injuries wereuncommon. Early in the U.S. experience, the two conditions were frequentlyconfused. Once the differentiation was appreciated and pulmonary edema from wetlung, injudicious fluid administration, and occasional fat embolism were alsoexcluded, very few patients were seen in field or evacuation hospitals who couldbe considered to be suffering from true blast injuries of the chest.
The single case of blast injury observed by surgeons of the 2d AuxiliarySurgical Group in their 2,267 thoracic and thoracoabdominal injuries wasconfirmed by post mortem examination. The 8 cases observed by Maj. Frank Tropea,Jr., MC, and Lt. Col. John M. Snyder, MC, in 603 chest injuries resulted in 3deaths; in 1 case, the diagnosis was confirmed by autopsy.
Capt. William W. Tribby, MC (7), in his study of 1,000battlefield deaths of U.S. Army troops in Italy, found 13 bodies in which therewere no penetrating injuries and in which the cause of death was presumablyblast injury. In four cases, the diagnosis was confirmed at autopsy, whichrevealed diffuse pulmonary hemorrhage in all cases and pulmonary edema in three.Because of the advanced state of decomposition of the bodies, microscopicconfirmation was possible in only one case.
Captain Tribby believed that several other casualties mightalso have died of blast injuries, for while penetrating wounds were present,they were not sufficient, in any instance, to account for the fatality. In onebody, for instance, the only injury was a penetrating wound of the right wrist.
Data prepared by the Medical Statistics Division, Office of The SurgeonGeneral, Department of the Army, show 1,021 blast injuries of nonbattle originin the U.S. Army for the 1942-45 period, of which 48 were fatal (table 1). Forthe same period, there were 13,200 battle-incurred blast injuries outside of the
33
[Preliminary data based on sample tabulations of individualmedical records]3
Area | 1942-45 | 1942 | 1943 | 1944 | 1945 | |||||
Admis- | Deaths | Admis- | Deaths | Admis- | Deaths | Admis- | Deaths | Admis- | Deaths | |
Number | Number | Number | Number | Number | Number | Number | Number | Number | Number | |
Continental United States | 256 | 7 | 47 | --- | 150 | 2 | 34 | 3 | 25 | 2 |
Overseas:4 | ||||||||||
| Europe | 325 | 18 | 3 | --- | 22 | --- | 145 | 5 | 155 | 13 |
| Mediterranean5 | 144 | 10 | 3 | --- | 64 | 1 | 52 | 1 | 25 | 8 |
| Middle East | 4 | --- | 1 | --- | 3 | --- | --- | --- | --- | --- |
| China-Burma-India | 11 | 2 | --- | --- | 2 | --- | 4 | 1 | 5 | 1 |
| Southwest Pacific | 128 | 7 | 5 | --- | 16 | --- | 27 | 3 | 80 | 4 |
| Central and South Pacific | 119 | 4 | 9 | --- | 22 | --- | 18 | --- | 70 | 4 |
| North America6 | 20 | --- | 4 | --- | 12 | --- | 4 | --- | --- | --- |
| Latin America | 9 | --- | 1 | --- | 7 | --- | 1 | --- | --- | --- |
Total overseas | 765 | 41 | 26 | --- | 149 | 1 | 255 | 10 | 335 | 30 |
Total Army | 1,021 | 48 | 73 | --- | 299 | 3 | 289 | 13 | 360 | 32 |
1Includes an unknown, but presumably a relativelysmall, number of cases CRO (carded for record only), mostly deaths. For the twoyears, 1943 and 1945, in which the number of CRO cases was known, CRO casesconstituted 3.2 percent of the nonbattle blast injury "admissions."
2Underlying cause of death, year of death, and area ofadmission.
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 5 admissions aboard transports, 1 in 1943and 4 in 1944.
5Includes North America.
6Includes Alaska and Iceland.
continental United States, of which 140 were fatal (table 2).Of the 6,284 blast injuries which occurred in 1944 and of which 76 were fatal,493 involved the chest, and 25 of these were fatal (table 3). Another eightinjuries, one of which was fatal, involved the thoracoabdominal region (table3). Only 68 of the survivors in both groups (65 with thoracic wounds, 3 withthoracoabdominal wounds) required evacuation (table 4). A remarkable variety ofagents (table 3) were responsible for these injuries, which are generallyconsidered to be caused only by high explosives. It must also be remembered thatdeaths which occurred in blast injuries were not always due to those injuries.This was true in 3 of the 25 deaths which occurred in 1944.
To complete the picture, it should be added that, as mighthave been expected, the Navy experience with blast injury was considerably moreextensive than that of the Army.
34
[Preliminary data based on sample tabulations of individualmedical records]
Area | 1942-45 | 1942 | 1943 | 1944 | 1945 | |||||
Admis- | Deaths | Admis- | Deaths | Admis- | Deaths | Admis- | Deaths | Admis- | Deaths | |
Number | Number | Number | Number | Number | Number | Number | Number | Number | Number | |
Europe | 9,651 | 106 | 1 | --- | 8 | --- | 5,405 | 61 | 4,237 | 45 |
Mediterranean3 | 1,345 | 14 | 15 | --- | 462 | --- | 620 | 12 | 248 | 2 |
Middle East | 1 | --- | --- | --- | --- | --- | 1 | --- | --- | --- |
China-Burma-India | 16 | --- | --- | --- | 2 | --- | 8 | --- | 6 | --- |
Southwest Pacific | 1.455 | 12 | 10 | --- | 33 | --- | 176 | 2 | 1,236 | 10 |
Central and South Pacific | 715 | 8 | 8 | --- | 102 | 1 | 73 | 1 | 532 | 6 |
North America4 | 4 | --- | --- | --- | 4 | --- | --- | --- | --- | --- |
Latin America | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 13,200 | 140 | 34 | --- | 616 | 1 | 6,284 | 76 | 6,266 | 63 |
1Excludes cases carded for record only.
2Among cases who reached a medical treatment facility.Underlying cause of death, year of death, and area of admission.
3Includes North Africa.
4Includes Alaska and Iceland.
5Includes 13 admissions aboard transports: 5 in 1943,1 in 1944, and 7 in 1945.
Experimental Studies
The British experience with German bombings early in WorldWar II led them to carry out a number of experimental studies on blast injuriesof the lung, under the direction of the Research and Experiments Department ofthe Ministry of Home Security. The most important of these studies was reportedby Zuckerman (8) in the Lancet of 24 August 1940. In the sameissue, Dean and his associates (9) reported on the clinical aspects ofthese injuries, and in the issue of 19 October of the same year, Hadfield,Swain, Ross, and Drury-White (10) described the pathologic process asconfirmed by autopsy studies.
Zuckerman's studies on blast injury, which are now classic, were carried outon pigeons and on animals varying in size from mice to monkeys. The experimentalanimals were exposed, in the open, (1) to blast from the explosions of 70 poundsof high explosives and (2) to explosions of hydrodgen and oxygen in balloons.Both sets of experiments produced essentially the same results.
In the experiments with high explosives, no animals were killed at distancesbeyond 18 feet, and none were hurt in any observed manner at distances beyond 50feet. When the bodies were protected by thick layers of rubber,
35
36
[Preliminary data based on sample tabulations of individualmedical records]
Area | THORAX | |||||
Total | Evacuated | Not evacuated | ||||
Admissions | Deaths | Admissions | Deaths | Admissions | Deaths | |
Number | Number | Number | Number | Number | Number | |
Europe | 416 | 20 | 63 | --- | 353 | 20 |
Mediterranean3 | 64 | 5 | 2 | --- | 62 | 5 |
Middle East | --- | --- | --- | --- | --- | --- |
China-Burma-India | --- | --- | --- | --- | --- | --- |
Southwest Pacific | 9 | --- | --- | --- | 9 | --- |
Central and South Pacific | 4 | --- | --- | --- | 4 | --- |
North America4 | --- | --- | --- | --- | --- | --- |
Latin America | --- | --- | --- | --- | --- | --- |
| 493 | 25 | 65 | --- | 428 | 25 |
| THORACOABDOMINAL REGION | ||||||
Europe | 5 | 1 | 3 | --- | 2 | 1 |
Mediterranean3 | 3 | --- | --- | --- | 3 | --- |
Middle East | --- | --- | --- | --- | --- | --- |
China-Burma-India | --- | --- | --- | --- | --- | --- |
Southwest Pacific | --- | --- | --- | --- | --- | --- |
Central and South Pacific | --- | --- | --- | --- | --- | --- |
North America4 | --- | --- | --- | --- | --- | --- |
Latin America | --- | --- | --- | --- | --- | --- |
| 8 | 1 | 3 | --- | 5 | 1 |
1Excludes cases carded for record only.
2Deaths occurring in 1944 among admissions in theaterindicated. In 3 cases, not evacuated, death was ascribed to causes other thanblast injury (see footnote 2, table 3).
3Includes North Africa.
4Includes Alaska and Iceland.
there was either no damage at all, or minimal damage, ascompared to the damage suffered by control animals. This observation, it mightbe interpolated, furnishes additional evidence for the use of body armor.
None of the animals or birds had any external signs of injury. In everyautopsy, the outstanding finding was traumatic pulmonary hemorrhage, whichvaried in degree according to (1) the distance of the animal from the charge and(2) the pressure exerted against the body. The lesions were bilateral unless theanimals were so placed that one side acted as a shield for the other.
37
Then the lesions occurred on the side facing the explosion.This fact, as well as the anatomic sites of the hemorrhage, permitted nointerpretation except that the lesions were caused by the impact of the pressurewave against the body wall. In all instances in which the lung injury wassufficiently severe to cause death, the lesions were detectable onroentgenologic examination, and blood was present in the upper respiratorypassages as well as in the lungs.
Mechanism
Blast may be defined as the compression and suction waves set up by thedetonation of a charge of high explosive. At every point in the neighborhood ofthe explosion there occur:
l. A wave of high pressure, lasting about 0.006 second, according toZuckerman's studies with 70-pound charges, followed by
2. A negative suction pressure wave, lasting up to 0.003 second, produced bythe reduction of the density of the air behind the positive compression wave tobelow the normal atmospheric pressure, which is about 15 pounds per square inch.The suction component of the blast wave is always much weaker than its pressurecomponent; it can never be greater than 15 pounds per square inch, whichcorresponds to a perfect vacuum. It is only because of the very short durationof both components that blast waves are not more destructive than they are.
As hot gases are ejected by a detonating shell, they compress the surroundingair into a shell or belt, which is thrown against adjacent layers of air. Thecompressed air within the belt is characterized by high pressure and high outward velocity. It is limited by an extremely sharp front, the so-called shockfront, which is less than one-thousandth of an inch and in which the pressurerises abruptly.
The initial velocity of the shock front as it travels away from the point ofdetonation is extremely high. Maj. Ralph W. French, MAC, and Brig. Gen. GeorgeR. Callender (11) estimated it as 3,000 feet per second at 60 feet from a4,000-pound light-case bomb where the pressure jump is 100 pounds per squareinch. The velocity then decreases rapidly down to the velocity of sound, whichis about 1,100 feet per second or 750 miles per hour. The velocity of the shockfront can be realized by comparing it with the velocity of gale winds from 50 to60 miles per hour; of hurricanes, 80 to 100 miles per hour; and of tornadoes, inwhich estimated velocities range from 200 to 230 miles per hour.
Because the pressure wave is highest in the region of the explosion and fallsoff rapidly the farther it moves from it, everything in the immediateneighborhood of a bomb explosion will suddenly be exposed to violent pressurewaves of many times atmospheric pressure, while everything 50 feet or more awaywill be exposed to only two or three times atmospheric pressure. The
38
velocity and duration of a pressure wave at any given pointare such that any object as large as the human body would undoubtedly becompletely immersed for an instant in a wave of almost uniformly raisedpressure.
The magnitudes of the pressure and suction components of a blast wave aredirectly correlated with the amount of explosive. Zuckerman's studies showed,however, that if a given positive pressure is caused by a given amount ofexplosive at a given distance, the same degree of pressure will be experiencedat twice that distance only when the amount of explosive is increased eighttimes.
Thus all objects in the immediate neighborhood of anexplosion are first subjected to violently increased wind and hydrostaticpressure, which may tear them to pieces and blow them far from the scene of theexplosion. If they are not shattered by the pressure wave and blown along in itsdirection, they may be pulled toward the center of the explosion by the weaker,but longer acting suction wave.
Pathologic Process and Causes of Death
The pathologic process which results from the contusive effect of a blastwave on the chest arises, according to Zuckerman's (8) experimentalstudies, from the pressure component of the blast, which bruises the lungs byits impact on the body wall. It varies from small ecchymotic areas on the lungsurface to such extensive lesions that the lung may appear hepatized.
In the 10 autopsies performed by Hadfield and his associates (10) oncivilians who died suddenly, or within a few hours, after short-range exposureto detonation of high explosives during aerial bombings, gross traumatic lesionswere entirely absent or were trivial in all but one instance. In eight cases,death was due to the effects of blast, though in three cases carbon dioxidesaturation of the blood was so extreme that it was considered the immediatecause of the fatality. Two casualties who were extricated from overlying debriswithout visible injuries were first thought to have died of blast. Furtherexamination showed that both deaths were caused by compression asphyxia.
As in Zuckerman's experimental studies, intrapulmonary capillary hemorrhagewas the single gross anatomic lesion common to all cases in which deaths weredue to blast. There was free capillary bleeding over large areas, and in theseareas, the bronchioles, atria, and alveoli all showed uniform and considerableoverdistention. In the fatalities due to compression asphyxia, hemorrhage wasrelatively slight. In these cases, the air passages contained only a smallamount of blood-stained fluid, which was not frothy, but capillary and venouscongestion and edema were striking. In the fatalities due to carbon dioxidesaturation, the pulmonary hemorrhage was of the same character as in the trueblast deaths, but the blood was fresh and pink, not dark. In both groups, theair passages contained quantities of frothy, serous fluid. Tribby (7)noted that in all 13 cases in which he believed death on the battlefield to bedue to blast, there was blood in the nose or mouth or in both in every instance.
39
Subpleural hemorrhage was not conspicuous in the blast deaths studied byHadfield and his group (10). The only casualty in the series who showedhemorrhagic rib markings in the pleura died from compression asphyxia, not fromblast. Bleeding into the walls of the smaller bronchioles was occasionallyobserved, but there was a conspicuous absence of hemorrhage into the largerstructures. There was no suggestion that hemorrhages were grouped around thebronchial system. Subpleural bullae, observed microscopically, had apparentlybeen produced by detachment of the visceral pleura and its subjacent elastictissue from the underlying lung by air escaping from ruptured alveoli.
The most severe hemorrhages were found in two young children, one an infant,the other 13 years of age; the possible explanation was the lesser rigidity ofthe thoracic wall in youth. Bleeding was minimal in the only notably obesecasualty in the group; it may be presumed that his excess flesh protected himfrom the most serious effects of the blast.
Even making due allowance for the rapidity of theextravasation of blood into the lungs as the result of blast, in no instance wasthe amount of blood found at autopsy sufficient to cause fatal circulatoryembarrassment. The conclusion was that blast probably produces death byinterference with some vital tissue or center, in which, because of the extremerapidity of the process, the structural changes that occur are not detectable.Hadfield and his associates (10), however, considered the presence ofhemorrhage a trustworthy indication that the patient had been struck at closerange by a wave of high pressure.
In general, autopsy studies on victims of blast in the Mediterranean theaterwere in accord with these observations (fig. 10).
Clinical Picture and Diagnosis
Symptoms and signs-The clinical symptoms and signs of blast injury asobserved in World War II were as follows:
l. Shock.-This was a universal finding and was oftenprofound. As a rule, the degree of shock was directly proportional to theseverity of the injury. It was often increased by, or was more serious becauseof, associated injuries in other parts of the body.
2. Restlessness.-This finding was often extreme and out of proportionto the evident severity of the injuries.
3. Chest pain, which was of two types.-Almost allcasualties from blast complained of pain located laterally and related directlyto respiratory movements. This type of pain was considered due to contusion andhemorrhage of the intercostal muscles, with resulting muscular spasm. The othertype of pain appeared in the more severe injuries. It was deep, central, and notrelated to respiratory movements. It lasted only a few days and was consideredto be caused by mediastinal hemorrhage.
40
4. Cough and expectoration, which were present in all butthe mildest injuries.-When the expectorated material first appeared, about24 hours after injury, it was thin and mucoid. Later, it became thick andmucopurulent. Often, it was streaked with dark blood, and in an occasional case,there was free hemoptysis. Expectoration usually lasted about 10 days. Almostall dead or dying casualties were found to have frothy, blood-stained fluid inthe nose and mouth.
5. Abdominal pain and rigidity.-These findings, which were present inonly a few blast injuries, were explained by extrapleural hemorrhages, which hadan irritative effect on the intercostal nerves and muscles. In an occasionalcase, the persistence and prominence of these findings provided an indicationfor laparotomy.
6. Partial fixation of the chest in the position of inspiration.-Movement,although limited, was equal on both sides.
In the absence of complications, the percussion note wasresonant. Breath sounds were usually weaker than normal, especially at thebases, and coarse bronchial rales were frequently heard in both lungs.
41
These observations are, in general, in correspondence withthose reported by Dean and his associates (9) in 27 patients whom theyexamined from 7 to 10 days after they had been close to bursting high explosivebombs. Three of the casualties had been immersed. External injuries includedextensive, but superficial, burns in 21 cases; fractures in 5 cases; andmultiple splinter wounds of the leg in 1 case.
There were no obvious chest injuries, and only six patientshad symptoms referable to the chest. None complained of chest pain orhemoptysis. In no instance did the symptoms develop on the day of bombing; allappeared between the second and fifth days. Sixteen patients had abnormalphysical signs, in 15 instances the characteristic fixation of the chest justdescribed; in 1 instance the intercostal spaces shared the fullness.
Roentgenologic examination-If the classical pathologicpicture was present, roentgenograms showed heavy mottling over large areas ofthe lung fields, corresponding with the interstitial and alveolar hemorrhagesobserved, and varying in size and density with the extent of the lesions. Theroentgenologic findings suggested those observed in patchy pneumonia. Theydisappeared within a week in mild injuries but persisted for weeks in severeinjuries. Roentgenologic abnormalities were present in 14 of the 27 patientsstudied by Dean and his associates.
Diagnosis-The first consideration in a suspected blast injury was ahistory of exposure, which, in spite of its importance, had to be interpretedwith caution. The effects of blast were often a factor in wounds from highexplosives, but the seriousness of the penetrating wounds was likely toovershadow them.
Other diagnostic criteria consisted of the presence of shock; the variousrespiratory symptoms described; the characteristic bulging of the lower portionof the chest, which was held almost immobile in the inspiratory position; andthe finding of blood in the lungs and air passages. Ruptured eardrums werepathognomonic. Roentgenologic findings were confirmatory.
The essential features of diagnosis were:
1. The absence of any significant external evidence of violence to the chest.
2. Hemorrhagic lesions in both lungs, as shown by the character of the fluidexpectorated or, less often, by hemoptysis, or by autopsy confirmation.
Management
The management of blast injury was never really satisfactory. The routineconsisted of the following measures:
1. The treatment of shock.-However deep was the stateof shock and however much associated injuries demanded intensive measures,resuscitation had to be carried out cautiously from the standpoint ofreplacement therapy. Injudicious use of any fluid, including blood, mightincrease pulmonary edema. An occasional patient, in fact, in whom edema andcongestion were particularly marked, responded to venesection, which wasemployed as a lifesaving measure.
42
Some medical officers, in order to reduce the amount of fluid infused,administered serum albumin in 100-cc. doses every 4 hours for three or fourdoses.
2. The continuous administration of oxygen.-This measure was usedespecially if cyanosis was a feature of the injury, or intermittent positivepressure oxygen.
3. The cautious use of morphine.-Morphine was used on strictindications and in doses of not more than gr. 1/8 if paincould not be controlled otherwise.
4. Positioning.-Since the bases of the lungs were usually involved inthe hemorrhagic process, the best position was with the shoulders raised. In theoccasional instance of unilateral blast injury, the patient was placed on thedamaged side, to keep the blood from trickling down the bronchi into theunaffected lung.
5. Local nerve block.-This measure was employed to control pain andpermit free respiration.
6. Other medication.-The administration of adrenalin was not withoutrisk, but it was occasionally used to overcome a temporary spasm of thebronchioles. Atropine was also occasionally used to diminish exudation. Asulfonamide, or penicillin when it became available, was used as preventivemeasure.
If surgery was necessary for other wounds, it was postponed as long aspossible, and then was done under local analgesia or with Pentothal sodium(thiopental sodium). General anesthesia was avoided.
Case History
The following case history concerns the only valid blast injury recorded inthe experience of the 2d Auxiliary Surgical Group:
Case 6-This soldier was riding in the back seat of acommand car when the rear wheels ran over a landmine. When he recoveredconsciousness some minutes later, he was lying on the ground, beside a tree. Hisonly complaint was a bilateral hearing impairment. Speech and cerebration werenot impaired, and he had no dizziness, headache, or blurred vision. The onlyexternal wounds were a small puncture wound anterior to the left ear, near thetemporal artery; a compound fracture of the right ankle; and a simple fractureof the left ankle.
Wire splints, extending above the knee, were applied in a collecting station.Later, the wound of the right leg was debrided, and plaster casts were appliedto both legs.
When the patient was admitted to the 38th Evacuation Hospital, 52 hours afterinjury, he was in good condition, was talking rationally and normally, and hadno complaints. Both pupils were equal and regular. The toes on both sides werewarm.
Four hours after admission, the pulse rate began to range between 125 and130. A little later, it rose to 140; at this time the respiration was 32 and theblood pressure 98/60 mm. Hg. After 500 cc. of plasma had been given, the pulsefell to 120, and the blood pressure was 98/50 mm. Hg. The respiration was 40 andthe temperature 100.4? F. The patient was very drowsy, but he could be aroused,and he was then mentally clear.
After a brief interval, his condition rapidly worsened; the blood pressuredropped, and the pulse rate became very rapid. After the foot of the bed hadbeen elevated and 500 cc. of blood and 750 cc. of plasma had been given, thesystolic blood pressure was recorded at 100 mm. Hg, but the diastolic pressurecould not be obtained. Shortly afterward, the systolic pressure fell to 70 mm.Hg.
43
About this time, moderate respiratory distress developed, and the patientbegan to spit up frothy blood. The breath sounds were spottedly distant, andcoarse rales were heard over the entire chest. There was considerableconsolidation of both bases, with a marked friction rub. The upper abdomen wasdistended, and no peristalsis could be heard. Voiding was involuntary.
Respiratory distress increased rapidly, and the patientbecame slightly cyanotic. Frothy fluid, streaked with bright red blood, wasbeing expectorated. After the administration of two ampules of Coramine (nikethamide)and 500 cc. of blood, examination of the chest revealed diminished breath soundsand tactile fremitus on the left, with a loud to-and-fro friction rub. Oxygentherapy was instituted. The blood pressure was now 90/40 mm. Hg. The patient,although struggling for breath, remained lucid enough to answer questions. Hisdyspnea did not seem of the obstructive type.
Death occurred suddenly, 72 hours after injury and 24 hours after admissionto the evacuation hospital. There was no response to artificial respiration.
Autopsy disclosed an entirely clear peritoneum and retroperitoneal space. Thethoracic cage and diaphragm were intact, as were both pleural cavities.
The lungs were expanded, and there were numerous areas of hemorrhage andcrepitation. There was no blood in the pleural cavity, but when the bronchialtubes were sectioned, they were found to contain frothy, serous fluid. The heartwas grossly normal except for a questionable auricular thrombosis. The thrombiin the pulmonary artery were regarded as a post mortem development.
Histologic examination of the lungs showed extensive areas of intra-alveolarextravasations of red cells intermingled with varying numbers of pigmentedmacrophages. Occasional alveoli were filled with fluid containing a few redcells. Other alveoli, some emphysematous, were interspersed among them. Some ofthe emphysematous alveoli had ruptured walls. The bronchial lumen containedlarge numbers of red blood cells.
The gross pathologic diagnosis (in addition to the fracturesof the lower extremities) was possible blast injury of the lung, possiblethrombosis of the pulmonary artery, and possible thrombosis of the cardiacauricle. The histologic diagnosis was severe pulmonary hemorrhage, rupturedalveoli, congestion of the spleen, and epicardiac hemorrhage.
Comment-This injury was difficult to evaluate from theclinical standpoint, and the diagnosis of blast syndrome was not made earlyenough for therapy to be precise and effective. One can only speculate whetherearlier recognition of the blast injury, with intensive treatment directedtoward it, would have altered the outcome.
Unfortunately, no special fat stain was used on the tissues.In some blast injuries, fat emboli were found in the lungs, as they were in somecompound fractures; they were more often seen after traumatic amputations fromlandmines. The diagnosis of blast injury of the lungs was made in this case onthe basis of the ruptured alveoli.
ASSOCIATED WOUNDS
A large proportion of thoracic injuries, as pointed outelsewhere (vol. I), were associated with injuries in other parts of the body,and these associated injuries were frequently more serious than the chestinjuries. That other injuries might exist had to be taken for granted in theapproach to all chest injuries. In other words, regardless of how specialized orrestricted an injury might seem, a total therapeutic approach was necessary. Incaring for a casualty, the surgeon had to conceive of him as a total organism,not as an assembly of individual injuries. The management of the chest injurytherefore had to be related to the management of all other injuries.
44
Compound Fractures
Compound injuries of the scapula and shoulder girdle might furnish numerousproblems. A missile of high velocity that struck the scapula a glancing blowcould cause extensive damage to the heavy muscles in the vicinity, in additionto shattering and fragmenting the bone. Although there was always considerablequestion as to how radical one should be in dealing with fractured ribs, therewas rather general agreement that fractures of the scapula, especiallysubscapular fractures, because of the less adequate drainage possible in thisarea, required more radical management than fractured ribs in other locations.Infection invariably resulted if the initial debridement was not complete. Wideexcision of the damaged muscles was necessary, because of the great danger ofanaerobic cellulitis in this area.
Fragments of the alar portion of the scapula, if left in situ, almostinvariably led to continued suppuration. Even when the scapula itself was notdamaged, it was necessary to bear in mind the possible involvement of thesubscapular space when reconstruction of the course of a missile that hadpenetrated the chest wall indicated this possibility. Abscesses in this areacould remain unrecognized for long periods of time.
Debridement, to be entirely adequate, would have required an incisionapproaching in magnitude the incision necessary for thoracoplasty and would alsohave required mobilization of the scapula. Such radical measures could not bejustified on the mere possibility that a clostridial infection might develop.The usual procedure was to make the incisions along the track of the missile atthe vertebral and axillary borders of the scapula, then to perform as complete adebridement as the exposure permitted. Great care had to be taken during surgeryin this area not to disturb the attachment of the serratus anterior; damage toit could be followed by serious functional losses. As much as possible of thescapula was preserved, particularly of the functionally important upper third.Less hesitancy was felt about removing as much as necessary of the lowertwo-thirds. Through-and-through Penrose drains were inserted, and a dependentdrain was placed in the subscapular space through an incision in the angle ofauscultation at the interior border of the scapula.
The patient was kept in the forward hospital in which operation had been donefor at least 6 days after operation. The wound was examined immediately if therewas a rise in the pulse rate or if he complained of local pain or if hisdemeanor changed, as it characteristically does in clostridial infections.
In six cases treated by the technique just described, Major Shefts and Capt.(later Maj.) Ernest A. Doud, MC, had no deaths. On the other hand, Maj. (laterLt. Col.) Reeve H. Betts, MC, and Maj. William M. Lees, MC, lost a patient fromclostridial infection after what was thought to be a very radical debridement.It was their custom thereafter to use large paravertebral incisions in injuriesof the subscapular area, even though intrapleural damage had not occurred.
45
When fractures ordinarily treated by traction occurred inassociation with wounds of the chest, some compromises in therapy werenecessary. Prolonged immobilization in traction was inimical to satisfactoryprogress in chest injuries. The patient had to be kept maneuverable. Precisefracture alinement could not be achieved at the expense of thoracic crippling.Fortunately, the orthopedic repertoire was extensive enough and flexible enoughto permit the application of a regimen that would reduce the fracturesatisfactorily without jeopardizing the outcome of the chest lesion.
Vascular Injuries
Injuries to the great vessels of the axilla and at the root of the neck werenot always recognized at initial wound surgery for chest injuries, and carefulobservation was necessary in base hospitals. Otherwise, important delayedsequelae of the vascular injuries might be overlooked. It was particularlynecessary to be suspicious whenever the point of entry of the missile was overthe apical region of the thorax. In one such case (p. 244), on the basis ofroentgenologic examination and the repeated aspiration of clotted blood from thechest, the casualty, who had sustained a penetrating wound of the right apicalregion 3 months before, was thought to have only a hemothorax. His progress wassatisfactory until he had a sudden attack of dyspnea and pain in the rightchest. Soon after, he went into profound shock and died promptly. Necropsyrevealed a rupture of a traumatic aneurysm of the first portion of the rightsubclavian artery.
It was always necessary, when operation was undertaken for early massiveempyema or clotted hemothorax, to remember the possible presence of a traumaticaneurysm of the subclavian artery, as in the case just described, or of theinnominate artery. Ignorance of the presence of the vascular lesion could easilyresult in catastrophe on the operating table.
In a case observed by Major Brewer at the 21st General Hospital, the patient,who had a foreign body in the chest wall, was conscious of a bruit that causedhim considerable annoyance. It also made him so apprehensive that he was afraidto perform even moderate activity. A thrill was present, but there was noerosion of the rib and no cardiac embarrassment. A large hemothorax hadpreviously been completely evacuated by repeated thoracenteses. After adiagnosis had been made of aneurysm of the intercostal artery and vein, thelesion was excised, for prophylactic reasons, and the foreign body in the chestwall was removed. The patient was able to return to full duty. It is believedthat this is the only case of the kind to be observed in the Mediterraneantheater during the war.
If ligation of the subclavian or axillary vessels had been necessary in aforward hospital, thoracic surgeons in general hospitals had to determinewhether or not an adequate collateral circulation existed. Measures to maintainthe circulation, or to increase it if it was deficient, were imperative.
46
The most useful measures to control edema were properpositioning of the hand and arm; intelligently applied massage; and procainehydrochloride block of the inferior cervical ganglion, repeated every 24 to 48hours as necessary.
As a rule, casualties with arteriovenous fistulas resultingfrom combined thoracic-vascular injuries were evacuated to the Zone of Interiorfor final surgery. If, however, evacuation was delayed for an inordinately longperiod and compression tests revealed a satisfactory collateral circulation,operation was done overseas. Quadruple ligation with excision was the procedureof choice.
Neurologic Injuries
Brachial plexus-The injuries of the brachial plexus frequentlyassociated with chest injuries constituted a neurologic problem. Carefulexamination in both forward and base hospitals was necessary in every chestinjury, to determine their presence or absence. It was often impossible tosettle this matter positively in the first few days after injury; during thisperiod, the neurologic findings present might be caused by contusion of thenerve as well as by laceration. If the injury was only a contusion, sensationand function would return promptly.
Head injuries-Head injuries were frequently associated with thoracicinjuries, and patients with head injuries were particularly prone to chestcomplications. Atelectasis and pneumonitis were ever-present threats,particularly in patients who were deeply comatose. Chest surgery was kept to anabsolute minimum in these circumstances. Careful nursing, drainage of thetracheobronchial tree by catheter and bronchoscope, and the administration ofpenicillin were all employed to avoid serious and fatal pulmonary complications.
Spinal injuries-In one series of 768 chest injuries studied by the 2dAuxiliary Surgical Group, there were 23 spinal cord injuries. Fractures of thevertebrae were also frequently associated with war wounds of the chest.
Roentgenologic examination of the spine was routine whenever there was anycomplaint of pain in the back or the cervical region. Some injuries weredetected only when the patients became ambulatory; they had fallen or beenthrown with considerable force when they were injured, and the simple vertebralfractures they had sustained were not noticed during their initial treatmentbecause of the greater importance of other injuries.
Associated injuries of the chest and the spinal cordconstituted one of the most distressing complexes seen in war surgery. The closeanatomic juxtaposition of the structures of the spine and thorax made itinevitable that major injuries should affect both regions in a considerablenumber of cases. Casualties with these so-called thoracospinal wounds were opento the complications inherent in both lesions. There was a marked predilectionfor paralyzed patients with chest wounds to develop infection of either thepleura or the
47
pulmonary parenchyma. Hemothoraces were much more likely to become infectedthan in patients without spinal cord lesions.
The most important component of management of these combined injuries wasintelligent nursing care (12). Patients were nursed in the face-downposition, whenever this was practical, and were turned at frequent intervals, toprevent the development of decubitus ulcers. Air mattresses were used when theywere available. Coughing and breathing exercises were carefully supervised.Repeated thoracenteses were performed, to dry up the pleural cavity as rapidlyas possible. If empyema developed rib-resection drainage was usually employed;these patients were seldom in condition for the more radical procedure ofdecortication.
Patients with combined spinal cord-thoracoabdominal injuries wereparticularly difficult to care for. Many of the wounds were tangential, a typelikely to be associated with far more pulmonary contusion and damage than theaverage penetrating or perforating wound. It was impossible for the patients tocough effectively because of paralysis of the lower intercostal and abdominalmuscles. The flared costal angle, the segmental motion of the chest wall, therelaxed abdomen, and the feeble bechic blast were all characteristic of thesecombined wounds. In many cases, tracheobronchial aspiration was necessary asoften as every 1 to 3 hours. Contrary to the usual observation, support of theabdomen and lower ribs by binders or adhesive strapping seemed to make coughingmore efficient.
It was observed in certain patients with combined injuries of the chest andspinal cord that the level of anesthesia caused by the spinal injury seemed toplay a considerable part in the outcome from the standpoint of the chest injury.Death sometimes occurred from intractable pulmonary edema. There was no responseto positive pressure oxygen therapy or endotracheal suction, even though thechest wounds were not in themselves potentially lethal.
Analysis of a number of these fatalities led to the conclusion that thelethal factor was paralysis of the abdominis recti muscles. Normally, the toneof the recti in a gentle cough, or their actual contraction during a vigorouscough, provides the diaphragm with opposing intra-abdominal pressure. When thepressure was lost by reason of paralysis of the recti, the cough becameineffective; the expulsive action resulted in nothing more than bulging outwardof the abdomen; and intractable pulmonary edema was the consequence.
This chain of events was clear in two of four cases ofcombined spinal cord-chest injuries treated in a field hospital in theMediterranean theater by Major Shefts and Captain Doud. All four patientspresented a typical picture of transverse myelitis, and all had pleuropulmonarydamage of about the same degree. The two fatalities were directly related to thelevel of cutaneous anesthesia. They occurred 4 and 5 days, respectively, afterdebridement of the chest wounds, from pulmonary edema that did not respond toany measures, including positive pressure oxygen and endotracheal suction. Onepatient had anesthesia from below the xiphoid process and the other from justbelow
48
the nipples. The two patients who survived had skinanesthesia below the umbilicus, one just below the umbilicus and the otherhalfway between the umbilicus and the symphysis pubis.
While a defeatist attitude was never permitted in combined spinalcord-thoracic injuries, it would have been unrealistic not to recognize theirpotential lethality. In many cases, therefore, the frank objective of treatmentwas to salvage these unfortunate men long enough to permit them to be returnedto the United States, so that they could spend their remaining days at home andwith their families. This was possible in many of the most apparently hopelesscases.2
References
l. Dunham, E. K., Stevens, F. A., Graham, E. A., and Keller, W. L.: Empyema.In The Medical Department of the United States Army in the World War. Washington:Government Printing Office, 1924, vol. XI, pt. 2, pp. 33-392.
2. Graham, E. A.: A Reconsideration of the Question of the Effects ofan OpenPneumothorax. Arch. Surg. 8: 345-363, January 1924.
3. Brewer, L. A. III, Bai, A. F., King, E. L., Wareham, E. E., and Farris, J.M.: The Pathologic Effects of Metallic Foreign Bodies in the PulmonaryCirculation. A Long-Term Experimental Study. J. Thoracic Surg. 38: 670-684,703-706, November 1959.
4. Yates, J. L.: Wounds of the Chest. In The Medical Department of the UnitedStates Army in the World War. Washington: Government Printing Office, 1927, vol.XI, pt. 1, pp. 342-442.
5. Sweet, R. H.: Transthoracic Resection of the Esophagus andStomach for Carcinoma. Analysis of the Postoperative Complications, Causes ofDeath, and Late Results of Operation. Ann. Surg. 121: 272-284, March 1945.
6. Heuer, G. J.: Traumatic Asphyxia; With Especial Reference to Its Ocularand Visual Disturbances. Surg., Gynec. & Obst. 36: 686-696, May 1923.
7. Tribby, W. W.: Examination of One Thousand American Casualties Killed InAction in Italy. Report to Surgeon, Fifth U.S. Army, 1944, 6 vols. [Officialrecord.]
8. Zuckerman, S.: Experimental Study of Blast Injuries to the Lungs. Lancet2: 219-224, 24 Aug. 1940.
9. Dean, D. M., Thomas, A. R., and Allison, R. S.: Effects of High-ExplosiveBlast on the Lungs. Lancet 2: 224-226, 24 Aug. 1940.
10. Hadfield, G., Swain, R. H. A., Ross, J. M, and Drury-White, J. M.: BlastFrom High Explosive. Preliminary Report on Ten Fatal Cases, With a Note on theIdentification and Estimation of Carboxyhaemoglobin in Formol-Fixed Material byArthur Jordan. Lancet. 2: 478-481, 19 Oct. 1940.
11. French, R. W., and Callender, G. R.: Ballistic Characteristics ofWounding Agents. In Medical Department, United States Army. Wound Ballistics.Washington: U.S. Government Printing Office, 1962, pp. 91-141.
12. Medical Department, United States Army. Surgery in World War II.Neurosurgery, Volume II. Washington, U.S. Government Printing Office, 1958, pp.31-65: 127-191.
2The reader is referred to chapterXI (p. 441)for long-term followup studies of the various types of wounds described in thischapter.
