U.S. Army Medical Department, Office of Medical History
Skip Navigation, go to content







AMEDD MEDAL OF HONOR RECIPIENTS External Link, Opens in New Window






Chapter 9



Shock Therapy (957 Casualties)

Beverly T. Towery, M. D., and John D. Welch, M. D.

From the standpoint of the therapy of traumatic shock,1 soldiers with major wounds of the abdomen presented more serious problems than any other group of casualties. There were a number of reasons:

1. Hemorrhage was frequently massive.

2. Contamination of the peritoneal cavity and retroperitoneal tissues carried the threat of fulminating infection.

3. Concomitant transdiaphragmatic injury of the thorax frequently caused serious disturbances of cardiopulmonary function.

4. These same injuries rendered the pleura liable to contamination by gastrointestinal contents or bile.

5. Multiple visceral and vascular injuries, which were common, often created surgical problems of great technical difficulty, with the result that the duration of anesthesia and operation was necessarily prolonged and further blood losses were likely to occur.

The problem of traumatic shock is considered in detail elsewhere in these volumes.2 Resuscitation therapy, however, played such an important role in the management of casualties with abdominal injuries that it must be discussed in some detail here. Furthermore, the important causes of shock deserve careful consideration in relation to therapy and therapeutic failure in these injuries.

The background of this discussion is a series of 957 casualties selected for this purpose from the whole series of 3,154 casualties with abdominal wounds who were treated by surgeons of the 2d Auxiliary Surgical Group during 1944 and 1945. Selection was on two grounds: (1) That all the injuries included perforations of hollow viscera and therefore carried the threat of fulminating peritonitis, and (2) that in all cases data relative to shock therapy were complete. The figures for replacement therapy, as well as the case fatality rates, must be interpreted with the reservation that the selected series perhaps contains an undue proportion of poor-risk cases, the explanation being that the data were more likely to be complete in such cases than in cases in which the surgical risk was better.

1The term "shock" throughout this discussion should be understood as synonymous with the term "traumatic shock." An attempt to distinguish between traumatic and hemorrhagic shock would serve no useful purpose.
2Medical Department, United States Army, Surgery in World War II. The Physiologic Effects of Wounds. Washington: U. S. Government Printing Office, 1952.


The initial care of casualties in World War II fell into three phases: (1) Preoperative resuscitation or the therapy of traumatic shock, (2) definitive surgery, and (3) postoperative care. Care could not be considered optimal unless these three phases were closely integrated to insure complete continuity of effort. The immediate aim of shock therapy was to restore the circulatory dynamics to a sufficient state of competence to permit a successful operation. If a fatal outcome from the deleterious effects of shock, continuing peritoneal contamination, and the establishment of widespread virulent infection was to be prevented, it was essential that both resuscitation therapy and surgery be accomplished with as little delay as possible. Moreover, replacement therapy frequently had to be continued during operation and throughout the first days after operation.

The medical officer responsible for triage and shock therapy played a unique role in the care of battle casualties. His chief asset was the ability to judge the gravity of the situation. His initial examination established the status of the patient with respect to the extent of injury and the severity of shock and permitted tentative decisions as to appropriate replacement therapy and the response to be anticipated from it. Thereafter he carefully followed the patient's progress during transfusion therapy so as to recognize promptly any failure of response or actual deterioration. Sound judgment was especially needed in reaching a decision as to the optimal time to operate upon a patient with continuing intra-abdominal hemorrhage. Often the shock officer had to decide who, among a number of wounded men, was in most urgent need of immediate surgery, as well as whose chance of survival would be least compromised by delay.

Intelligent triage during the period of resuscitation was difficult to achieve because it depended upon awareness, experience, and judgment which were not easily acquired. Nevertheless, skill in this, as in surgical technique, was an extremely important factor in the successful treatment of the seriously wounded.

The objective of treatment was thus a continuity of therapeutic endeavor which afforded optimum coordination of replacement and surgical therapy and which recognized the fundamental importance of operation in the resuscitation of the severely wounded. The operation was of special importance in resuscitation when, because of continuing intra-abdominal hemorrhage or early fulminating infection, transfusion therapy alone had failed to bring about sustained improvement. Obviously, more than blood replacement was required to correct these conditions. The only hope for a successful outcome-and frequently the hope was slight-lay in supplementing massive transfusion therapy by prompt and skilled surgery. The care of these casualties thus constituted a challenge to all those responsible for all phases of their management in forward hospitals.



The classification of shock into categories for statistical purposes is never entirely satisfactory, since the objectivity of different observers naturally varies considerably and their interpretations vary accordingly. For purposes of comparison and reference, however, four degrees of shock (table 18) were recognized in the 957 instances of traumatic perforation of the gastrointestinal tract upon which this discussion is based: (1) No shock, or incipient shock, in which the lower level of the systolic blood pressure was 101 mm. Hg; (2) moderate shock, in which the systolic blood pressure was 71 to 100 mm. Hg; (3) severe shock, in which the systolic blood pressure was 41 to 70 mm. Hg; and (4) profound or preterminal shock, in which the systolic blood pressure ranged downward from 40 mm. Hg to 0. More than half of the 957 patients in this series (table 18) exhibited degrees of shock ranging from moderate through preterminal, and more than a quarter exhibited severe or profound shock.

TABLE 18.-Administration of replacement therapy in relation to degree of shock in 957 traumatic perforations of the gastrointestinal tract

Degree of shock




Case fatality rate

Timelag (average)

Replacement therapy (average)


During operation
















None or incipient (101 to 120 mm. Hg.)












Moderate (71 to 100 mm. Hg.)












Severe (4 to 70 mm. Hg.)












Profound (0 to 40 mm. Hg.)


















The chief shortcoming of a classification based on the level of the blood pressure lies in the fact that the blood pressure was often maintained at relatively normal levels for a considerable period of time following injury. As a result, the soldier's condition might seem better than it really was (cases 1 through 4).3 Oligemia and reduction of the peripheral blood flow in the patients classified as not in shock or in incipient shock were likely to be disproportionately greater than the level of the systolic pressure would indicate. This was not true, however, of the other categories. All the experience in forward hospitals indicated that when once the blood pressure had fallen below normal levels, certain therapeutic and prognostic implications were inherent in the initial reading.

3For the protocols on these cases and the other cases referred to in this chapter, see a Clinicopathologic Study of Abdominal Wounds, at the end of the chapter.



The average timelag between wounding and operation in these 957 casualties with traumatic perforations of the gastrointestinal tract varied from 10.4 to 11.6 hours (table 18). Aside from the fact that it was slightly less in patients with no shock and with incipient shock than in those in other categories, there was no correlation between timelag and the severity of shock as indicated by the level of the blood pressure. Obviously, in wounds of equal severity, the total amount of blood lost and the severity of shock would increase with the passage of time (case 4). A patient admitted 10 hours after injury, for example, might exhibit signs of severe shock, whereas if he had been admitted 4 hours after injury, only moderately severe shock might have been observed. 

The correlation between the degree of shock and the severity of the wound is borne out by an analysis of the cases in this series. Not only did the patients in the deepest shock consistently exhibit more severe injuries than those whose blood pressure approached normal, but multiple visceral injuries were also more common in the categories of severe shock (cases 5 through 8). The high case fatality rate in patients with combined gluteal and abdominal wounds was almost certainly related to the frequency in such injuries of massive hemorrhage from the iliac and femoral vessels (cases 1, 5, 6, and 9). Lacerations of the spleen and major vascular injuries furnished other illustrations of the relation of vascular injuries to the degree of shock. Of the 140 patients with gastrointestinal perforations in the category of profound or preterminal shock, 43 (30.7 percent) had associated injuries of this sort, as did 21 of the 121 patients (17.4 percent) in the category of severe shock. In contrast, this type of injury was present in only 39 of the 250 patients (15.6 percent) in moderate shock, and in only 37 of the 446 patients (8.3 percent) in incipient shock or not in shock.


Massive hemorrhage was frequently present in injuries of the abdomen, as the result of disruption of a major vessel or a laceration of the spleen or the liver. Even in the absence of such major injuries, widespread damage of smaller vessels could account for a considerable blood loss. Blood loss was undoubtedly the most important factor in the production of traumatic shock in patients with abdominal injuries. It was also chiefly responsible for the depression of the hematocrit frequently observed in battle casualties. On the other hand, it had to be remembered, in the appraisal of severe shock, that the administration of plasma also tended to reduce the hematocrit and the oxygen-carrying capacity of a given volume of blood. If an excessive amount of plasma had been given after hemorrhage, the total volume of the circulating blood might be only moderately reduced; yet a marked anemia might be present, with a hematocrit reading of only 15 to 18 percent. This observation, incidentally, furnishes additional proof of the superiority of whole blood over reconstituted plasma in the resuscitation of battle casualties (p. 22).


In addition to hemorrhage, the loss of plasma or of plasma components into the traumatized or contaminated area of the abdomen contributed materially to the production of shock. This loss occurred as the result of transudation or exudation of fluid from inflamed peritoneal surfaces and into visceral tissues under the influence of mechanical and chemical trauma and beginning infection. While the precise importance of fluid loss from this source has been disputed, certain observations made by the surgeons of the 2d Auxiliary Surgical Group caused them to be inclined to regard it as significant. An extremely high case fatality rate was observed when the pleura was widely contaminated by gastric or intestinal contents, and the shock present in this type of case was apt to be especially severe (cases 4 and 10). Similarly, diffuse peritoneal contamination consistently increased the severity of the circulatory collapse associated with evisceration of the intestines (case 11). When, on the other hand, evisceration occurred without appreciable peritoneal soiling, shock was often surprisingly mild, and the response to replacement therapy was more favorable than in the presence of diffuse contamination of the peritoneum.

Changes in the splanchnic capillary bed produced by irritation and infection also appeared to be of some importance in reducing the effective circulating blood volume, though there was no agreement concerning the quantitative influence exerted by these alterations. Mann4 had noted in 1915 that mechanical trauma to the bowel was associated with an increased capacity of the splanchnic vascular bed. Similar observations were made by the surgeons of the 2d Auxiliary Surgical Group who found that, in patients with perforations of the gastrointestinal tract, a considerable volume of blood was likely to be pooled in dilated and engorged venules and capillaries and that the accumulation was increased when the bowel was manipulated at operation.

Another source of fluid loss was dehydration, which existed in some degree in all patients with gastrointestinal perforations, whether or not vomiting had occurred. Because of paralytic ileus, which was a natural consequence of peritoneal contamination or infection, small-bowel contents did not reach the colon, and the normal reabsorption of water did not occur.

Although the loss of plasma or its components into the contaminated peritoneal cavity resulted in a rise in the hematocrit of the circulating blood, hemorrhage, which was an almost universal sequela of penetrating wounds of the abdomen, ultimately had the opposite effect. The hemodilution which follows hemorrhage would seem to account for the fact that even in severe shock associated with widespread soiling of the peritoneum, hemoconcentration was seldom observed and was, in itself, a factor of negligible significance in the pathogenesis of shock in battle injuries.

Furthermore, a patient who showed clinical evidence of severe shock might occasionally present no significant decrease in the hematocrit level, and there was frequently little apparent correlation between the hematocrit reading and the severity of peripheral circulatory failure. On the other hand, con-

4Mann, F. C.: Shock and Haemorrhage: An Experimental Study. Surg., Gynec. & Obst. 21: 430-441, October 1915.


siderable hemodilution was often found in association with severe shock. These low hematocrit levels, as already mentioned, were usually observed in cases in which the initial blood loss had been severe and in which large amounts of plasma had been used to prevent death within a short time after injury. The subsequent management of these patients was always difficult, a fact that was not always readily apparent at the time of the initial examination.

The maintenance of a normal or slightly elevated hematocrit level in a patient with a penetrating wound of the abdomen did furnish some evidence that the effects of peritoneal contamination had equaled or exceeded the effects of hemorrhage. So many factors, however, influenced the hematocrit in wounds of the abdomen that its usefulness as a means of quantitative assessment of the loss of blood or of plasma components from the circulation was quite limited.


Aside from any theoretic considerations, increasing experience led the surgeons of the 2d Auxiliary Surgical Group to emphasize the important role of local loss of fluid, particularly blood, in the initiation of the traumatic shock which followed penetrating wounds of the abdomen. Undeniable support for their position was furnished by the proved efficacy of plasma and whole-blood replacement in the resuscitation of these severely wounded men.

Replacement therapy was frequently begun in the battalion aid station with the administration of reconstituted plasma. Whenever severe shock was apparent soon after injury, plasma was usually given, occasionally in excessive amounts. Plasma received under these circumstances is included in the calculation of the average figures for replacement therapy in this series (fig. 25; table 18).

The total amounts of blood and plasma used for resuscitation were proportionate to the severity of the shock exhibited by the casualties. Furthermore, the volume required in each category of shock agreed fairly well with the estimate of blood loss made by the Board for the Study of the Severely Wounded.5 These facts lend support to the contention that the level of the blood pressure is, in spite of its shortcomings, an important criterion in the evaluation of the degree of shock and the extent of blood loss in severely wounded casualties.

Although application of average data to individual casualties would naturally be expected to reveal discrepancies, nevertheless the values calculated as average for this relatively large series may be taken as a general guide to the requirements which had to be met in preoperative resuscitation. Patients suffering from severe shock often had to be given extremely large quantities of plasma and blood. The administration of 5,000 to 6,000 cc. was not unusual. Uncontrollable hemorrhage during operation also called for the

5Medical Department, United States Army, Surgery in World War II. The Physiologic Effects of Wounds. Washington: U. S. Government Printing Office, 1952, pp. 21-74.


FIGURE 25.-Average replacement therapy (whole blood and plasma) in relation to degree of shock in 957 abdominal injuries with perforation of gastrointestinal tract.

use of very large amounts of fluid. One patient in this series, for example, received a total of 8,250 cc. (1,750 cc. of plasma, 6,500 cc. of whole blood), chiefly because intra-abdominal bleeding could not be effectively controlled during the early phase of the operation.

Adequate replacement therapy was not available in World War I, and surgeons quickly recognized the futility of attempting operation in the presence of severe traumatic shock. Had the same lack of replacement therapy prevailed in World War II, the severity of shock would have precluded operation in many of the 261 patients in this series whose systolic blood pressures were 70 mm. Hg or less (table 18). Operation was attempted in the field hospital in every instance, however, after adequate replacement therapy, and 107 (41 percent) survived to be evacuated to the rear, after the usual period of convalescence.

Accurate data are not available concerning deaths before operation in patients in severe shock after perforation of the gastrointestinal tract. Presumably the number was small. During a 4-month period in 1944-45, a single shock team lost only 6 patients (2.1 percent) of the 285 it was attempting to prepare for operation.

During World War I, Maj. J. W. Vaughan, MC, consulting surgeon, 32d Division, recorded that of 256 patients sent from the triage to an operating station for seriously wounded, 41 died in spite of attempts to combat shock


by various methods, leaving only 215 who received operative treatment.6 In marked contrast, under similar circumstances in a field hospital platoon in 1943, there were no preoperative deaths among 297 consecutive patients admitted for surgery.

The saving of life in World War II among casualties in severe and profound shock was largely attributable to liberal, and frequently massive, transfusions. In numerous instances operation would have been impossible without such therapy. The improvement which followed prompt and intensive replacement therapy was often striking, particularly among patients who had suffered very severe shock of relatively short duration. Improvement was not observed uniformly in this category, it is true, but it occurred with sufficient frequency to emphasize the therapeutic potentialities of liberal replacement therapy in the resuscitation of severely wounded men.

Failures of Replacement Therapy

The fact that significant or sustained improvement did not invariably occur after replacement therapy suggested that, in spite of the importance of the fluid loss in the pathogenesis of shock and the unique role of replacement therapy in its management, other factors, which were probably not completely amenable to this type of therapy, had to be taken into account.

One of the most important of these additional factors was massive peritoneal contamination, with early fulminating peritonitis, the deleterious effects of which were often augmented by severe hemorrhage, a prolonged timelag, or both.

The case fatality rate in the various categories of shock in this series was directly related to the degree of shock (table 18). The steadily ascending rate emphasized the increasing gravity of the prognosis as the degree of shock increased. About three-quarters of the deaths in the 3,154 cases in the total series occurred by the end of the second postoperative day. Casualties who died within this period often pursued a relentlessly downhill course characterized by persistent or recurrent shock. It was soon learned that failure of the patient to manifest sustained benefit from adequate resuscitative measures, particularly when replacement therapy had been liberal and intensive, was clear evidence that the prognosis was grave (cases 8, 9, and 12).

The course of such a patient was characteristic. When he was received at the field hospital he was in shock, which often was severe. The blood pressure was extremely low. The pulse was rapid and of poor volume. The extremities were cold. The skin was pale, and if the degree of shock was profound there was often more or less purple-red mottling or cyanosis. Venous filling was apt to be poor, even after a tourniquet had been applied. The rectal temperature was seldom above 100° F. and was frequently subnormal. The patient was aware of his extreme weakness, complained little if at all, and wanted

6The Medical Department of the United States Army in the World War. Washington: U.S. Government Printing Office, 1927, vol. XI, pt. 1, pp. 109-110.


to be left undisturbed. The sensorium was likely to be clouded, but in the absence of acute morphinism true coma was not common except as a terminal manifestation. Appraisal of the history often suggested that the timelag, although perhaps not intolerable in itself, had become intolerable because of the extent and character of the wound. Considerable transient improvement might be attained by liberal transfusion therapy, but the clinical response was likely to be incomplete or poorly sustained, and increasingly severe shock often supervened after operation. In fact, one of the most constant characteristics in this type of case was the tendency to recurrent shock.

The terminal decline as a rule occurred shortly after operation, but in many instances it was sufficiently delayed for widespread bacterial growth to occur. Death was usually attributable to shock or to shock and peritonitis, but the peritonitis, it must be emphasized, differed from the type of purulent peritonitis familiar in civilian practice (p. 198). It was characterized by massive contamination of the peritoneum and by the gross pathologic changes observed during the early stages of bacterial growth. In addition, the circulatory and metabolic changes associated with severe traumatic shock, as well as the overwhelming nature of the peritoneal infection, appeared to be capable of inhibiting to an appreciable degree the local leukocytic response. Autopsy revealed edema and discoloration of the intestine, opacity and dullness of the serosal surfaces, a moderate amount of turbid, serosanguineous exudate, and relatively little fibrin (cases 7, 8, 10, and 11). A purulent exudate of significant amount was not usually seen, probably because of the brief duration of the inflammatory process.

The peritoneal reaction which was the result of early massive soiling thus differed materially from the usual bacterial peritonitis of 4 to 6 days' duration. Early in the war, the deleterious local effects of such soiling were not always appreciated because the surgeon's judgment was based on knowledge limited to the classical variety of peritonitis.

Preexisting traumatic shock was undoubtedly the explanation of the promptness and frequency with which death occurred after wounding in this type of peritonitis. When massive peritoneal contamination was further complicated by continuing intra-abdominal hemorrhage, successful resuscitation and surgery were often impossible.

In some cases, in spite of massive transfusion therapy, careful surgery, chemotherapy, oxygen therapy and all other available measures, the fatal outcome was merely delayed for a day or two. Its apparent inevitability, in spite of what seemed to be adequate replacement therapy, introduced the question of the possible role of toxic factors in the production of so-called irreversible shock. The toxic theory is particularly interesting because of the massive peritoneal contamination and the associated circulatory failure present in many of these severely wounded men. The analogy to anaerobic myositis was suggestive, especially because in a few instances pure cultures of Clostridium welchii were obtained from the peritoneal cavity. Similar infections of the pleural cavity were occasionally observed (case 13). These isolated observa-


tions, however, cannot be construed as valid evidence that the same factors were responsible for severe shock in the presence of peritoneal contamination. The role of toxicity cannot be denied on the basis of the evidence at hand, but a toxic factor, other than infection per se, remains to be isolated and established.

It should be realized that the exigencies of warfare often precluded the accumulation of data which would fully explain the apparent failures of replacement therapy in the management of shock. In the absence of these data, unwarranted conclusions were frequently drawn relative to the effect of what were termed toxic factors. When comparatively complete clinical and post mortem data were critically analyzed, these conclusions usually could not be substantiated. The protocols found at the end of this chapter (cases 1 to 4, inclusive; 9; 10; 14; and 15) disclose various findings which, regardless of their relationship to shock, limited the chance of survival of the seriously wounded. The investigations of the Board for the Study of the Severely Wounded7 afforded no evidence of the existence of toxic agents and led the investigators to suggest that the term "irreversible shock" should be abandoned because of its implications in this respect.

Excessive prolongation of the period of peripheral circulatory failure was apparently the most important reason for failure of liberal replacement therapy to accomplish resuscitation in some cases. The importance of the duration of shock cannot be arrived at by a simple analysis of statistical data (table 18). There was no significant correlation, as already pointed out, between the severity of shock and the duration of injury before operation. Moreover, it was frequently impossible to determine the exact duration of the observed state of shock from the field medical record.

It was readily apparent, however, that the more severe the peripheral circulatory failure, the more rapidly did a state of profound or so-called irreversible shock ensue. A relatively short period of delay in resuscitation might prove fatal to a patient in severe shock, although it could easily be tolerated by a patient in moderate shock. Traumatic shock, satisfactorily alleviated by the prompt transfusion of 1,500 to 2,000 cc. of whole blood, could constitute irreversible shock in the absence of such therapy. It is a rather general custom to label shock as irreversible when the therapy at hand has proved to be unsuccessful, but this is a vague characterization defined by therapeutic trial, not by exact knowledge of the fundamental pathologic physiology.

All the World War II experience indicated that peripheral circulatory failure is followed, within a variable period of time (the variation depending upon its severity), by alterations in general cellular metabolism which cannot be corrected by transfusion therapy alone. These changes were thought to depend upon the interruption of the supply of oxygen and other nutrients to large groups of cells, as, for example, those of the liver and kidney, rather than upon changes in the capillary endothelium. Shock therapy was effective only

7See footnote 5, p. 124.


insofar as it restored adequate circulatory dynamics and alleviated tissue anoxia. Changes in cellular metabolism could thus be prevented or arrested, but they could not be completely reversed. In this respect, replacement therapy was always prophylactic.

If the metabolic changes associated with severe shock were thoroughly understood, the knowledge might serve to resolve the enigma of so-called irreversible shock without invoking factors other than fluid loss in its pathogenesis. Unfortunately, no simple methods were available during World War II-nor are they available now-by which the clinician could determine accurately the onset of irreversible cellular changes during the course of traumatic shock. The only solution of the problem was, therefore, to treat existing circulatory failure as promptly and as vigorously as possible.


The experience of the 2d Auxiliary Surgical Group with casualties in shock and in need of resuscitation following abdominal wounds apparently paralleled the total war experience. Local fluid loss undoubtedly was the most important factor in the initiation of traumatic shock. The most efficacious means of resuscitation was prompt and adequate replacement therapy, in which whole blood was the most important component. The nature of battle wounds of the abdomen demanded that prompt and expert surgery constitute an integral part of resuscitative therapy, in order to arrest hemorrhage and peritoneal contamination with the least possible delay. Many factors, not all of which were completely understood, augmented and perpetuated shock. One, which was extremely important, was massive contamination of body cavities. The circulatory disturbances of severe shock promptly led to bodily changes which could not be completely reversed by replacement therapy. Delay in initiating such therapy, as well as improper utilization of available measures, whether as the result of military exigencies or for other reasons, resulted in failures of resuscitation.

Many abdominal injuries were lethal, at least in relation to methods of resuscitation and surgery then at hand. They remained, at the end of the war, as a challenge for the future. It is in the resuscitation of gravely wounded men that the therapy of traumatic shock will meet its most severe test. Advances will depend upon a more fundamental knowledge concerning the relationship of massive peritoneal contamination to traumatic shock and death, as well as upon a more thorough understanding of the cellular derangements produced by severe shock. The failures in replacement therapy which occurred in World War II point to the need for further study of the problem but must not be interpreted to permit the neglect of available measures of resuscitation in the management of severely wounded men with abdominal injuries.



Introductory Note

In March 1944, the author of this study (Maj. Beverly T. Towery) had the good fortune to assume the direction of Shock Team No. 6, 2d Auxiliary Surgical Group, which was then engaged in the resuscitation of nontransportable casualties in the field hospitals of the Fifth United States Army, Mediterranean Theater of Operations. As explained elsewhere, the evacuation of gravely wounded casualties was always specifically interrupted in a forward medical echelon for the treatment of shock, followed by surgical intervention, after which the patient was retained in the field hospital for a period of convalescence.

At this time, almost no post mortem examinations were being carried out upon the fatal cases in this group of patients, although the high incidence of severe shock and the resulting deaths among these "casualties of immediate urgency" provided a remarkable opportunity for the study of the clinical as well as of the morphologic characteristics of wound shock. The tissues and clinical protocols which the microscopists received were often too inadequate to permit an intelligent interpretation of the findings. Conversely, reports of microscopic diagnoses usually failed to find their way back to the prosectors.

With the enthusiastic encouragement of the late Lt. Col. Tracy B. Mallory, MC, a systematic program was set up for the clinical and post mortem study of patients who died in field hospitals, as follows:

1. The author, with only one or two exceptions, performed the post mortem examinations.

2. He was responsible for providing the microscopist with a record of the gross anatomic findings.

3. With this report, he also included provisional diagnoses and a brief summary of the case, based upon the autopsy findings and the clinical data available at the time of the post mortem examination.

4. The microscopists provided histologic descriptions of the tissues, added microscopic diagnoses, and frequently wrote brief comments upon their observations.

5. The original observations made by the prosector were retained in the protocols, and the provisional diagnoses were not amended to conform to the subsequent reports of microscopic findings. Such inconsistencies as might have arisen were discussed and rectified in the final summarizing statements attached to the end of each protocol.

During the 13-month period in which this program was continued, it was possible to accumulate relatively complete clinical and post mortem data upon 47 men who had been wounded under varying circumstances of terrain, climate, and tactical situation in Italy, France, and Germany. Thirty of these casualties (64 percent) had sustained wounds of the abdomen. The 15 protocols which follow have been selected from this group of cases, in an attempt to define


clearly the magnitude of the problem of shock among patients with abdominal injuries. It is believed that these records constitute representative source material.

The author was personally responsible for part or all of the resuscitation in most of the cases in the total series. As collator, however, he has drawn freely upon the clinical observations of numerous colleagues in respect to replacement therapy, anesthesia, surgical management, and postoperative care.

This study clearly showed that hemorrhage and peritoneal contamination were the most important determinants of the outcome of intra-abdominal trauma. Regardless of the ultimate detrimental effects of peritoneal contamination, the early onset of severe shock was almost invariably due to brisk hemorrhage. Widespread peritonitis unquestionably increased the likelihood of a fatal outcome, though, aside from the attendant local loss of fluid, the mechanism of the deleterious effects of this complication remained unexplained in this study. A morphologic evaluation of these phenomena proved to be a difficult undertaking and one which, on theoretical grounds, was not particularly rewarding. From a practical standpoint, however, the demonstration of a microscopic lesion, such as diffuse pulmonary fat embolism, sometimes provided crucial data in a fatality which otherwise might have remained obscure.

Resuscitation by means of vigorous replacement therapy removed, at least in part, the immediate risks of shock but seemed to accentuate certain microscopic changes contingent upon partial or complete recovery from it. Hemoglobinuric nephropathy was perhaps the most striking of these lesions, and the importance of post traumatic renal failure will be readily apparent on a perusal of these records. It is ironical that the development of this particular lesion seemed to depend upon a measure of success in resuscitation.

One of the major aims of this study was to obtain evidence concerning the fundamental pathogenesis of irreversible shock. The results were illuminating. In effect, it was evident that a careful post mortem examination very often revealed findings which provided some other logical explanation for the fatality. Among these findings were unrecognized perforations of the gastrointestinal tract, the autotransfusion of contaminated blood, unsuspected infection, pulmonary edema, and cerebral emboli. On the basis of the experience obtained in this investigation, one would therefore be inclined to minimize the importance of shock which does not have a reasonable explanation in terms of hemorrhage, exudation, and infection.


Case 1


This patient received a bullet wound of the left hip on 2 June 1944, at 1000 hours. When he reached a field hospital at about 1600 hours, after having

8The author gratefully acknowledges the generous assistance in this study of the various pathologists whose microscopic studies (sometimes identified by initials) appear in the protocols. He is particularly indebted to the late Lt. Col. Tracy B. Mallory, MC, and to 1st Lt. Leslie S. Jolliffe, MC, and Capt. Joseph G. Rothenberg, MC. First Lt. Kathryn T. Driscoll, ANC, provided invaluable clerical assistance.


received 250 cc. of plasma, his blood pressure was 110/60 mm. Hg. He did not appear to be in shock, but examination of the abdomen revealed generalized rigidity. He was given 500 cc. of plasma and 500 cc. of blood before operation, and his condition seemed relatively good. His blood belonged to group A.

At operation, much blood was found in the peritoneal cavity. The missile had perforated the abdomen by way of the left buttock. When exploration was attempted, profuse hemorrhage occurred from the left iliac vessels, and the aorta and inferior vena cava were temporarily ligated with tape. After a great deal of difficulty, it was possible to ligate the left external iliac vein, and the internal iliac vein and artery.

The patient's condition was extremely poor during this procedure, and he was given 4,500 cc. of group O blood in an attempt to replace the blood loss. Extensive exploration of the abdomen was not carried out, but there was no evidence of perforation of the intestine or of peritonitis. At the end of the procedure, which lasted approximately 7 hours, the blood pressure was 100/40 mm. Hg; it had risen considerably after bleeding had been controlled.

The right leg became moderately cyanotic after operation, and, on both the first and second postoperative days, right lumbar sympathetic blocks were performed with procaine. Some improvement was noted. On the second postoperative day, the temperature was 98.6° F., the pulse was 100, and the respiration was 24. When the patient was catheterized, only 60 cc. of dark-brown urine was obtained, although the fluid intake had been adequate. Its specific gravity was 1.025. The patient became somewhat confused and died at 2000 hours 4 June 1944, 58 hours after wounding.


Autopsy was performed 5 June 1944, 12 hours after death (Cori, Italy). 

Gross observations.-The delay in performing the autopsy, without benefit of refrigeration, had produced marked changes in the body. The pericardial cavity contained 30 cc. of pink serous fluid, and 120 cc. of similar hemolyzed blood-stained fluid was present in each of the pleural cavities. The peritoneal cavity was not remarkable, except for generalized pink (hemoglobin) staining of the serous surfaces.

The lungs showed bilateral and symmetrical edema and deep purple-red discoloration, with considerable reduction in crepitus. No bronchial obstruction was present nor were any emboli present in the carefully dissected branches of the pulmonary artery.

Two perforations were observed in the retroperitoneal portion of the ascending colon. They were caused by passage of the bullet through this viscus about 8 cm. distal to the ileocecal valve. There was also extensive retroperitoneal hemorrhage in this region, with a rather scanty fibrinopurulent exudate.

The kidneys were swollen and moderately increased in size. The cut surface bulged above the edge of the capsule. The cortex was rather pale, but


its pallor was modified by hemoglobin staining of all the tissues. Hydroureter and renal-vascular occlusion were not present.

The finer detail of the pelvic vessels was greatly obscured by extensive suffusion of tissues by markedly hemolyzed blood.

Sections of muscle taken from the upper thigh on the left were not grossly remarkable except for moderate pallor.

Gross anatomic diagnosis-(1) Penetrating gunshot wound of left buttock, with two perforations of retroperitoneal portion of ascending colon and early retroperitoneal abscess; lacerations of left common iliac vein, external iliac vein, and internal iliac artery and vein, all secondarily ligated; and (2) severe, exsanguinating hemorrhage, secondary to these lacerations. 

Comment.-Death may be ascribed to shock, secondary to severe and prolonged hemorrhage. Since death occurred on the second postoperative day, the perforations of the colon may have had relatively little to do with the fatality. These perforations, however, served to contaminate the wound track with anaerobic organisms, and clinical manifestations would probably have supervened shortly, had the patient survived. How important this contamination was in causing death must remain conjectural. The oliguria present before death suggested a renal lesion, but in all probability the inadequate renal blood flow (hypotension) during the postoperative period was alone responsible for the decreased urinary output.

Microscopic observations-Histologic examination of the kidney showed the lower nephron segments to be extensively plugged with pigmented casts. In the cortex, some tubules contained masses of intensely blue-staining granules and spherules suggesting calcareous material. These masses were not found in the pyramid. There were many bacilli in the interstitial tissues, and fixation was too poor to judge epithelial degeneration.

Examination of the bowel showed much of the mucosa to be necrotic. All layers of the wall were infiltrated with leukocytes, and the serous surface was covered by a thick fibrinopurulent membrane.

Comment-The severe lower nephron nephrosis of the hemoglobinuric type is suggestive of a transfusion reaction. The story of the administration of 4,500 cc. of group O blood to a group A recipient within a period of a few hours is interesting and may have been the responsible factor. The clostridial infection was evidently only agonal and cannot be held responsible for the fatality. There was no massive necrosis of voluntary muscle which could have produced myohemoglobinuria. The calcareous masses in some of the tubules are assumed to represent past history; they could be remnants of a previous sulfonamide reaction. (T. B. M.)


Death probably was due to a combination of factors in this case: (1) Severe shock during a prolonged operation; (2) perforation of the colon with contamination of the retroperitoneal tissues; and (3) renal failure, with consequent


pulmonary edema. Of these factors, the first seems to be of greatest importance, since death occurred within a relatively short time after operation. Uremia per se was probably not of primary importance in causing death; under the conditions at hand it is extremely difficult to assess the importance of infection in this case.

Case 2


This patient received shell-fragment wounds of the left thigh and right costal margin 19 March 1945 at 0230 hours. A tourniquet was applied to the left leg at 0300 hours. When the patient reached a field hospital at 0630 hours, after having received 500 cc. of plasma, the blood pressure was 110/70 mm. Hg, and the pulse was 80. Before operation, he was given 600 cc. of plasma and 500 cc. of whole blood. His blood belonged to group AB.

Anesthesia was begun at 1105 hours 19 March. Laparotomy revealed a laceration of the right lobe of the liver and severe destruction of a 10-cm. portion of the transverse colon. A spur colostomy was made through the right upper quadrant, and drains from the hepatic laceration were brought out through the wound at the right costal margin. The operation required 3½ hours, and during it the patient received 2,000 cc. of whole blood.

In addition to his abdominal wound, this patient had a fracture of the lower third of the left femur, with severe comminution and posterior angulation of the distal fragment.

The day after operation, the patient was semistuporous. The blood pressure was 84/50 mm. Hg and the pulse 140. He received 500 cc. of plasma. Medium rales were heard in both lungs. Moderate cyanosis was relieved by the administration of oxygen. There was profuse drainage of bile-stained fluid from the drains at the right costal margin. During the first 12 hours after operation, the patient excreted 275 cc. of urine. His respirations were quite shallow, and he coughed only occasionally and ineffectually. He died at 1630 hours on the second postoperative day, 62 hours after wounding.


Autopsy was performed 4 hours after death (St. Avold, France).

Gross observations-The abdomen was moderately distended, in spite of the presence of several drains. The peritoneal cavity contained about 300 cc. of blood-tinged, purulent fluid, which was present chiefly above and posterior to the colostomy spur; a moderate amount of yellow-gray purulent exudate was also found in this area. There was considerable dilatation of the stomach and of the upper bowel down to the point at which the small bowel was partially obstructed as it lay posteriorly and lateral to the proximal segment of the colostomy. Distal to this point, the bowel was not distended, though the loops


were lightly bound together by fibrinopurulent exudate. There was no leakage into the peritoneal cavity from the colostomy, the transverse spur of which appeared to be under slight tension.

The right auricle was considerably distended by blood. The tricuspid valve appeared slightly dilated. There were petechial hemorrhages over the surface of the posterior aspect of the heart, particularly along the coronary sulcus. On section, the myocardium appeared paler than normal.

There were severe partial atelectasis of both lungs; the aeration of all lobes except the left upper lobe was moderately to markedly decreased. There was well-established nodular consolidation of the entire right lower lobe, the hilar portion of the right middle lobe, and the posterior portion of the right upper lobe. Near the hilus of the right lower lobe there appeared to be beginning abscess formation.

The spleen was of usual size and was covered by a thin layer of fibrinopurulent exudate. On section, it was purple red and moderately firm, but the pulp scraped away readily.

There was a small laceration of the lateral margin of the right lobe of the liver, but no appreciable collection of bile was present within the peritoneal cavity. In the dome of the right lobe was a large, irregular area of yellow mottling. There was no evidence of suppuration in the tissue adjacent to the laceration.

The kidneys were somewhat smaller than normal, and the capsular surfaces were smooth and pale pink, with small irregular patterns of vascular congestion. The cut surface was not remarkable. The adrenals were normal.

Gross anatomic diagnosis-(1) Perforating shell-fragment wound of abdomen, via upper right flank, with exit just to right of umbilicus, with moderately severe laceration of right lobe of liver, and severe perforation and laceration of right transverse colon, with spur colostomy; (2) acute, moderately severe, fibrinopurulent peritonitis, secondary to perforation of transverse colon; (3) severe pulmonary atelectasis, involving all lobes, with early acute bronchopneumonia of right lung and left upper lobe, and possible abscess of right lower lobe; dilatation of right auricle and subepicardial petechial hemorrhages probably due to atelectasis and asphyxia (possibly terminal);  (4) perforating shell-fragment wound of left lower thigh, with complete, compound, comminuted fracture of lower third of left femur and laceration of capsule of left knee joint and of suprapatellar bursa; and (5) lacerating shell-fragment wound of left chest wall near 12th rib.

Comment-The immediate cause of death in this case appears to have been atelectasis and bronchopneumonia. The post mortem findings suggest that these complications began soon after operation. The semistuporous state and the diffuse peritonitis and distention of the stomach and upper small bowel were probably responsible for the poor respiratory exchange and ineffectual cough. In the presence of severe bone trauma, stupor and pulmonary rales suggest the possibility of fat embolism.


Microscopic observations.-In the alveoli of the lungs were large amounts of eosinophilic coagulum and varying degrees of polymorphonuclear infiltration into the coagulum. A second section revealed dissolution of the alveolar walls with extreme infiltration and accumulations of polymorphonuclear leukocytes. In the central portions of these accumulations were bacterial groups and necrotic debris. Some alveolar walls which were still present appeared necrotic. Thrombi were seen in the vessels in these areas. On the pleural surface was a thin, fibrinoid material with enmeshed polymorphonuclear leukocytes. Large amounts of eosinophilic coagulum were seen in alveoli along with polymorphonuclear leukocytes.

In the section studied from the liver, small vacuoles were seen throughout the parenchyma. About the central vein, which was slightly dilated, slight sinusoidal congestion was observed.

The sinusoids of the spleen were dilated and congested.

The tubular cytoplasm of the kidney was granular. Many of the nuclei were indistinct or absent. The glomeruli were not remarkable.

Vacuolation of cortical cells was most noticeable in the zona glomerulosa of the adrenal glands.

Microscopic diagnosis-(1) Lobular pneumonia with abscess formation, (2) moderate fibrinopurulent pleuritis, (3) severe acute purulent peritonitis, (4) edema of small intestine, and (5) traumatic necrosis of liver.


This case is an instance of severe abdominal injury. After a relatively short delay, the patient reached the field hospital in surprisingly good condition. In spite of the many favorable circumstances, however, he died of pulmonary atelectasis and bronchopneumonia on the second postoperative day. Peritonitis was probably also a contributory factor in the fatality. It is unfortunate that microscopic studies of the lung were not sufficient to exclude pulmonary fat embolism.

Case 3


This patient received a bullet wound of the left abdomen 18 April 1945 at 1800 hours. When he was admitted to a field hospital at 2215 hours, his blood pressure was 130/80 mm. Hg. He received 500 cc. of group A (his blood group) blood before operation. His condition during the preoperative period was relatively good.

As the peritoneal cavity was being opened, the blood pressure fell abruptly to 0/0 mm. Hg. The systolic pressure rose to 90 mm. Hg when 1,500 cc. of whole blood were given rapidly. As abdominal exploration was continued, a large quantity of blood was found in the peritoneal cavity. There was a severe laceration of the proximal jejunum; a small segment was resected and an end-to-


end anastomosis was performed. A lacerated segment of the colon was exteriorized through a stab wound in the left upper abdomen. Operation was completed 19 April at 0220 hours.

Immediately after operation, the systolic blood pressure was 170 mm. Hg, which was attributed to a mild transfusion reaction during the operation. The blood pressure then progressively declined; it was 100/66 mm. Hg at 0830 hours 19 April and only 22/0 mm. Hg at 1200 hours. At this time, the patient appeared moribund. He had vomited large quantities of blood, and it was thought that the fall in blood pressure was caused by bleeding in the upper gastrointestinal tract.

In an attempt to control this bleeding, a second operation was undertaken at 1400 hours 19 April. At this operation, no blood was found in the peritoneal cavity. The original jejunal anastomosis was examined through an opening made in the bowel just distal to it. At this time, although he had had 3,000 cc. of blood just before the operation and during its course, the patient was in severe shock. No point of bleeding could be found along the line of resection. The anastomosis was, nevertheless, reenforced by an additional row of through-and-through sutures, and the jejunotomy was closed. A gastrotomy was then performed, and a large volume of clotted blood was removed from the stomach. There was no evidence of gastric bleeding from the mucosal surface, which was examined as completely as possible through the incision.

After the second operation, the blood pressure was 80/0 mm. Hg. By 1830 hours, it had fallen to 20/0 mm. Hg, and it did not again exceed this level. The patient expired in profound shock at 2220 hours 19 April 1945, 28 hours after wounding.

During his brief hospitalization, this patient received 5,500 cc. of whole blood, all group A and all carefully crossmatched with his blood.


Autopsy was performed 13 hours after death (Poxdorf, Germany).

Gross observations-The body was not remarkable except for considerable dependent lividity.

The lungs showed only post mortem hypostasis.

The abdominal cavity was not remarkable except for the thin, opaque, fibrinous exudate which covered several loops of small bowel. This portion of the bowel was not distended, and there was no appreciable collection of purulent exudate.

The spleen was moderately enlarged and rather soft and appeared somewhat less congested than is usual after massive transfusion therapy.

The gastrotomy, jejunotomy, and jejunal anastomosis 8 cm. distal to the ligament of Treitz were all snugly sutured, and there was no evidence of leakage of bowel contents. When the gastrointestinal tract was opened, about 300 cc. of old blood were found in the stomach, and a small amount of similar blood was also found in the loop of jejunum at the site of anastomosis. The duo-


denum, which was free of blood, was deeply bile stained. In an attempt to locate a possible source of bleeding, the mesenteric vessels leading to the jejunojejunostomy were cannulated and infused with water, but no vascular leak was disclosed. Since no blood was present distal to the jejunostomy, the upper gastrointestinal tract was examined minutely. No source of hemorrhage was present. A small, shallow excoriation of the distal esophagus and another in the first portion of the duodenum showed no evidence of tissue reaction, and no vessels were exposed. No varices were present in the lower esophagus. 

The kidneys were somewhat smaller than usual but were not remarkable on section. The left kidney and adrenal were partly surrounded by the blood in the retroperitoneal area.

Gross anatomic diagnosis-(1) Penetrating bullet wound of abdomen with severe laceration of proximal jejunum, secondary resection and jejunostomy, jejunotomy; perforation of transverse colon, colostomy; and (2) exsanguinating postoperative hemorrhage from upper jejunum secondary to lacerating wound of jejunum, with the operations described.

Comment-A very careful examination of the upper gastrointestinal tract failed to reveal any source of hemorrhage. The most likely sequence of events, therefore, is that the patient bled extensively from the initial jejunojejunostomy. It may be that no bleeding was found at the second operation because of the very low blood pressure at the time. The failure of response to transfusion therapy before and during the second operation is probably related to the considerable time during which the blood pressure remained at a very low level. From the evidence obtained at autopsy, the second row of sutures about the jejunum was apparently successful in preventing further hemorrhage, but fatal hemorrhage had already occurred.

The histologic picture of the kidneys will be of interest in this case, in view of the massive therapy with type-specific blood. During the 10 hours the patient was on the ward between operations, there was no recorded output of urine. During most of this time, the blood pressure was below the level necessary for renal filtration.

Microscopic observations-The sections from the lungs showed confluent patches of bronchopneumonia. The exudate was hemorrhagic in a few foci but chiefly showed polymorphonuclear leukocytes. In another section, scattered, small areas of atelectasis were seen. Two small vessels were thrombosed. The alveolar septa were hyperemic.

Sections from the liver showed the centers of the lobules to be congested. The liver cells in these regions contained granular brown pigment (lipochrome). 

The pulp of the spleen was moderately congested. The malpighian bodies were large and conspicuous, with active secondary follicles, in some of which much free and phagocytized nuclear debris was present.

The only change noted in a section from the gastroesophageal junction was autolysis of the gastric mucosa.

The epithelium lining some of the collecting tubules and Henle's loops in


the kidney showed degeneration and desquamation. With this exception, both glomeruli and tubules appeared normal.

Comment-The degenerative changes in the kidney in this case were not pronounced and may be entirely post mortem changes. One would be more inclined, however, to attribute them to the prolonged, severe shock which this patient sustained.

The fact that this patient received 5,500 cc. of blood without histologic evidence of incompatibility is a tribute to those responsible for the transfusions. In similar cases, the experience has been quite the contrary, especially when group O blood has been given to a group A patient. (J. G. R.)


In this case, postoperative hemorrhage from a jejunal anastomosis was responsible for prolonged, severe shock and eventual death. Microscopic examination of the kidneys failed to reveal the presence of pigment nephropathy, which might have been expected because of its frequency under similar circumstances. The following factors may offer a partial explanation for the absence of such a lesion: (1) The use of group-specific blood instead of the customary low-titer group O blood in the resuscitation of a group A patient, and (2) the failure of the blood pressure to reach the glomerular filtration pressure throughout the prolonged period of shock. Presumably, pigment casts do not appear unless some degree of glomerular filtration is resumed or is maintained after the initial insult.

Case 4


This patient received an accidental bullet wound on 4 May 1945, at 0945 hours. The wound of entrance was located at the right costal margin in the midclavicular line, and the wound of exit was just lateral to the inferior tip of the left scapula. When he was received in a clearing station immediately after injury, the blood pressure was 120/60 mm. Hg, and the pulse 130. He was given 125 cc. of plasma, and the 5th to 12th intercostal nerves on the right side were blocked with procaine. The patient was then moved forward with the field hospital which was in convoy at the time. Upon his arrival at the new installation, the blood pressure was 90/50 mm. Hg, and the pulse was 140.

A right thoracentesis was not productive, but 635 cc. of blood were removed from the left chest. The patient complained of extreme pain in the abdomen, right chest, and both shoulders. The response to the administration of 750 cc. of plasma and 635 cc. of blood by autotransfusion was poor.

Epidural block with procaine was performed, with good results, but severe bilateral pains in the shoulders persisted. When the patient was turned on his side before this procedure, there was a transitory fall in the blood pressure, but the block was accomplished without difficulty. Just before operation, the blood pressure was 84/50 mm. Hg. The patient's blood belonged to group B.


Gas-oxygen-ether endotracheal anesthesia was begun at 1815 hours. The wound of exit was debrided, and a thoracotomy was performed through the left ninth intracostal space, with resection of a fragment of the ninth rib. Splenectomy and suture of a laceration of the greater curvature of the stomach were carried out through a transdiaphragmatic incision. Laparotomy was then performed, with closure of a perforation in the upper portion of the lesser curvature of the stomach. The missile track had traversed the left lobe of the liver.

Just before thoracotomy, when the patient was turned on his right side, there was a marked fall of blood pressure. Within a short time, he was given 1,500 cc. of whole blood and 250 cc. of plasma. The agglutinin titers of the blood which he received at this time were not known; no blood with known agglutinin titer was available.

On 5 May, the day after wounding, the patient's condition was poor, because of atelectasis of the left lung. After tracheal aspiration, there was an increased density of breath sounds on the left side and a lessening of cyanosis. The abdomen was flat. There was considerable drainage of bile. Approximately 600 cc. of bloody fluid drained from the left chest. The hematocrit was 37 and the hemoglobin 13 gm. percent; the plasma protein was 5.8 gm. percent. The systolic blood pressure was 85-90 mm. Hg. Intravenous fluids consisted of 500 cc. of plasma and 500 cc. of blood. The urine output since operation was 700 cc. The serum showed no evidence of hemolysis.

On 6 May, the patient was somewhat improved after tracheal aspiration, with removal of dark, blood-tinged mucus. Aeration of the left chest was better, but a beginning pneumonitis was suspected. The temperature was 101°-102° F. by rectum. The urinary output was 200 cc.; albumin 1 plus was present. The blood pressure was 100/60. The patient was irrational and required sedation, for which paraldehyde was used. He perspired profusely, and venous distention was pronounced. All fluids were discontinued.

On 7 May, the left chest was clearing. Bronchial breath sounds were heard over the right base. Oxygen was administered by mask.

At 1300 hours, the patient was moved to the X-ray table. Here he rapidly became cyanotic, in spite of continued oxygen administration, and failed to respond after endotracheal suction. The superficial veins were greatly dis tended. He died 7 May 1945 at 1400 hours, about 76 hours after wounding.


Autopsy was performed 1 hour and 15 minutes after death (Adelstetten, Germany).

Gross observations-There was considerable evidence of dehydration. The veins of the arms and neck were quite distended. In addition to the healing surgical incisions, the wound of entry lay just inferior to the right costal margin, and the wound of exit was at the level of the ninth rib just lateral to the tip of the left scapula.

The peritoneal cavity contained a small amount of clotted blood. The surfaces were smooth and glistening and there was no general fibrinopurulent


exudate. Along the drainage tracks to the posteroinferior aspect of the liver and the anterior portion of the stomach was a moderate deposit of yellow, fibrinopurulent exudate. There was no fluid or bile in the subhepatic space. The diaphragm was at the level of the fifth rib bilaterally.

The right chest contained about 75 cc. of clotted blood. A moderate amount of air escaped from the left chest when the cavity was opened and pressure was applied to the sternum. The chest on this side contained 400 cc. of bloody, turbid fluid, which was enmeshed in fibrinous strands and which compressed the lung more or less uniformly. Anteriorly, a fibrinopurulent-lined pocket was present, with a capacity of approximately 250 cc.; it appeared to have contained air. The left diaphragm was sutured without leakage.

The pericardial cavity contained about 40 cc. of clear, straw-colored fluid. The heart appeared considerably distended, especially on the right. The right auricle was distended by blood, which brought the trabeculae into sharp relief. The tricuspid valve measured 13.0 cm. in circumference. There were scattered subepicardial petechiae, chiefly over the posterior cardiac aspect along the coronary sulcus.

The left lung was moderately reduced in size and air content. The surface was covered by a shaggy, yellow-pink, fibrinopurulent exudate in which were embedded foreign particles with the appearance of food. Both lungs were greatly congested and edematous, particularly in the dependent portions. Edema was more severe on the right, and the right lung was somewhat heavier than the left. Both, however, were considerably increased in weight. Section of the right lung revealed it to be practically airless. Dark, serosanguineous fluid oozed freely from the cut surface. The right lower lobe contained no areas of consolidation, but small areas of early consolidation were present in the subpleural portion of the right upper lobe. There was no consolidation of the left lung, but all bronchi contained a copious, blood-tinged, serous fluid streaked with purulent exudate. No demonstrable isolated bronchial obstruction existed; rather, obstruction was generalized. There was moderately severe contusion of, and hemorrhage into, the parenchyma of the inferior portion of the left lower lobe.

The liver was not generally remarkable. The missile had entered the left lobe at the falciform ligament anteriorly and had emerged near the lesser curvature of the stomach. There was the usual degree of infarct necrosis along the missile track. There was no gross evidence of suppuration.

There were two sutured perforations of the fundus of the stomach; no leakage was noted along either suture line.

The kidneys, which were similar in appearance, were both considerably swollen, particularly in the anteroposterior diameter. The smooth cortical surface was pale yellow gray and faintly mottled. On section, the bulging cortex appeared pale and opaque. The pyramidal striations were moderately increased. The renal pelves and vessels were negative.

Gross anatomic diagnosis-(1) Bullet wound of left thoracoabdominal area, with moderately severe perforating wound of left lobe of liver; severe


perforating wound of stomach with acute fibrinopurulent peritonitis localized to subhepatic region; laceration of spleen, splenectomy; healing perforation of left posterolateral diaphragm; severe contusion of lower lobe of left lung; complete compound fracture of left ninth rib; (2) severe left fibrinopurulent pleuritis, secondary to perforating wound of stomach and perforation of left diaphragm; (3) moderate partial left pneumothorax secondary to contusion of lower lobe of left lung and complete compound fracture of left ninth rib; (4) possible hemoglobinuric nephropathy or shock kidney, as manifested by severe bilateral pulmonary congestion and edema; acute congestive heart failure, predominantly right sided; clinical oliguria; and (5) acute purulent early bronchopneumonia of upper lobe of right lung.

Comment-No group O blood in which the agglutinin titer had been determined was available for this patient. The only blood which could be used did not fall into the category of low-titer blood. Under such circumstances, autotransfusion would ordinarily have been particularly desirable, but in this case, because of the injury of the stomach with soiling of the left pleura, it was incorrect.

This patient appeared to have been extremely sensitive to changes in position. On two separate occasions, the blood pressure fell sharply when he was turned on his side. Before thoracotomy, the blood pressure remained at excessively low levels for a period of about 80 minutes, in spite of liberal transfusion therapy. During such periods, the renal blood flow must be markedly reduced, and it is suggested that this was a crucial factor in the production of pigment nephropathy in this case.

The delay before operation was unusually long. Part of it was caused by the fact that the field hospital was in process of moving when the patient reached the clearing station. During the delay, his condition deteriorated considerably, and liberal replacement therapy was necessary before operation. Unfortunately, however, operation was not urged at the earliest possible moment because of the mistaken belief that there had been no perforation of a hollow viscus. Obviously, the optimum time of operation would have been when the patient first reached the clearing station, only a short time after injury.

Microscopic observations-In one section of the lungs, the tissue was packed with red blood cells. Several vessels contained thrombi. In another section, the congestion was almost as severe, and the bronchiolar walls contained a purulent exudate. In a third section, which was free from congestion, the alveolar walls showed patchy approximation and stretching. There was a fibrinopurulent exudate on the pleura.

Rare round cells were found beneath the epicardium. No myocardial changes were noted.

In one section of the liver was a large zone of infarct necrosis, bordered by a region of inflammatory exudate, beyond which was normal-appearing liver tissue. The peritoneal surface showed a thick, purulent exudate. Both in the infarcted area and in the exudate were masses of bacteria.


Sections from the kidneys showed amorphous material and lightly staining spherical bodies in capsular spaces and proximal convoluted tubules, which were moderately dilated. Brown, granular casts were present in the distal tubules and loops of Henle. Nuclear pyknosis and cell shrinkage were noted in relation to these casts. There was moderate swelling of the capsular and convoluted tubular epithelium. A few cysts filled with pink-staining coagulum were present.

One zone of cortical necrosis was noted in the adrenal glands. Otherwise, there was a decrease in the usual vacuolar appearance of the cortical cells. Adjacent fat tissue showed regions in which cells had lost their vacuolated structure.

Microscopic diagnosis-(1) Focal necrosis of the adrenal gland; and (2) moderately severe, early hemoglobinuric nephrosis.

Comment-The microscopic findings are in general agreement with the gross findings. Although the kidney lesion was classified as hemoglobinuric nephrosis, death appears to be related to trauma, peritoneal infection, and pulmonary circulatory obstruction rather than to uremia.


This patient affords an example of a severe thoracoabdominal wound in which the military exigency afforded an opportunity to observe the steady deterioration which can be caused in a wounded man's condition by the delay in operation. Resuscitation and operation were made much more difficult in this case by the passage of a relatively short period of time. Failure to realize that a wound of the stomach was present was partly responsible for the delay and was also responsible for the use of autotransfusion in the presence of contamination of the pleura by gastric contents.

Microscopic examination confirmed the presence of pigment nephropathy, which accounts for the oliguria and gross anatomic changes in the kidneys. Pulmonary edema and mild congestive heart failure were again noted as a terminal manifestation of the renal failure which follows severe traumatic shock.

Case 5


This patient received a bullet wound of the left buttock 1 June 1944, at 0300 hours. He was given 15 mg. of morphine tartrate at 0305 hours and was brought to the collecting station at 0600 hours. Here his wound was re-dressed, in preparation for evacuation to the rear. The record indicates that he seemed in relatively good condition, but at 0630 hours, while he was urinating, he lapsed into deep shock and stopped breathing. After the use of artificial respiration and the administration of 1,750 cc. of plasma, supplemented by two large doses of caffeine sodium benzoate, breathing was resumed. At this time (0815 hours), it was noted that the left pupil was dilated and that there was


weakness of the face on the left side, with equivocal aphasia. At 0900 hours, improvement was continuing, and at 0930 hours the patient was sent to the rear. He reached the clearing station at 1000 hours and was taken at once to the field hospital. At 1005 hours, his blood pressure could not be obtained by auscultation or palpation. In spite of a plasma transfusion and oxygen inhalations, he continued poorly oriented. Death occurred at 1200 hours. The left pupil was notably larger than the right.

This patient had received a total of 500 cc. of plasma in the field hospital. He expired just as whole blood was obtained. His blood pressure was not obtainable at any time after he reached the field hospital.


Autopsy was performed 2 hours after death (Cori, Italy).

Gross observations-Externally the body was not remarkable except for extreme pallor.

The brain was negative, except that the cerebral vessels were exceedingly pale. There was no gross area of infarction. The lungs were negative, except for some patchy atelectasis.

The peritoneal cavity contained an enormous quantity of blood, estimated at 4,000 to 5,000 cc. The source was the left hypogastric artery, which was transected at the point at which it crosses the brim of the pelvis. There was a large hematoma in the areolar tissue of the retroperitoneal space and along the course of the missile, which was through the medial aspect of the ramus of the ischium. A second fragment of the bullet lay free in the left upper quadrant of the abdomen. The mesentery of the sigmoid colon and a loop of the mesentery of the small bowel were perforated, and there were two perforations of the jejunum. It was impossible to judge the amount of fecal contamination of the peritoneum with any accuracy because of the large amount of blood present in the peritoneal cavity. Contamination, however, was not thought to be great.

Gross anatomic diagnosis-Bullet wound of abdomen, entering by way of left buttock, with laceration of left hypogastric artery and exsanguinating intraperitoneal hemorrhage; two perforations of jejunum; compound comminuted fracture of ramus of left ischium.

Comment-There is no reasonable doubt in this case that death was caused by massive intraperitoneal hemorrhage. The cerebral changes were due to the severe degree of cerebral anoxia. It is regrettable that whole blood was not available at the time this man reached the field hospital, although its liberal use might not have changed the outcome in the face of longstanding and profound shock.

Microscopic observations-Examination of sections from the lung showed extensive alveolar atelectasis with overdistention of the respiratory bronchioles. This finding is consistent with a story of prolonged artificial respiration. A


few fat droplets were found in small arteries and arterioles, but the number was not sufficient to make them of functional significance.

The choroid plexus appeared edematous. The kidney and brain were negative.


This case illustrates several important points in connection with shock among battle casualties.

1. It may be exceedingly difficult to detect the fact that severe circulatory collapse is about to occur, particularly when hemorrhage is in progress.

2. Once the blood volume has reached a critical level, a relatively minor happening, in this instance the act of sitting up to urinate, may be followed immediately by the onset of profound shock.

3. It is almost impossible to treat shock caused by severe and progressive hemorrhage with plasma alone.

4. Laparotomy to control hemorrhage offers the only hope of success in such a case as this, and yet, as in this case, it may be impossible to prepare the patient adequately for operation in the face of severe bleeding.

Case 6


This patient, who was 22 years of age, received a shell-fragment wound of the left buttock 3 October 1944, at 2230 hours. After a delay of unknown duration, he was brought to a field hospital in profound shock. Even after he had received 250 cc. of plasma and 2,500 cc. of group O blood (it is believed that his blood belonged to group AB), the blood pressure was only 98/60 mm. Hg and the pulse was 130, but operation was begun without further delay because intra-abdominal hemorrhage was suspected. The patient received 50,000 units of penicillin intramuscularly, but no sulfonamides were given in the field hospital.

Exploration of the abdomen revealed multiple perforations of the small and large bowel. In the peritoneal cavity, there was an estimated 1,000 to 1,500 cc. of blood which was found to have come from the region of the bladder on the left side posteriorly. The retroperitoneal space was opened, and, after considerable difficulty, the bleeding was brought under control by the ligation of two branches of the left internal iliac vein. The perforations of the small bowel were sutured, and a sigmoid colostomy was performed through the left lower quadrant of the abdomen. The gluteal wound was debrided.

The patient was in a critical state throughout the operation. In spite of the administration of 4,000 cc. of group O blood and 750 cc. of plasma, his blood pressure was often as low as 40/0 mm. Hg, and at the end of the operation, which took 4 hours, the blood pressure was only 90/48 mm. Hg.

On the first postoperative day, the blood pressure was 100/50 mm. Hg. The urine was deep red brown and contained 2 to 3 red blood cells per high-power field.


The second day after operation, the blood pressure was 120/70 mm. Hg. The patient was moderately dyspneic. Examination of the abdomen revealed no evidence of peritonitis. Catheterization produced only 10 cc. of urine. Isotonic glucose solution and 500 cc. of a 2-percent solution of sodium bicarbonate were given intravenously.

On the third postoperative day, dyspnea was severe. The patient was drowsy, and his speech was slurred. The blood pressure was 180/100 mm. Hg. Moist rales were first noted throughout both lungs about 2½ hours before death. Dyspnea rapidly worsened, and death occurred at 1830 hours 5 October 1944, 44 hours after wounding. The total output of urine from the beginning of operation until death was about 210 cc.


Autopsy was performed 12 hours after death (Elayes, France).

Gross observations-The peritoneal cavity contained 200 cc. of bloody fluid. There was no exudate other than the generalized matting of loops of small bowel by fibrin. In the pelvis was a small amount of clotted blood. The lungs were indurated and edematous. Only the kidneys were saved for microscopic study. They were somewhat swollen and appeared to be congested. Small subcapsular hemorrhages were scattered over both renal surfaces, but on section it was found that they did not extend into the cortex. The cut surfaces of the kidneys were rather pale.

Gross anatomic diagnosis-(1) Perforating shell-fragment wound of abdomen with multiple perforations of ileum, multiple perforations of sigmoid colon, and multiple lacerations of branches of left internal iliac vein; and (2) pigment nephropathy, hemoglobinuric type, possibly due to massive transfusion of group O blood, with severe pulmonary congestion and edema, manifest clinically by severe oliguria, moderate arterial hypertension, and severe pulmonary edema.

Comment.-Although the blood group of this patient is not positively known, it is believed that his blood belonged to group AB. If this be true, his AB cells were susceptible to agglutination and hemolysis by the anti-A and anti-B iso-agglutinins found in the plasma of group O blood. The difficulty of providing enough group-specific blood is apparent, as are the shortcomings of universal donor blood under such circumstances.

Microscopic observations-Examination of the kidney tissue revealed that a very large number of collecting tubules, and a smaller number of distal convoluted tubules, were plugged with pigmented casts. There was slight dilatation of the proximal nephrons. The interstitial tissues were slightly edematous. In some areas, proliferation of the tubular epithelium was evident, the casts being encapsulated within the tubules. Here and there, small granulomas had formed in the interstitial tissues. There were scattered foci of lymphocytic infiltration.

Microscopic diagnosis-Severe hemoglobinuric nephropathy.



This case provides another example of posttraumatic renal failure, with the microscopic finding of pigment nephropathy. Although definite proof is not available, it is believed that the transfusion of a large volume of group O blood contributed to the production of the renal lesion. Perhaps, in the presence of severe shock, the kidney is particularly liable to injury by a relatively small amount of hemoglobin in the plasma. A moderately severe degree of hypertension was observed before death; its possible relationship to the pathologic physiology of renal failure is of considerable interest.

Case 7


This patient, a German prisoner of war with a bullet wound of the abdomen, was first seen 6 March 1945 at 1430 hours. The precise time of wounding was unknown but was probably several hours earlier. He reached a field hospital at 1715 hours, after having received 750 cc. of plasma. At this time, he was in profound shock, and his blood pressure could not be obtained. He was stuporous and at times semidelirious, and his appearance was striking because of a peculiar cherry-red flush of the face and extremities and a red mottling of the skin of the trunk. He was moderately dehydrated. The response to 1,500 cc. of plasma and 1,500 cc. of blood was poor; the blood pressure rose only to 78/40 mm. Hg, and no clinical improvement was evident. His blood group was O.

Operation was begun 6 March at 2210 hours. Laparotomy revealed extensive contamination of the peritoneal cavity by 2,000 cc. of bloody fecal material. There were multiple perforations of the jejunum, and the transverse colon was found almost completely torn away from its attachments except for a narrow band of tissue along the mesenteric border. It was necessary to resect about 15 cm. of the small bowel; three perforations in the jejunum were sutured, and a colostomy was performed. The patient received 1,000 cc. of whole blood during the operation.

At the conclusion of the operation, the blood pressure was at a very low level, and there was never any significant rise. Often it was unobtainable. The patient was stuporous, and death occurred 11 hours after operation.


Autopsy was performed 4 hours after death (St. Avold, France).

Gross observations-The mottled, pink-gray discoloration of the skin of the trunk noted before operation was still present.

The peritoneum showed the effects of severe contamination. A small amount of seropurulent fluid was present. The serosal surfaces of the bowel were opaque, dark and congested, and patches of fibrinopurulent exudate were present.

The heart was rather soft. The right side was distended by blood.


The pleural cavities were not remarkable, except for an old fibrous pleuritis at both bases.

Both lungs showed considerable hypostatic congestion and partial atelectasis of the dependent and inferior portions.

There was a moderately large laceration of the left lobe of the liver.

The spleen was about three times the usual size and was dark purple red. The consistency was firm. The pulp, which was uniformly dark purple, did not scrape away easily.

The gastrointestinal tract showed the sutured lacerations and anastomoses of the small bowel already described.

The kidneys were of the same size, but the right kidney appeared somewhat more congested and purple than the left. The cortical surfaces of both kidneys showed focal punctate hemorrhagic areas, more marked on the right side. On section, these hemorrhagic areas were seen to extend varying distances into the cortex in the form of small, blood-suffused columns. On section, the parenchyma bulged above the cut edge of the capsule. A few petechiae were present in the mucosa of the renal pelvis.

The cortex of the left adrenal gland exhibited several radial hemorrhagic areas in which there was possibly destruction or necrosis of the tissue. The medullary area was negative.

Gross anatomic diagnosis-(1) Perforating gunshot wound of abdomen, with severe laceration and avulsion of transverse colon; multiple perforations of small bowel; moderately severe laceration of left lobe of liver; (2) severe generalized peritonitis, secondary to injuries of transverse colon and small bowel; (3) extensive hemorrhagic focal necrosis of kidneys and possibly of left adrenal, causes unknown; (4) moderately severe splenomegaly, possibly secondary to massive transfusion therapy and peritonitis; and (5) hypostatic pulmonary congestion and edema, with bilateral partial atelectasis.

Comment-The exact lapse of time between wounding and operation in this case was unknown but seems to have been at least 16 to 18 hours. The patient was moribund on admission, as the result of shock and extreme peritoneal contamination.

The case illustrates the difficulty which occasionally was encountered in the management of patients with overwhelming contamination of the peritoneum. Once contamination of this degree has occurred, effective therapy is difficult even though the patient is seen relatively soon after wounding.

The peculiar gray-pink mottling of the skin suggests a considerable degree of capillary damage. The pathogenesis of the hemorrhagic columns in the renal cortex is obscure, but is possibly related to the severity of the shock and infection.

Microscopic observations-There were a few atelectatic patches in the lungs. The alveolar septa were hyperemic. The bronchial walls were infiltrated with lymphocytes. One small artery was thrombosed. Another section showed scattered small areas of hemorrhage into the peribronchial alveoli.


The centers of the lobules of the liver were congested. The portal triads were infiltrated with moderate numbers of lymphocytes.

The malpighian corpuscles had active germinal centers; many contained foci of degeneration, with nuclear debris. The pulp and sinuses were engorged with blood.

A moderate number of distal convoluted and collecting tubules in the kidneys contained pigmented casts. Protein precipitate was seen in many of the convoluted tubules. Some of the tubules contained hyalin casts. Neither inflammatory reaction nor edema was present in the interstitial tissues. A small number of basophilic structures, probably sulfonamide crystals, were seen. Vascular congestion was pronounced.

There was pronounced congestion of both the inner zone of the cortex and of the medulla. Small foci of lymphocytes were seen in the medulla. 

The serosal layer of the intestine was thickened and edematous. It was infiltrated with lymphocytes, histiocytes, and polymorphonuclear leukocytes. The mucosa was covered with blood, serum, and epithelial debris.

Microscopic diagnosis-(1) Hemoglobinuric nephropathy, moderately severe; and (2) focal necroses in spleen.

Comment-In the absence of other factors which produce pigmented casts in the renal tubules, the renal lesion in this case must be attributed to the blood transfusions which the patient received. His blood group was O, and he received 1,500 cc. of O bank blood. If the blood typings were correct, the crossmatches negative, and the blood free of hemolysis before transfusion, there is no evident explanation for the nephropathy. (J. G. R.)


This patient sustained an extremely severe wound of the abdomen with widespread peritoneal contamination, and his status on arrival at a field hospital reflected the relatively long lapse of time since the injury. He died within a short time, in spite of replacement therapy and prompt surgical care. Failure to survive such a wound seems to depend upon (1) the severity of the hemorrhage and (2) the deleterious effects of massive peritoneal contamination. It seems likely that appropriate bacteriologic studies would have revealed significant findings in this instance.

Case 8


This patient received a perforating bullet wound of the abdomen 30 March 1945, at 1740 hours. When he reached a field hospital at 1945 hours, he was in profound shock, with a blood pressure of 42/20 mm. Hg and a pulse of 128. His response to 500 cc. of plasma and 1,500 cc. of whole blood was not good. At 0030 hours 31 March, his blood pressure was only 80/60. His blood group was O.


Laparotomy was performed about 4½ hours after the patient was received at the field hospital. The peritoneal cavity was full of blood, and there were perforations of the stomach and proximal ileum, as well as a laceration of the left lobe of the liver. Severe bleeding was in progress from the gastrohepatic ligament; it was stopped as promptly as possible. The perforations in the intestine were then sutured, and a large defect in the transverse mesocolon was repaired. The spleen and the left kidney, both of which were badly lacerated, were removed. The wounds of entrance and exit were debrided. Throughout the operation, the patient's condition was critical in spite of the administration of 250 cc. of plasma and 2,750 cc. of whole blood.

He failed to improve after operation and died at 0005 hours 1 April 1945, about 24 hours after operation and 28 hours after injury.


Autopsy was performed 15 hours after death (Grossostheim, Germany). 

Gross observations.-There was considerable post mortem lividity of the neck and face. The abdomen was moderately distended.

The left chest contained 600 cc. and the right chest 200 cc. of blood-tinged fluid. The peritoneal cavity contained about 800 cc. of cloudy, blood-tinged fluid. The loops of small bowel were loosely bound together by strands of fibrinopurulent exudate, particularly in the vicinity of the sutured perforations of the ileum. No free purulent exudate was present in the peritoneal cavity. The adipose tissue of the colon and omentum exhibited numerous focal areas 0.5 to 1.5 cm. in diameter, of chalky, gray-white discoloration, particularly in the upper abdomen.

The heart was not remarkable except for the presence of subepicardial ecchymosis posteriorly along the coronary sulcus and over the wall of the left auricle.

The lungs were both markedly increased in weight and showed hypostatic congestion and atelectasis. The tracheobronchial tree was filled with a tenacious, blood-tinged mucoid exudate which was occasionally flecked by purulent material. Some of the smaller bronchi contained similar material. There was no evidence of established pneumonic consolidation.

The left lobe of the liver, which was the site of an extensive laceration, as well as a portion of the right lobe, was pale, yellow gray, and of abnormally soft consistency.

The body of the pancreas was partially torn away from its attachments, and there had been considerable autolysis of the contiguous tissues. The portion of pancreas adjacent to the bullet track was pale, reddish gray, and extremely soft in consistency. Surrounding this area were evidences of old hemorrhage. No purulent exudate, however, was present in the pancreatic area.

The gastrointestinal tract showed only healing sutured perforations. 

Gross anatomic diagnosis.-(1) Perforating gunshot wound of left abdo-


men, with severe laceration of left lobe of liver, possible necrosis of most of left lobe and of portion of right lobe; two healing perforations of proximal ileum; laceration or contusion of body of pancreas with autolysis of pancreatic tissue and acute fat necrosis throughout peritoneal cavity; severe laceration of spleen, surgical removal; severe laceration of left kidney, surgical removal; (2) severe bilateral pulmonary atelectasis, secondary to bronchial obstruction by mucus; (3) subepicardial hemorrhage, probably due to asphyxia, secondary to pulmonary atelectasis; (4) acute fibrinopurulent peritonitis, secondary to perforating wounds of intestine and liver; and (5) moderately severe bilateral pleural effusion.

Comment-Aside from the severe trauma to several organs, the insult of extreme pulmonary atelectasis appears to be immediately responsible for the death, which seemed inevitable from the time of wounding.

Microscopic observations-Numerous hemorrhages were present in the fat tissue around the adrenal gland. Several small hemorrhages were also seen in the medullary portion. A portion of the cortex stained a very faint pink resembling an area of infarction. The sinusoids were congested.

Numerous hemorrhagic areas were seen in two slides of the lung tissue. Areas of atelectasis were present. Cells of the heart-failure type were present in all areas. In a third slide, the tissue was for the most part normal, but in one area were alveoli filled with protein-rich fluid in which were many erythrocytes.

Granules of brown pigment were seen in the muscle bundles of the heart. Some hemorrhage was visible in the fat of the pericardium.

Extensive areas of hemorrhage and necrosis of liver substance were observed along one side of the first slide examined. In the immediate vicinity of the traumatized area were considerable accumulations of polymorphonuclear leukocytes and lymphocytes. In some areas were evidences of small abscess formation. Numerous lymphocytes were seen in nearly all the periportal areas. In a second slide, extensive areas of tissue destruction were present. The liver cells were dark-staining cords without nuclei.


The findings in this case are characteristic of the massive abdominal injuries which are by no means rare among severely wounded men. Wounds of this kind produce profound shock by destruction of tissue, hemorrhage, and massive peritoneal contamination. Death almost always occurs with little delay in spite of liberal replacement therapy and prompt surgical care. If the patient does survive the immediate postoperative period, the renal damage of the period of profound shock may prove fatal within a few days.

The microscopic observations in this case suggest that there was widespread necrosis of liver tissue which was not restricted to the missile track. Since the kidneys were not examined microscopically, the question of a renal lesion cannot be settled.


Case 9


This patient received a perforating bullet wound of the right buttock 30 March 1945, at 1700 hours. When he was received at the field hospital at 1945 hours, he appeared moribund; neither pulse nor blood pressure could be obtained. In the space of an hour, he was given 250 cc. of plasma and 2,500 cc. of group O blood, but the response was poor. Laparotomy was undertaken in the belief that hemorrhage was continuing. Just before operation, the blood pressure was 50/30 mm. Hg. The patient's blood belonged to group O. 

Exploration of the wound track from the right gluteal fold to the right iliac fossa revealed (1) a complete laceration of the right external iliac artery and vein, (2) a complete laceration of the right gluteal artery and vein, (3) division of the right femoral nerve, and (4) a perforation of the terminal ileum. All the injured vessels were ligated, and the ileal perforation was closed. During the operation, an extensive expanding retroperitoneal hematoma was noted. At the end of the operation, the blood pressure was 70/40 mm. Hg; 2,500 cc. of group O whole blood had been given while it was in progress.

The patient's condition continued critical after operation. On the first postoperative day, the hemoglobin was 55 percent of normal; the red blood cells numbered 3,290,000 per cubic millimeter. On the second day, the volume of urine was 225 cc. On the third day, no urine was excreted. The right leg was gangrenous, with a line of demarcation just below the knee. On the fourth day, no urine was excreted. The blood pressure was 160/80 mm. Hg. Venesection was carried out for the relief of pulmonary edema, but dyspnea continued. The patient died 3 April 1945, at 1635 hours, about 96 hours after wounding.


Autopsy was performed 1 hour and 25 minutes after death (Grossostheim, Germany).

Gross observations-The sclerae were slightly icteric. There was considerable brawny edema of the whole of the right leg, with an effusion into the knee joint and beginning dry gangrene below the knee.

The pericardial and pleural cavities contained small amounts of straw-colored fluid. There was a large bilateral retroperitoneal hematoma, which was larger on the right. Blood had disseminated upward as high as the perirenal tissue bilaterally and downward to fill the areolar tissue about the bladder. The right external iliac artery and vein and the right gluteal artery and vein had been ligated.

The heart was pale. Subepicardial petechial hemorrhages were distributed over the surface. The right ventricle was unusually soft. The tricuspid valve measured 13.5 cm., the mitral valve 10.3 cm., and the aortic valve 5.8 cm. The coronary vessels were normal.

Both lungs were partially atelectatic. The cut surface was very wet and exuded a large quantity of serosanguineous fluid. The bronchi were filled with edema fluid.


The liver was apparently slightly increased in size, but on section the parenchyma was not remarkable.

The spleen was moderately increased in size. The consistency was somewhat softer than normal and the malpighian bodies were prominent.

The gastrointestinal tract was negative, apart from a healing laceration of the ileum.

The kidneys, both of which presented the same appearance, were somewhat smaller than usual. The surfaces were smooth and were deep purple red, with a peculiar, finely mottled, nutmeg appearance. On section, the cortex was somewhat pale in comparison to the color of the renal surface. The peripheral portions of the pyramids as seen on section were dark, due to the presence of fine, closely placed, red-brown striae. The renal vessels were negative.

Gross anatomic diagnosis-(1) Perforating bullet wound of abdomen via right buttock, with laceration of right external iliac artery and vein, laceration of right gluteal artery and vein, division of right femoral nerve, healing perforation of terminal ileum, and incomplete fracture of right ilium; (2) severe retroperitoneal hemorrhage, secondary to laceration of iliac vessels; (3) edema and dry gangrene of lower right leg, secondary to ligation of iliac and gluteal vessels; (4) pigment nephropathy possibly due to shock and ischemia of right leg, as evidenced clinically by oliguria, slight hypertension, and pulmonary edema; (5) pulmonary congestion, bilateral, severe, with partial atelectasis, possibly due to uremic pneumonitis and left heart failure secondary to pigment nephropathy; (6) multiple petechial subepicardial hemorrhage; and (7) icterus, slight.

Comment-This case epitomizes the outcome in many patients who have suffered severe shock for long periods of time. In such instances, the ultimate decline is manifest chiefly as cardiorenal decompensation with increasing pulmonary edema, particularly when relatively large quantities of fluid are given. Ischemic muscle necrosis, massive transfusion therapy, and the extensive retroperitoneal hematoma present in this case may have contributed to the pigment which will almost certainly be found in the renal tubules on microscopic examination. Whatever may be the source of the pigment, it would appear that in this case the severe degree of shock was the most important factor in the development of the renal lesion.

The blood which this patient received was fresh; it was drawn in the field hospital just before it was administered. All flasks were inspected for hemolysis, and none was apparent. Routine crossmatchings showed no incompatibility, and the group of all the blood used was carefully checked just before transfusion.

Microscopic observations-A few small areas in the pulmonary alveoli contained fibrin and a few polymorphonuclear and mononuclear cells. The interlobular septa were edematous. On frozen section a large number of fat emboli were observed in the alveolar capillaries and arterioles.

The portal triads were infiltrated with moderate numbers of lymphocytes and eosinophiles.


The pulp of the spleen was moderately congested. There were increased numbers of eosinophiles in the malpighian bodies.

A very large number of the distal convoluted and collecting tubules in the kidney sections were plugged with pigmented casts. Sulfonamide crystals were seen here and there in the tubules. A few small foci of lymphocytes were found in the interstitial tissues. Epithelial proliferation had occurred about some of the casts in the pyramid. No fat emboli were seen on frozen section. 

Sections of the pancreas showed cystic dilatation of a small number of acini, which contained pink-staining material.

Microscopic diagnosis-(1) Severe pulmonary fat embolism; and (2) severe pigment nephropathy, hemoglobinuric type.


Because of his severe vascular injury and the resulting hemorrhage, this patient remained in shock for a prolonged interval. In spite of the fact that resuscitation and operation were accomplished with a fair degree of success, postoperative oliguria occurred, and death followed, from pulmonary edema, on the fourth day after operation. Microscopic examination confirmed the presence of pigment nephropathy and revealed, in addition, a high degree of pulmonary fat embolism. It is suggested that the fat embolism may have been the more important factor in the pulmonary edema which occurred on the fourth day after operation. It seems unlikely, however, that this patient could have survived the renal lesion.

The amount of trauma to the bone appears to have been relatively mild in this case, and it may be that injury of gluteal adipose tissue contributed to the pulmonary embolism.

Case 10


This patient received a perforating shell-fragment wound of the thorax and abdomen 14 December 1944, at 1020 hours. When he reached a field hospital at 1245 hours, the blood pressure was 80/34. Hg, the pulse was 140, and he was in moderately severe shock. He complained of shortness of breath and of right-sided chest pain. The wound of entry was at the left lateral costal margin and the wound of exit in the fourth interspace in the right anterior axillary line. Right thoracentesis produced 960 cc. of blood. The right 4th to 12th intercostal nerves were infiltrated with procaine, and pain in the chest lessened. At 1525 hours, after the administration of 500 cc. of plasma and 1,000 cc. of whole blood, the blood pressure was 122/64 mm. Hg. A catheterized specimen of urine showed occasional erythrocytes.

Thoracotomy was begun at 1600 hours, under endotracheal gas-oxygen-ether anesthesia. The thorax was opened in the right fifth interspace, and fragments of the fifth costal cartilage were removed. Five hundred cubic centimeters of blood were aspirated from the right pleural cavity. A small laceration


of the lower lobe of the right lung was closed with a mattress suture. After a 6-cm. laceration of the right diaphragm had been sutured, the thoracotomy incision was closed. The sucking wound of the anterior right chest was debrided and closed.

Laparotomy revealed much blood in the peritoneal cavity. The source appeared to be the severely lacerated and comminuted left lobe of the liver. There was an extremely severe laceration of the left transverse colon, which involved about three-fourths of its circumference for a distance of 7 cm. This portion of the large bowel was exteriorized through a left subcostal incision. No perforations were evident in the remainder of the gastrointestinal tract, and at this time there was no evidence of an acute inflammatory process within the peritoneal cavity. The spleen and kidneys were not injured. During the course of the operation, the patient's condition remained poor, although he received 1,500 cc. of whole blood, 750 cc. of plasma, and an autotransfusion of the 960 cc. of whole blood which had been aspirated from the right chest before operation. The operation required 2 hours.

Immediately after operation, the blood pressure was 40/0 mm. Hg, the pulse 104, and the respiration 24. The respirations were described as gurgling. The patient reacted from anesthesia at 0100 hours 15 December 1944, but respirations ceased at 0625 hours, about 20 hours after wounding.


Autopsy was performed 7 hours after death (Meisenthal, France).

Gross observations-The peritoneal cavity contained 300 cc. of foul-smelling, bloody fluid. There was an extensive laceration of the left lobe of the liver. The omentum, which was dirty and hemorrhagic, was displaced about the colostomy. Gentle pressure upon the stomach caused leakage of gas and gastric contents through a small aperture in the midportion of the omentum just inferior to its attachment to the greater gastric curvature. The loops of small bowel, particularly those in the lower portion of the peritoneal cavity, exhibited evidences of marked peritoneal contamination. The serosa was injected, bright red, and more opaque than normal. No fibrinopurulent exudate was recognized.

The right pleural cavity contained 300 cc. of blood, but there was relatively little fibrinous exudate. There was a defect in the fifth costal cartilage and a fracture of the sixth costal cartilage on the right.

The heart was somewhat dilated. The myocardium was softer than usual in consistency.

The appearance of both lungs was the same. They were considerably increased in weight and exhibited generalized atelectasis and loss of resiliency. The laceration of the right lower lobe had been closed, and contusion of the pulmonary parenchyma was not marked. On section, much blood and serous fluid dripped from the cut surfaces. The main bronchi and their smaller branches were occluded by a tenacious, mucosanguineous exudate which appeared to be responsible for the observed atelectasis. There was no evidence


of bronchopneumonia, but some of the smaller bronchi contained moderate amounts of mucopurulent exudate.

The spleen was small and soft but not otherwise remarkable.

The liver was of normal size. There was marked destruction of the left lobe, but no evidence of suppuration, and little autolysis was observed along the track of the missile. No exudate was present, nor was there a discernible zone of leukocytic infiltration. Thrombosis of the hepatic veins was noted only in the immediate vicinity of the laceration and was not striking. On section, the right lobe of the liver was not remarkable.

The stomach was greatly distended by partially digested food. There was a small (8-mm.) unsutured perforation in the greater curvature 10 cm. proximal to the pylorus. The contamination of the lesser peritoneal sac was minimal, and gastric contents appeared to have been extruded into the main peritoneal cavity through a small perforation in the omentum. A small (3-mm.) perforation of the ileum was located about 90 cm. from the ileocecal valve.

The kidneys were not enlarged, but the smooth cortical surfaces were pale, and on section the cortex showed faintly hemorrhagic markings through the pale yellow-gray substance. The renal pyramids were faintly striated by red-brown markings and had a peculiar pale-pink coloration.

Gross anatomic diagnosis-(1) Thoracoabdominal shell-fragment wound, with entrance at left costal margin and exit via right chest at level of fifth costal cartilage, with small perforation of greater curvature of stomach; small perforation of terminal ileum; severe laceration of transverse colon, loop colostomy; severe laceration and comminution of left lobe of liver; sutured perforation of right leaf of diaphragm; moderately severe laceration of lower lobe of right lung, closed; multiple compound fractures of right fifth and sixth costal cartilages and left sixth and seventh costal cartilages; (2) severe bilateral bronchial obstruction with pulmonary atelectasis; (3) acute generalized peritonitis, secondary to perforating wounds of stomach, terminal ileum, and transverse colon; (4) moderate right hemothorax, secondary to thoracoabdominal wound; and (5) congestion and edema of lungs, probably secondary to bronchial obstruction and atelectasis.

Comment-The severe laceration of the liver and the extreme peritoneal contamination were ominous findings at operation. It is remarkable, however, that there was so little evidence of peritonitis at this time. A severe fibrinopurulent peritonitis was present at autopsy. The difference was perhaps due simply to the passage of time, though it must be remembered that unclosed perforations of the stomach and ileum had permitted additional contamination of the peritoneal cavity.

The large amount of bloodstained mucoid exudate present in the bronchi at autopsy was remarkable. It did not seem reasonable to try to account for it simply on the basis of the contusion of the right lung. The severe atelectasis appears to have been immediately responsible for death, and its development may have been favored by the fact that after operation the patient received two injections (each of 0.008 gm.) of morphine sulfate.


The use of autotransfusion under the circumstances of this case constituted an error of judgment. In the presence of a large wound of the colon, the blood could scarcely have escaped contamination, and its use undoubtedly contributed to the fatality.

Microscopic observations-The lung tissue showed large areas of atelectasis. The alveolar septa were hyperemic. Many pigmented macrophages were seen within the alveoli.

One section of the liver was essentially normal. The other showed extensive traumatic infarction.

Examination of the spleen showed the malpighian bodies to be numerous and large and the pulp slightly congested.

A few cystic tubules in the kidney were lined with flattened epithelium containing pigmented material. The epithelial cells in some of the collecting tubules contained granules of golden-brown pigment. A single focus of lymphocytes was seen in the interstitial tissues.

One of the interlobular septa seen in the section from the pancreas was infiltrated with polymorphonuclear leukocytes and a few histiocytes.

A section of voluntary muscle showed hemorrhage and polymorphonuclear leukocyte infiltration into the tissues between the muscle bundles. 

A small amount of purulent exudate covered a small area in the serosa of the jejunum. In the spaces between the valvulae conniventes were desquamated and necrotic epithelial cells. The mucosa was infiltrated with a moderate number of lymphocytes and a few eosinophiles.

Section from the stomach showed hemorrhage into the submucosa and many mononuclear cells containing formalin-pigment (formaldehyde pigment) precipitate. In one small area was a collection of polymorphonuclear leukocytes.


This is an example of a very severe thoracoabdominal wound in which death occurred within a relatively short time after injury. Extreme destruction of the liver and widespread peritonitis appear to have been the most important factors in the fatality. The patient's critical state and the widespread visceral damage explain why two small perforations of the gastrointestinal tract were missed at operation. Other important lethal factors were undoubtedly the continuing peritoneal contamination, the autotransfusion of contaminated blood, and the rather liberal administration of morphine.

Case II


This patient received a shell-fragment wound of the abdomen 17 October 1944 about 1130 hours. When he reached a field hospital at 1350 hours, his blood pressure was 128/70 mm. Hg, and his condition appeared surprisingly good. The foreign body had entered the abdominal cavity through the right


upper quadrant and had made its exit through a large defect in the left upper quadrant, with the evisceration of 14 to 16 cm. of the transverse colon, several loops of small bowel, and most of the omentum. Two hundred and fifty cubic centimeters of plasma were administered rapidly, and a transfusion of group O whole blood was begun. During roentgenologic examination, after he had received about 300 cc. of blood, the patient had a chill. The transfusion was stopped immediately, and a second flask of blood was substituted for the first. On the operating table, shortly after the chill, the blood pressure was 90/60 mm. Hg, and operation was deferred until it reached 124/70 mm. Hg. This delay amounted to about 2 hours.

Laparotomy revealed a small amount of blood and some bile in the peritoneal cavity. There was a large laceration of the liver, a perforation of the second portion of the duodenum, a perforation of the midjejunum, and multiple perforations of the transverse colon. The duodenal perforation was sutured, and posterior gastroenterostomy performed. The jejunal perforation was sutured. A segment of the colon was exteriorized. It was noted that during operation bleeding tended to be excessive and hemostasis was difficult. Before the abdomen was closed, 50,000 units of penicillin were placed in the peritoneal cavity.

The patient's condition was so critical after operation that he was immediately given 1,000 cc. of blood. The blood pressure had fallen to 90/60, and shortly afterward it fell to 80/50 mm. Hg. The day after operation, the exteriorized colon was opened; its appearance was not remarkable. Coarse rales appeared in the chest, and considerable quantities of purulent material were aspirated from the trachea. Death occurred 19 October 1944, about 48 hours after operation.

During this period of survival after injury, this patient received a total of 3,000 cc. of plasma and 3,500 cc. of blood. Before the first plasma transfusion was started, a sample of blood was taken. The group was O, and all the blood which he received was compatible by routine crossmatching.


Autopsy was performed 2 hours after death (Lunéville, France).

Gross observations-The right pleural cavity contained 200 cc. of serous fluid. The peritoneal cavity contained 300 cc. of thin, purulent exudate. A generalized fibrinopurulent peritonitis was present. There apparently had been no appreciable hemorrhage from the liver after operation, and the serous surfaces were not bilestained. Neither the sutured lesions of the intestine nor the gastroenterostomy was remarkable. The colostomy was also not remarkable. There was a large defect of the anterior peritoneum medial to the colostomy in the left upper quadrant, and there appeared to be an early infection in the properitoneal tissues of the upper abdomen.

The heart was negative except for slight atherosclerosis of the coronary arteries.


There was considerable atelectasis of all lobes of the lungs, particularly in the lower and posterior portions. The right upper lobe, in addition, was firm and had a slightly nodular consistency. On section, there was early peribronchial consolidation. The bronchi contained a moderate amount of mucopurulent, blood-tinged exudate.

The spleen was three times normal size. It was soft, and on section the pulp was easily scraped away. The secondary follicles were prominent.

A small laceration of the right lobe of the liver was not remarkable. Although local congestion and focal hepatic vein thromboses were noted adjacent to the missile track, the hepatic defect was partly filled by plasma clot.

The kidneys were moderately swollen and congested. On section, the cortex was slightly pale.

Gross anatomic diagnosis-(1) Perforating shell-fragment wound of upper abdomen, with evisceration of transverse colon, small bowel, and omentum; multiple perforations of transverse colon; perforation of jejunum; perforation of second portion of duodenum; moderately severe laceration of liver; (2) generalized severe fibrinopurulent peritonitis; (3) generalized severe congestion, edema, and atelectasis of lungs; (4) early acute bronchopneumonia of right upper lobe; (5) acute splenomegaly, secondary to peritonitis and possibly to multiple transfusions; and (6) moderately severe passive congestion of kidneys.

Comment-Death in this case appears to have been caused by severe visceral trauma and the effects of massive contamination of the peritoneal cavity. The chill which occurred during transfusion was attributed either to chilling of the patient or to the presence of pyrogens in the transfusion equipment. It was not thought to be due to incompatibility of the blood. The appearance of the urine, however, suggested that hemoglobin-derived pigment might have been present. Plasma was given in an unusually large volume during operation, in an effort to limit the amount of blood needed, and thus to lessen the danger of a transfusion reaction.

Microscopic observations-The heart was negative, except for slight atherosclerotic changes in the coronary arteries.

The pulmonary tissue showed an extensive bronchopneumonia, confluent in places. The bronchial and alveolar exudate in some lobules was composed of polymorphonuclear leukocytes. In some areas, many red blood cells were mixed with the leukocytes. In one section in which the alveoli contained protein-rich edema fluid, a few hyalin membranes were seen in the respiratory bronchioles.

Sections from the liver showed large areas of hemorrhage and traumatic infarction. One section showed a moderate degree of vacuolation of the liver cells in all parts of the lobule.

Groups of fat cells were seen in some of the malpighian bodies of the spleen. Considerable nuclear debris was enmeshed in a fibrinous exudate which covered the capsule.


Sections from the kidneys showed that a moderate number of collecting tubules and a small number of distal convoluted tubules contained pigmented casts. Some of the proximal nephrons were dilated. There was no epithelial proliferation, interstitial edema, or inflammatory reaction.

Microscopic diagnosis-Moderately severe hemoglobinuric nephropathy. 

Comment.-This case represents another instance of hemoglobinuric nephropathy in a patient supposedly belonging to blood group O. In the experience of the 6713th Blood Transfusion Unit at Naples, which covered many thousands of donors, listings of the blood type on the identification tag were found to be incorrect in about 10 percent of the soldiers, and one wonders whether that happened in this case. Unless hemolyzed blood is injected, there would seem no reason for a transfusion reaction in a group O recipient when O blood is given. In crossmatching, there is one possible source of error in reading the result if the possibility is not borne in mind. Agglutination hemolysis may occur, and crossmatching may be reported by an unwary observer as compatible. (J. G. R.)


This patient had a very severe abdominal wound, with massive evisceration of abdominal viscera, and death occurred within a relatively short time after injury. The blood pressure remained at a low level during operation and most of the postoperative period. It is thought that the most important lethal factors were (1) severe abdominal injury with evisceration and consequent peritonitis, (2) prolonged anesthesia and operation, and (3) bronchopneumonia and pulmonary edema.

Uremia per se probably was not important in producing death so soon after operation, and the importance of the renal lesion as a lethal factor is minimized by at least two facts, namely, the moderate renal changes present and the fact that the patient excreted 300 cc. of urine during a period of 48 hours in which the blood pressure ranged from 90/60 mm. Hg downward.

It was customary in World War II to disregard the blood group listed on the identification tag, for the reasons already stated. This patient was regrouped in the field hospital, and his blood type was found to be O. He did, however, receive blood which had been drawn 6 to 9 days earlier, and small amounts of free hemoglobin may have been present in the 3,500 cc. of blood which he received.

Case 12


This patient received a shell-fragment wound of the right hip and sacral region 9 October 1944, at about 0800 hours. He was brought into the field hospital at 1700 hours with a rapid, weak pulse and blood pressure of 70/50 mm. Hg. He had received 500 cc. of plasma. Before operation he received


another 500 cc. of plasma and 2,500 cc. of group O (his blood group) blood, but the blood pressure never rose above 90/60 mm. Hg. Through error, he did not receive penicillin before operation.

Operation was unavoidably delayed until approximately 24 hours after injury. Exploration revealed an extensive perforating wound of the right buttock with widespread destruction of the gluteal muscles and fractures of the right wing of the ilium and the right side of the sacrum. The missile track was filled with semiliquid, necrotic muscle, which was pale pinkish gray. The necrosis evident along the walls of the missile track seemed to extend for a considerable distance into the muscles of the lumbar region and posterior upper thigh. So extensive was the process that complete excision of damaged muscles was impossible. At operation, sections were taken from the right gluteus medius at the base of the wound and were fixed for microscopic study. At no time during operation were gas bubbles or crepitus noted in the muscle or in the subcutaneous tissue adjacent to the wound.

Laparotomy revealed two small perforations of the terminal ileum, but the site of entry of the foreign body could not be identified. The right colon was intact. Small spicules of bone had been driven into the soft tissue, and these fragments may have been responsible for the injury to the small bowel. Sigmoid colostomy was performed. No perforations were found on careful examination of the intraperitoneal portion of the rectum. Before the abdomen was closed, 10 gm. of sulfanilamide crystals were placed in the abdominal cavity. The wound of the hip was widely debrided, and a fasciotomy was performed along the right lateral gluteal fold. Zinc peroxide, which was sprinkled liberally throughout the gluteal wound, could not be activated because no reliable source of heat was available. Operation was completed at about 1200 hours 10 October. During its course, the patient received 1,000 cc. of blood.

Immediately after operation, the patient was given 2.5 gm. of sulfadiazine sodium by vein. On the third day, it was found that the penicillin which had been ordered intramuscularly in dosages of 25,000 units every 3 hours was being given subcutaneously. As soon as the error was discovered, 100,000 units were given intravenously.

For the first day or two after operation, the patient looked remarkably well. There was little systemic evidence of severe infection, the pulse remained within the normal range, and the maximum temperature elevation was 99.6° F. On the second postoperative day, the blood pressure was 126/90 mm. Hg. The patient appeared slightly drowsy and was oliguric. An attempt was made to render the urine alkaline by placing 4 gm. of sodium bicarbonate in the stomach every 2 hours. An intravenous infusion of 1,000 cc. of 1.25 percent sodium bicarbonate solution also was given daily. Although the total daily fluid intake averaged 3,000 cc., the greatest single volume of urine was about 100 cc., and the total output from operation to death, which occurred on the sixth day after wounding, was about 400 cc. The urine was dark amber brown and contained many granular casts and many erythrocytes. A few


sulfadiazine crystals were noted soon after operation. The hemoglobin by the copper sulfate method was 12.7 gm. percent and the plasma protein 5.1 gm. percent.

The patient received an average of 250 cc. of plasma daily, and during the immediate postoperative period he also received 750 cc. of carefully crossmatched O-type blood.

On 13 October, the blood pressure was 178/120 mm. Hg, but it declined from this level and during the last 3 days of life was in the range 140/80 to 160/85 mm. Hg. On 13 October, the leukocyte count was 6,000 per cubic millimeter, with a moderate increase in stab forms. There was also moderate pitting edema, and the temperature reached 100.2° F. On the night of the fifth postoperative day, moist rales were noted throughout the chest, and intratracheal aspiration was performed. The patient died the following morning, 15 October 1944, at 0800 hours, approximately 144 hours after injury.


Autopsy was performed an hour after death (Lunéville, France).

Gross observations-Externally the body showed moderate pitting edema over the back and sacrum. There was no icterus.

The peritoneal, pleural, and pericardial cavities were negative.

The right side of the heart was greatly distended by blood; the tricuspid valve measured 15 cm. There was a considerable degree of atherosclerosis of the coronary arteries, most marked in the right circumflex artery.

There was complete atelectasis of all lobes of the lungs except the right upper lobe. There was loss of crepitus, and the involved lobes were rubbery in consistency. Consolidation was thought to be present but was difficult to detect because of the marked degree of atelectasis. On section, the lung was purple and beefy red. There was only a moderate increase in the amount of serous fluid which oozed from the cut surface. The bronchi contained a large quantity of hemorrhagic purulent exudate.

The spleen was twice normal size and moderately firm. On section, it did not seem remarkable, but the purple-red pulp scraped away somewhat more easily than would have been expected from the firmness of its consistency.

The liver was moderately enlarged. On section it was not remarkable except for the presence of scattered pinpoint areas of hemorrhage.

The gastrointestinal tract showed the operative changes already mentioned. 

Both kidneys were considerably swollen, particularly in the anteroposterior diameter. Their size was estimated at 1½ times normal. The subcapsular surface was uniformly homogeneous, yellow gray, and of an opaque, slightly granular appearance. On section, the parenchyma bulged above the edge of the capsule. The cortex appeared somewhat thickened and was of the same yellow gray as the surface. The pyramids were gray brown; near their bases the tissue had an opaque granularity like that of the cortex. The tips of the pyramids were congested and the mucosa was hemorrhagic, as was the mucosa of the renal pelves generally. The renal arteries showed early atherosclerosis.


Examination of the adrenals showed liquefaction and destruction entirely out of proportion to the brief time which had elapsed since death. There was no evidence of medullary hemorrhage.

There was moderate edema of the bladder mucosa.

In the lumbosacral area was an extensive phagedenic wound track extending from the right posterolateral rib margin downward and across the midline to the left gluteal region. There was considerable undermining of the subcutaneous tissue, with involvement of the adjacent muscle and connective tissue about the periphery of the track. The odor was foul. The exudate was watery and grayish green, except that in many places it was slightly pink, the color suggesting the presence of blood or muscle pigment. The exposed sacrum and the wing of the ilium formed the floor of the wound. There was no evidence that the peritoneal cavity had been invaded.

At operation, an attempt had been made to provide drainage for the wound by making incisions along the right gluteal fold and in the posterior right flank. The edges of these incisions had the appearance of those of the original wound. Infection had extended upward to the level of the 12th rib along the lumbar muscles and downward into the posterior muscles of the upper thigh. Grossly there was no apparent predilection for muscle tissue, and adipose and connective tissue were extensively involved. In the midst of the wound, however, pale-pink, edematous, and necrotic muscle fibers were present.

Gross anatomic diagnosis-(1) Perforating shell-fragment wound of right lumbosacral and gluteal regions, with extensive destruction of soft tissue, and compound, comminuted, incomplete fracture of right ilium; compound, comminuted, incomplete fracture of right posterior aspect of sacrum; two healing perforations of terminal ileum; (2) extensive phagedenic infection of wound of gluteal region, with necrosis of muscle and connective tissue, bacterial cause unidentified; (3) nephrosis, possibly hemoglobinuric in type, with pulmonary congestion and edema, moderately severe, manifest clinically by oliguria, uremia, and arterial hypertension, moderately severe; (4) purulent bronchitis, bilateral, severe, with complete atelectasis of left lung, right lower and middle lobes, due to bronchial obstruction by exudate; (5) extensive bilateral necrosis of adrenal medulla, possibly due to overwhelming infection; and (6) moderate, generalized atherosclerosis.

Comment-The clinical and gross post mortem findings in this case are typical of pigment nephrosis. In view of the widespread destruction of skeletal muscle, myoglobin rather than hemoglobin may have been the offending pigment. Whatever the pigment, the kidney always appears to be particularly vulnerable to it during periods of severe shock.

This type of infection is of considerable interest. Bubbles of gas were never observed, but the invasive and destructive properties of the infecting organisms were evidently very great. A mixed infection which includes microaerophilic streptococci deserves consideration. The delay in effective penicillin therapy may or may not be important. The delay of 24 hours in operation certainly was.


Microscopic observations.-Sections of the lungs showed small patches of atelectasis. Several lobules were filled with exudate composed largely of red blood cells, stringy protein deposits, and small numbers of polymorphonuclear leukocytes and mononuclear cells. The character of the alveolar exudate varied in different lobules. In some it was chiefly composed of erythrocytes. In others, the red blood cells were mixed with granular, pink-staining protein precipitate. In still others, there was an admixture of these elements with mononuclear cells. A few small bronchi contained polymorphonuclear leukocytic exudate. Slight edema of the interlobular septa and blood were observed in scattered alveoli, but nothing else of note was seen. On frozen section, fat emboli, in moderate numbers, were present in capillaries in the alveolar septa.

The liver cords in the centers of the lobules were atrophic. The cells contained fat vacuoles and granular brown pigment. A few foci composed of histiocytes were seen. The portal triads were infiltrated with moderate numbers of lymphocytes.

The splenic pulp was moderately congested. There was a large amount of formalin-pigment (formaldehyde pigment) precipitate.

A large number of distal convoluted and collecting tubules in the kidneys were plugged with pigmented casts. The proximal nephrons were slightly dilated. Some tubules contained epithelial debris mixed with pigment. There was a moderate degree of epithelial proliferation, most evident in the medulla, with the formation of a few granulomas in the interstitial tissues. Only a very slight inflammatory reaction, consisting of lymphocytes, was seen. A small number of sulfonamide crystals were identified in the tubules.

The appearance of the voluntary muscle varied in different sections, and even in different parts of the same section. Many muscle fibers were swollen. Some took an orange stain. Others had lost striations and nuclei. In places, the interstitial tissues were edematous. There was a large amount of polymorphonuclear leukocytec infiltration in the interstitial tissues in some areas and a small amount in others. In still others, there was an absence of inflammatory reaction. A few vessels were thrombosed.

Sections examined after straining by the MacCallum-Goodpasture technique showed large numbers of organisms, usually mixed. There were present large, fat, gram-positive bacilli, some of which were comma shaped. No spores were seen. The other prominent organism was a gram-positive coccus. Some of these cocci were small, some large.

Microscopic diagnosis-Moderate pulmonary fat embolism. 

Comment.-Histologically the voluntary muscle lesion in this case appears to be a combination of clostridial and suppurative myositis. Without cultural proof, an exact diagnosis is not possible. One can guess, however, that the germ-positive rods were Clostridium sordellii and that the cocci were streptococci. 

The renal lesion is typical of hemoglobinuric nephropathy. The very destructive muscle lesion might conceivably have liberated myoglobin, but whether this would be absorbed and then excreted through the kidney, as in the crush syndrome, it is not possible to say. In addition, clostridial organ-


isms are hemolytic, and, if hemolysis were sufficiently severe, this process, too, might produce hemoglobinemia and hemoglobinuria. In a recipient belonging to group O and receiving only group O blood, there would seem no reason for a transfusion reaction, unless hemolyzed blood had been injected. This fact would tend to favor a relationship between the myositis and the nephropathy.


This case provides another example of posttraumatic renal failure and pigment nephrosis. The type of wound infection encountered here, however, was not common, and the myositis per se may have been important in the development of the renal lesion. In attempting to assess the role of this infection in the pathogenesis of the nephrosis, other factors should not be overlooked, including prolonged and severe shock due to trauma and hemorrhage, massive transfusion therapy, and sulfonamide administration. Probably the single most important lethal factor was the prolonged period of shock, and it may well be in this regard that the myositis exerted its deleterious effect upon the kidney.

Case 13


This patient suffered a shell-fragment wound of the right costovertebral region 18 May 1944 at 1300 hours. After a delay of 38 hours, during which he received 1,000 cc. of plasma and several injections of morphine, he arrived at a field hospital 20 May at 0300 hours. The delay in evacuation was due to an offensive breakthrough; combat units outstripped supporting medical units. 

At this time, the patient's condition was critical. The blood pressure was 88/52 mm. Hg, the pulse 98, and the respiration 26. He was dehydrated, febrile, and semistuporous. In 4 hours after he reached the field hospital, he received 1,400 cc. of plasma and 1,000 cc. of group O whole blood; 10,000 units of polyvalent gas-bacillus antitoxin; and 100,000 units of penicillin by vein and 100,000 units by instillation into the right pleural cavity, from which 1,000 cc. of bloody, foul-smelling fluid had been aspirated. These measures produced little improvement; at 0600 hours the blood pressure was 108/30. 

At operation, a laceration of the diaphragm was closed. The chest was also closed after a catheter had been inserted in the right pleural cavity. Laparotomy revealed a laceration of the right kidney. The kidney was not removed, and drainage was provided for the perirenal, subdiaphragmatic, and subhepatic regions on the right. The patient was apparently in fairly good condition throughout the procedure but died suddenly as operation was being concluded.



The autopsy was performed 1 hour after death (Minturno, Italy).

Gross observations-The skin and sclerae were definitely icteric. The abdomen was moderately distended and tympanitic. No bubbles of gas were


noted in the tissues at the site of entrance of the missile, which was beneath the 12th rib in the right costovertebral region. The shell fragment had passed forward through the reflection of the diaphragm and then through the dome of the liver, coming to lie beneath the skin in the right midclavicular line at the level of the sixth intercostal space.

The right pleural cavity was remarkable in that the right lung was almost completely collapsed and the entire cavity was lined with a thick, fibrinous exudate. This cavity contained about 200 cc. of bloody fluid. The stomach was greatly dilated; it extended downward to a point 3 cm. below the umbilicus. It was so distended with gas that it appeared tense and drumlike. The hepatic flexure of the colon was plastered to the anterior surface of the right kidney by fibrinous exudate. There was moderate bile staining of the lesser peritoneal sac and of the retroperitoneal tissues of the right renal region.

There was practically total collapse of the right lung, which was bound down by the exudate just described. This exudate covered all exposed surfaces, and in the thicker portions, which lay posteriorly, there were many bubbles of gas. The amount of gas present was out of proportion to the degree of subcutaneous emphysema present, and it was thought that the gas represented the product of bacterial growth in the exudate. This lung contained practically no air, and on section it was found to be almost completely consolidated, the collapsed tissue being suffused with hemorrhagic fluid. There was almost complete atelectasis of the left lower lobe and moderately severe patchy atelectasis of the left upper lobe, in which peripheral emphysematous blebs were observed. The bronchi contained moderate amounts of serous bloody fluid, but there was no evidence of obstruction of a major bronchus.

A missile had passed through the dome of the right lobe of the liver. Incision through Glisson's capsule at a point of hemorrhagic discoloration released bloody fluid containing bubbles of gas. On section, the wound track measured 5 to 6 cm. in diameter. The walls were made up of pale-yellow, opaque, necrotic liver parenchyma, which contained small, irregular bubbles of gas and had the appearance of fermented culture media. Bubbles of gas were expelled from the blood vessels of the liver.

Except for the extreme gastric dilatation already described, and thinning of the wall by pressure of retained air and gas, the stomach was not remarkable. 

The passing missile had caused a severe contusion of the central portion of the right kidney, and on section a spherical hemorrhagic area was seen, which was 4 to 5 cm. in diameter. Overlying this area was considerable capsular reaction, but the cortex of the kidney did not appear to be directly lacerated by the missile.

The spleen was moderately swollen and somewhat softer than normal. 

The capsule overlying the relatively normal portion of the kidney stripped away with some difficulty and slight tearing of the underlying parenchyma. The renal pelvis on the right was bile stained to a degree not noted on the left. 

Gross anatomic diagnosis.-(1) Shell-fragment wound of right thoraco-abdominal region, with right hemopneumothorax and extreme atelectasis of


right lung; empyema of right pleural cavity, possibly caused by clostridia; laceration of right lobe of liver with extensive necrosis due to anaerobic infection; and (2) severe atelectasis of left lower lobe and partial atelectasis of left upper lobe, secondary to right pneumothorax and gastric dilatation; severe acute dilatation of stomach; severe contusion of right kidney, and moderately severe contusion of hepatic flexure of colon; sutured laceration of right diaphragm.

Comment-When this patient was received in the field hospital, he was suffering from overwhelming infection and severe respiratory embarrassment. Gastric dilatation and atelectasis of the left lung may have accounted for his sudden death on the operating table. A roentgenogram taken before operation showed a large gas bubble in the region of the stomach.

Microscopic observations-The heart veins were dilated and engorged with blood.

Two sections of the lungs showed massive collapse with patent bronchi. Many dust cells were present. The pleural surfaces were covered with a fibrinopurulent exudate which showed no evidence of organization. Exudate stained with the MacCallum-Goodpasture stain showed many gram-positive spore-bearing bacilli morphologically consistent with clostridia.

One section of the liver showed extensive traumatic necrosis. A second showed central congestion, acidophilia, and slight vacuolation.

The splenic pulp was very cellular.

One section of kidney tissue showed only moderate diffuse dilatation of all portions of the nephrons. A second showed massive hemorrhagic traumatic necrosis. Many of the collecting tubules were plugged with casts; some were hyaline, some consisted solely of red blood cells, and others showed all grades of transition.

The pleural surface of the pericardium was covered with a fibrinopurulent exudate. The pericardial surface was normal.

Comment-This is the first case to come to autopsy in the Naples laboratory with what appears to be a clostridial infection of the pleural cavity. (T. B. M.)


Even if this patient had survived operation, it is doubtful that he could have tolerated such overwhelming infection. His response to replacement therapy before operation was slow and incomplete and was in itself indicative of the profound circulatory changes which characterize clinical gas-bacillus infection.

Case 14


This patient received a bullet wound of the mid-upper abdomen 17 April 1945, at 1800 hours. He reached a field hospital 18 April, at 0130 hours, without replacement therapy during evacuation. The blood pressure at this


time was 80/60 mm. Hg and the pulse 120. Improvement was prompt after the administration of 500 cc. of plasma and 500 cc. of whole blood, and the blood pressure before operation was 128/80 mm. Hg. The patient's blood belonged to group O.

Laparotomy through a left paramedian incision revealed the omentum to be badly lacerated and torn. The proximal jejunum, which was transected, was repaired by end-to-end anastomosis. The splenic flexure of the colon, which was badly lacerated, was resected and the loops were brought out in the left abdomen as a colostomy. The posterior wound of exit in the left flank was debrided. Exploration showed that the left kidney was lacerated, but it was not removed. The perirenal space was drained, and drains were brought out through the inferior portion of the abdominal incision. The patient's condition was satisfactory during operation, in the course of which he received an additional 500 cc. of blood.

His condition the day of operation continued fairly good. He voided 600 cc. of urine. The following day, 19 April, his status worsened rapidly; he became stuporous and was unable to void. The following laboratory values were reported: Hematocrit, 51; hemoglobin, 17.2 gm. percent; and plasma protein, 6.5 gm. percent.

On 21 April, the patient's condition was fair, though drainage of small-bowel contents was noted through the posterior wound in the left flank, and it was thought that the jejunal anastomosis might have broken down.

On 22 April, the patient was very ill, and there was evidence of a severe peritonitis. He died on this day at 1000 hours, approximately 112 hours after injury.


Autopsy was performed 4 hours after death (Poxdorf, Germany).

Gross observations-Externally the body showed dehydration and moderate weight loss. The abdomen was quite distended. Foul-smelling, watery fluid oozed from the wound in the left flank.

A considerable quantity of gas escaped when the abdominal cavity was opened. The cavity contained about 2,000 cc. of foul, blood-tinged fluid in a large pocket which extended from the left upper abdomen across the midline to the right para-umbilical region. The walls of this pocket were made up of fibrinopurulent exudate, and the serosa of the adjacent bowel was opaque and dark in color. Fluid present in the pelvic portion of the peritoneum contained small particles of undigested food. Strands of purulent exudate were scattered throughout the peritoneal cavity. The diaphragm was elevated to the level of the third rib bilaterally; both leaves were intact.

There was much old blood in the left perirenal tissues.

The left cardiac ventricle showed moderately severe concentric hypertrophy.

The lungs revealed only slight congestion and atelectasis of the inferior portions of the lower lobes.


The spleen was somewhat softer than normal.

In the posterior wall of the stomach were two unclosed perforations. The larger, which measured 4 cm. in diameter, lay near the attachment of the gastrocolic omentum in the midportion of the greater curvature. The second, which measured 1 cm. in diameter, overlay the tip of the pancreas. The stomach was empty and contained no blood, but apparently considerable bleeding had taken place into the peritoneal cavity from the gastric perforations. The jejunojejunostomy was not remarkable.

There was a small area of trauma at the extreme tip of the tail of the pancreas. Two small foci of fat necrosis were present in the interlobular adipose tissue of the pancreas. There was no gross evidence that pancreatic juices had entered the peritoneal cavity to any appreciable extent.

On the anterolateral aspect of the lower pole of the left kidney was a 3-cm. stellate laceration.

Gross anatomic diagnosis-(1) Perforating bullet wound of abdomen, with transection of proximal jejunum, healing jejunostomy; severe laceration of transverse colon, colostomy; two large perforations of posterior wall of stomach; moderately severe laceration of left kidney, with retroperitoneal hemorrhage; minimal laceration of tail of pancreas; (2) severe, acute, generalized peritonitis, secondary to unclosed perforations of stomach; and (3) gastric fistula via wound of exit in posterior left flank, secondary to unclosed gastric perforations.

Comment-The technical error of omission in this case emphasizes the necessity for thorough exploration of the abdomen in any combat-incurred wound. Unfortunately, the lesser peritoneal sac was not explored at operation.

Microscopic observations-On the pleural surface of the lung, an exudate of polymorphonuclear leukocytes, blood, and fibrinoid material was noted. The alveolar walls were approximated, and blood or a slight amount of eosinophilic coagulum was present in some alveoli. The vessels were engorged. Alveolar hemorrhage was most apparent subpleurally. A second section showed approximation of the alveolar walls with foci of intra-alveolar fibrinoid material and clumps of polymorphonuclear leukocytes; these changes were moderately extensive in certain areas. Several small bronchioles showed ulceration of the epithelium with blood and polymorphonuclear leukocytes in the lumens. Other bronchioles were partly filled by accumulations of polymorphonuclear leukocytes.

A fibrinoid material with intermingled polymorphonuclear leukocytes was present on the capsular surface of the spleen. The sinusoids were large and filled with blood.

There was vacuolation of the liver cells, chiefly about the central vein of the lobule. The veins appeared to be slightly dilated.

Hemorrhage had occurred into the interlobular fat of the pancreas, and there was a focal collection of fibrinoid material with enmeshed leukocytes.


On the serosal surface of the stomach was an extensive fibrinoid exudate with scattered polymorphonuclear leukocytes. At the attachment adjacent to the serosa, the exudate showed early organization, as evidenced by fibroblastic and capillary proliferation.

The tubules of the kidney showed degeneration to the point of necrosis, with granularity of the cytoplasm and loss of nuclear detail. The glomeruli were small. Another section showed irregular necrosis and hemorrhage into the parenchyma. About the periphery of these areas were moderate accumulations o£ polymorphonuclear leukocytes.

Microscopic diagnosis-(1) Moderate lobular pneumonia; (2) fibrinopurulent pleuritis; (3) severe traumatic renal necrosis and hemorrhage; (4) slight fatty metamorphosis of liver; (5) moderate interlobular pancreatic hemorrhage; and (6) pulmonary atelectasis.

Comment-The microscopic findings were in essential agreement with the gross diagnosis. The peritonitis was well illustrated by the exudate seen on the capsular surface of the spleen and the serosal surfaces of the stomach and pancreas. (E. N. B.)


This case is an example of one of the numerous pitfalls which can occur as the result of failure to carefully explore all abdominal wounds. In spite of the severity of the abdominal wound, the patient tolerated operation surprisingly well. Failure to close two perforations of the stomach, however, led to death on the fourth postoperative day.

Case 15


This patient sustained a shell-fragment wound of the right lumbar region 1 June 1944, at 1510 hours. He reached a field hospital at 1800 hours, after having received 1,000 cc. of plasma. The blood pressure at this time was 86/44 mm. Hg and the pulse 120. Remarkable improvement occurred after the administration of 1,000 cc. of plasma and 500 cc. of whole blood. Just before operation, the blood pressure was 120/88 mm. Hg, and there was no excessive bleeding from the large wound.

At operation, the wound of the right lumbar region was found to be perforating and very large. The missile had entered the abdomen just to the right of the third lumbar vertebra and had caused extensive destruction of the body of this vertebra and of the cauda equina. There was a massive defect in the lumbar muscles on the right, with extreme damage to the cecum, which required resection. An ileotransverse colostomy was performed. The wound was debrided and drained, and a large pack was placed in the retroperitoneal muscle defect. During operation, in spite of continued transfusion of whole blood, the patient's condition became critical. He did not rally after operation and died 2 June 1944.



Autopsy was performed 4 hours after death (Cori, Italy).

Gross observations-There was no apparent invasive infection of the large lumbar wound, which measured 6.0 by 8.0 cm. and which made a gaping defect in the right paravertebral region. The floor of this defect was formed by the posterior portion of the parietal peritoneum. A large retroperitoneal hematoma was present.

The pericardial and pleural cavities were not remarkable. There was no increased fluid in the peritoneal cavity, although the loops of small bowel were loosely bound together by strands of fibrinopurulent exudate.

The heart showed minimal coronary atherosclerosis, with some increased thickness of the left ventricular myocardium.

The lungs showed only moderate congestion and edema of both lower lobes. 

The liver, spleen, and adrenal glands were essentially negative. The gastrointestinal tract showed only a recent healing ileotransverse colostomy.

The kidneys were normal, and the intact right ureter bridged the lumbar defect. There was moderate hydroureter and slight hydronephrotic dilatation of the right renal pelvis. The orifice of the ureter was partly occluded by two blood clots at the level of the wound (third lumbar vertebra). No blood was present in the renal pelvis. The bladder showed somewhat increased trabeculations of the mucosal surface.

Examination of the vertebral column showed extensive destruction of the third lumbar vertebra, with contusion of the filaments of the cauda equina.

Gross anatomic diagnosis.-(1) Severe shell-fragment wound of right posterior abdomen, with multiple lacerations and perforations of cecum and ascending colon, secondary resection of cecum and ileotransverse colostomy; compound comminuted fracture of body and transverse and spinous processes of third lumbar vertebra, with severe retroperitoneal hemorrhage; moderately severe contusion of right ureter, with moderate hemorrhage into ureter and slight secondary hydronephrosis; (2) severe, acute, purulent, generalized, early peritonitis, secondary to wounds of cecum and ascending colon; and (3) severe, multiple contusions and lacerations of cauda equina, secondary to fracture of third lumbar vertebra.

Comment-The massive trauma, blood loss, and severe peritoneal contamination made the outlook in this case very poor from the first. It is surprising that the patient responded as well as he did to 1,000 cc. of plasma and only 500 cc. of blood. In retrospect, it seems probable that he received too much plasma, and, more important, too little blood before operation. 

Microscopic observations.-The lungs showed diffuse partial atelectasis and marked congestion. A few fat droplets were present in the arterial tree, chiefly in the arterioles.

The kidney showed a focus of interstitial and intratubular hemorrhage consistent with a traumatic lesion. Scattered through the cortex were spaces lined with very flat epithelium or swollen endothelium which contained orange-


staining material. No pigment precipitates were found in any part of the tissues which could be recognized definitely as a nephron. The nature of the lesion was not apparent, but it was not regarded as important to clinicopathologic correlation.

Microscopic diagnosis-Minimal pulmonary fat embolism. (T. B. M.)


It is remarkable, in retrospect, that this patient lived as long as he did, because of (1) extreme tissue destruction and consequent loss of blood, (2) the difficulty of controlling blood loss during operation, and (3) the overwhelming contamination of the peritoneal cavity. Early postoperative deaths were not uncommonly seen in severely wounded patients after such a long and tedious operation as this patient underwent.