Battle Casualties in Korea: Studies of the Surgical Research Team, Volume IV
Renal Changes in Patients Dying of War Wounds: Korea,1950-1952
A Preliminary Report
Captain Thorwald R. Anderson, MC, USAR
Lieutenant Colonel Arthur Steer, MC, USA
Shortly after the onset of war in Korea there was concern lest the use of group "O" blood for all transfusions in the combat zone should cause serious reactions and a high incidence of acute renal failure.* When, after the first 6 months of combat, it became apparent that this fear was exaggerated, study of autopsy material was continued in an effort to learn more about the relation between fatal battle wounds and renal complications. The results of these studies were published in the annual reports of the 406th Medical General Laboratory1, 2, 3 and played a part in indicating the need for and justifying the establishment in 1952 of a renal failure center in Korea. Subsequently, publication of the study of the physiologic effects of wounds in the Mediterranean Theater in World War II4 showed that the experience in Korea was a confirmation of previous experience. It is probable that acute renal failure had always been an important complication of severe battle trauma, but had received little notice masked as it was by the direct effects of trauma and the attendant shock and infection. Of 324 autopsies performed on men who died after receiving wounds in battle in Korea, 111 showed the histologic changes of acute renal failure.
Materials and Methods
During the period 1 July 1950 to 1 January 1953, the protocols and tissuesof autopsies on 324 battle casualties who died in military medical installationsin the Far East were studied by the Pathology Department of the 406th MedicalGeneral Laboratory in Tokyo. Autopsies on patients who died of battle woundswere irregularly performed in the surgical hospitals in Korea and generallyin 1950 and 1951 only
*In this report acute renal failure, acute renal insufficiency, lower nephron nephrosis, hemoglobinuric nephrosis and acute tubular necrosis are considered synonymous in that they refer to a similar pathologic process.
236
when death was unexpected or unexplained. Autopsies were regularly performedon battle casualties who died in evacuation hospitals in Korea and in thehospitals in Japan. There was a considerable increase in the percentageof autopsies performed on men who died of wounds in hospitals in Koreaas the battle situation became stabilized. In 1950 the ratio of autopsiesto deaths from wounds was unknown but quite low; in 1951 it was approximately3 per cent and in 1952 it was approximately 25 per cent. In 1950 and thefirst few months of 1951, a policy of rapid evacuation of patients to hospitalsin Japan was in effect and much of the information obtained during thistime came from these hospitals.
As the percentage of autopsies increased in the hospitals in Korea,protocols and clinical histories were given in greater detail and probablywith greater accuracy. In part, this reflected the static battle situationwhich permitted more thorough study of patients and, in part, also reflectedthe influence of the various research teams in Korea which stimulated investigationalong many lines.
Table 1. Regional Distribution of War Wounds in 324 Autopsies
Region | 1950 | 1951 | 1952 | Total |
Head, neck, or spine | 44 | 22 | 9 | 75 |
Thorax | 4 | 6 | 11 | 21 |
Abdomen | 19 | 10 | 13 | 42 |
Abdomen and thorax | 12 | 8 | 11 | 31 |
Extremities | 18 | 17 | 19 | 54 |
Abdomen and extremities | 9 | 9 | 17 | 35 |
Multiple | 19 | 18 | *29 | 66 |
| Total | 125 | 90 | 109 | 324 |
*Including 6 with severe burns.
An attempt to determine whether the regional distribution of wounds(Table 1) was similar to that reported for other wars5 failedbecause the present autopsies, yielding more accurate information, frequentlyimplicated multiple regions of injury while previous surveys, based onsimple inspection and information giving only the presumed major woundof entrance, reported only a single region of injury.
Autopsy material from men who died of battle wounds was received fromthe 1st Medical Field Laboratory in Korea and from all hospitals in Koreaand Japan. A special search was made in all cases for evidence of renalchanges using paraffin sections stained with
237
hematoxylin and eosin. Selected cases were further studied using thefollowing stains: fat, Schiff periodic acid, Feulgen, Heidenhain, LaPehne,Dunn-Okajima, and Wilder reticulum. The protocols of the 111 autopsieswith histologic evidence of renal failure were studied to determine whatpart blood transfusion, shock, severity of injury, site of injury, andinfection played in the pathogenesis of the renal lesions.
Clinical Course
The relation between severe trauma and acute renal failure was obscureand required cautious interpretation of clinical findings. The severelyinjured patient frequently showed clinical evidence of renal failure whichwas transient and disappeared with recovery from shock. Persistence ofclinical findings was therefore required to establish the clinical diagnosisof acute renal failure. Consequently, diagnosis became difficult in patientsso severely injured that death occurred within a few days after injury.Often in such patients there was little, if any, clinical evidence thatdeath was hastened or made inevitable by superimposed renal failure althoughrecognizable renal changes were found at autopsy. Further, evidence ofnephropathy was found so frequently in patients who died of war woundsthat, in some cases, it could be considered, like pulmonary edema and septicsplenitis, a system or organ failure associated with the delayed deathpeculiar to trauma and the battle environment.
In the study on the physiologic effects of wounds in World War II4it was concluded "that the clinical syndrome of renal insufficiency whichfollows shock is remarkable chiefly for the scarcity and mildness of itssymptoms" and that "the symptom complex of oliguria, pigment excretion,azotemia and hypertension established the diagnosis." Search for such evidenceis not always practical under battle conditions, and therefore, the pathologist'sreport of a renal complication sometimes surprised the clinician. Reviewof the protocols showed in retrospect that especially in 1950 the diagnosismight have been made more frequently (Tables 2 and 3) but not in all cases.Some of the patients died of their wounds before clinical evidence of renalfailure developed, while in others the renal changes were probably a terminalevent.
The severity of the azotemia varied considerably, and frequently wasnot sufficient of itself to establish a diagnosis of renal failure. Sincerapid tissue breakdown and destruction could also produce azotemia, anante-mortem diagnosis of acute renal failure usually was not made withoutother evidence. Similarly, oliguria was usually present
238
Table 2. Recorded Ante-mortem Evidence of Renal Failure in 111 Patientswith Renal Changes at Autopsy
Ante-mortem Findings | 1950 | 1951 | 1952 | Total |
Azotemia | 0 | 2 | 1 | 3 |
Oliguria | 4 | 6 | 6 | 16 |
Oliguria and azotemia | 12 | 9 | 21 | 42 |
Oliguria and hypertension | 1 | 0 | 0 | 1 |
Uremia | 12 | 3 | 2 | 17 |
Oliguria, azotemia and hypertension | 1 | 2 | 3 | 6 |
Total with ante-mortem findings | 30 | 22 | 33 | 85 |
Total with renal changes at autopsy | 43 | 35 | 33 | 111 |
Table 3. Clinical Diagnoses of Renal Failure in 111 Patients withRenal Changes at Autopsy
Clinical Diagnosis | 1950 | 1951 | 1952 | Total |
Lower nephron nephrosis | 12 | 15 | 27 | 54 |
Possible lower nephron nephrosis | 1 | 2 | 0 | 3 |
Uremia | 3 | 3 | 2 | 8 |
| Total with diagnoses of renal failure | 16 | 20 | 29 | 65 |
| Total with renal changes at autopsy | 43 | 35 | 33 | 111 |
during shock and in patients living only a few days this was not ofgreat aid in establishing a diagnosis of acute renal failure.
A clinical diagnosis of acute renal failure was made more frequentlyin 1952. Many of these patients were treated at the Renal Failure Centerbut there was also a greater awareness among medical officers in otherhospitals of the possibility of renal complications and a greater effortwas made to treat this complication.6
Possible Etiologic Factors
Blood Transfusion. During the period of this report, more than200,000 bottles of group "O" blood were sent to Korea, most of which wasused for the treatment of the battle wounded. Blood labeled "Low Titer"could be given to all patients and blood labeled "High Titer"
239
was intended for group "O" patients only. It was neither practical norpossible to cross-match patients in forward hospitals or to determine theRh type of the recipients. Very few reports were received of transfusionreactions. In only 4 of the 324 autopsy procotols was such a reaction reportedand in only 2 of these was the reaction considered of sufficient severityto warrant naming transfusion reaction as a contributory cause of death.However, all four patients showed typical renal changes at autopsy.
Progressively larger amounts of blood were used in the treatment ofbattle casualties as the war continued. This was true also for the amountof blood given to the battle casualties who were found to have renal changesat autopsy (Table 4). These patients received more blood than the averageof approximately 4.3 bottles of blood given to all battle casualties whowere transfused,7 indicating that these patients were amongthe more seriously injured. The studies of the Surgical Research Team indicatedthat serious transfusion reactions due to the use of group "O" blood wererare.8
Table 4. Blood Transfusions to Patients Who Died of War Wounds andHad Histologic Evidence of Renal Changes-87 Autopsies
| 1950 | 1951 | 1952 | Total |
Number of patients | 35 | 22 | 30 | 87 |
Total bottles of blood | 203 | 198 | 360 | 761 |
Average-bottles per patient | 5.8 | 9 | 12 | 8.7 |
Extremes-bottles per patient | 1-52 | 2-28 | 2-40 |
|
Battle Trauma. Although it was desirable and perhaps theoreticallypossible to distinguish the renal effects of traumatic shock, tissue damageand infection, this could not be accomplished. Many variables influencedand determined the response to therapy and the occurrence of acute renalfailure. These included the degree and duration of shock, the intervalbetween injury and treatment, the site and severity of injury, the presenceof infection, the climate and weather, the battle situation, the numberof casualties requiring treatment at the same time, etc. Information concerningmany of these factors was either fragmentary or unavailable.
Shock. Because of the large amount of blood given to the patientswho died and showed renal changes at autopsy, it was assumed that mosthad suffered shock of a significant degree and duration. The protocolsof 19 patients who died in 1952 stated that traumatic shock
240
had been present in all, with a duration of 11/2to 20 hours and an average duration of slightly over 3 hours. The averageinterval between injury and surgery was 7 hours. For the other cases, onlyan estimate of the severity of shock could be made. This was based on theblood pressure when recorded, the pulse rate, the hematocrit, the typeand severity of injury, the interval between injury and treatment, andthe amount of blood transfused. Using such estimates, of 75 patients whocould be evaluated, 58 had suffered severe shock probably for a considerableduration and 17 had suffered moderate shock.
Severity of Injury. The autopsy protocols of the 111 men showedrenal changes were reviewed to determine the severity of their wounds.Twenty-five were judged to have wounds of moderate severity and 86 wereclassified as severely wounded. Those with wounds considered to be moderatewould ordinarily have been expected to respond to modern forms of treatmentif some other factor or combination of factors, such as prolonged shock,infection, gangrene, renal failure, secondary hemorrhage, had not complicatedthe course of recovery. Those judged to have severe battle wounds had aguarded or poor prognosis because of the amount and type of tissue damagedby the missile or because the wounds involved vital organs. A similar estimateof the patients who did not have renal changes at autopsy is not available.
Site of Injury. Study of the protocols of the 324 patients whodied after receiving wounds in battle indicated that there was a relationbetween the site of injury and the ocurrence of renal failure (Table 5).Two groups were compared in an attempt to explain this. The first groupconsisted of all patients who died of wounds of the head,
Table 5. Regional Distribution of Fatal Wounds and Incidence of AutopsyEvidence of Acute Renal Failure
Region Wounded | Number Wounded | Number with Renal Changes | Per Cent with Renal Changes | |||
1950 | 1951 | 1952 | Average | |||
Head, neck or spine | 75 | 4 | 5 | 9 | 0 | 5 |
Thorax | 21 | 0 | 0 | 0 | 0 | 0 |
Abdomen | 42 | 26 | 63 | 50 | 69 | 62 |
Abdomen and thorax | 31 | 12 | 57 | 25 | 27 | 39 |
Extremities | 54 | 28 | 61 | 65 | 32 | 52 |
Abdomen and extremities | 35 | 13 | 44 | 56 | 24 | 37 |
Multiple | 66 | 28 | 37 | 56 | 38 | 42 |
| Total | 324 | 111 | 34 | 39 | 30 | 34 |
241
neck, spine or thorax. Few of these patients had histologic changesof acute renal failure at autopsy although many were exposed to the variousfactors accepted as possible causes of acute renal failure. Patients withhead wounds usually suffered little shock and often did not receive bloodtransfusions but patients with thoracic wounds did suffer shock and weretransfused. Infection occurred in all types of wounds. The second groupconsisted of all patients who died of wounds of the abdomen or extremities.This group had a high incidence of acute renal failure. Abdominal woundsgenerally had a relatively high mortality rate while extremity wounds,as a group, had a low mortality rate. Both severe extremity wounds andabdominal wounds were associated with shock, large blood transfusions andinfections. It was not apparent why a majority of patients who died afterreceiving wounds of the abdomen or extremities showed autopsy evidenceof acute renal failure while such evidence could not be found in any patientwho had a fatal wound involving the thorax alone.
Infection. Infection present at the time of autopsy was difficultto evaluate as a factor in the pathogenesis of the renal changes. It appearedthat infection frequently progressed because of lowered resistance in partdue to renal failure but one could not be certain that infection had playeda role in causing the renal failure or in making the renal changes moresevere. A common clinical observation at the Renal Failure Center6was that the blood nonprotein nitrogen level frequently rose more rapidlyafter dialysis than would be anticipated and that often this unexpectedrise was an indication of the presence of unrecognized infection. Studyof 24 patients who were treated at the Center and who had renal changesat autopsy (Table 6) indicated that patients with infections tended tohave a high blood nonprotein nitrogen level.
Table 6. Relation Between Azotemia and Infection
| Highest Recorded Blood Nonprotein Nitrogen Level | ||||
Finding at autopsy | 35-100 | 100-200 | 200-300 | 300-400 | 400-500 |
Infection present | 2 | 4 | 3 | 9 | 1 |
Infection absent | 0 | 3 | 1 | 0 | 1 |
Severe infection was frequently present and often involved more thanone region (Table 7). Peritonitis and cellulitis were the most frequentfindings. The incidence and severity of infections found at autopsy ofpatients who did not have renal lesions are not available.
242
Table 7.-Presence of Infection at Autopsy
| 1950 | 1951 | 1952 | Total |
No. of patients with renal lesions | 43 | 35 | 33 | 111 |
No. of patients with infections | ? | 27 | 23 | ? |
| Peritonitis | 28 | 14 | 15 | 56 |
| Cellulitis | 12 | 8 | 10 | 30 |
| Pleuritis | 11 | 2 | 5 | 18 |
| Pericarditis | 2 | 0 | 0 | 2 |
| Brain abscess | 0 | 2 | 0 | 2 |
| Meningitis | 0 | 1 | 0 | 1 |
| Pyelonephritis | 0 | 0 | 4 | 4 |
Pathology
Although certain features of interest were observed in the study ofthese cases, little was added to our knowledge of the pathogenesis of orthe actual site of injury in this condition variously known as lower nephronnephrosis,9 hemoglobinuric nephrosis,10 acute tubularnecrosis,11 acute renal failure,12 etc. All of thecases met the requirements for histologic diagnosis although there wasconsiderable variation in the severity of the lesions.
Weight of Kidneys. In most instances the kidneys were heavierthan normal and in the few patients who had one kidney removed surgically,the remaining kidney showed hypertrophy. The weight of both kidneys, andof single remaining kidneys, after nephrectomy ranged from 200 to 725 gramswith a median weight of 450 grams. The kidneys of patients who lived morethan 13 days after wounding usually weighed more than 600 grams and thoseof patients who lived less than 10 days after wounding usually weighedless than 400 grams.
Histologic Observations. Necrosis of the epithelial cells ofthe tubules was a constant finding but no consistent relation could berecognized between the amount of tubular necrosis and the severity of injury,duration of life after injury, severity of renal failure, number of pigmentedcasts, or number and size of foci of inflammation.
Pigmented casts were also frequently present but no relation could befound between the number of casts in the cortex or the medulla and theduration of life after wounding in 31 of the patients who died in 1952.Casts were found in larger number in the medulla than in the cortex.
Foci of interstitial inflammation in the kidney were absent to minimalin patients who died within 5 days after injury, were numerous
243
and extensive in patients who lived from 6 days to 2 weeks after beingwounded and tended to be fewer but larger in patients living more than2 weeks after injury (Table 8). The infiltrates were found predominantlyin the region of the cortico-medullary junction but when renal failurewas severe and of long duration they were also found in the cortex aboutnecrotic distal convoluted tubules. Because of the epithelial degenerationand the cellular reaction, involvement of the proximal portion of the nephroncould not be recognized with certainty.
Table 8. Severity of Interstitial Infiltration and Interval BetweenInjury and Death
Interval Between Injury and Death (days) | Number of Patients | Severity of Interstitial Infiltration | |||
Minimal | Mild | Moderate | Severe | ||
3-5 | 10 | 8 | 1 | 1 | 0 |
6-10 | 8 | 3 | 2 | 2 | 1 |
11-15 | 8 | 1 | 2 | 2 | 3 |
16-19 | 5 | 1 | 2 | 1 | 1 |
Mallory4 reported significant fat droplet alterations inthe epithelial cells of renal tubules in men dying 18 to 96 hours afterinjury and traumatic shock. He interpreted these changes as being morphologicevidence of parenchymatous injury and functional impairment. A search wasmade for similar changes in the kidneys of 23 of the patients who diedin 1952 and showed renal changes at autopsy. Only one of these men diedin the critical 18- to 96-hour interval after injury. Fat droplets measuringone to several micra in diameter were observed in many of the kidneys.They were found most frequently in the ascending tubules and less frequentlyin distal convoluted and collecting tubules. They were most prominent insections from three patients who lived less than 6 days after injury buton the other hand were minimal in sections from three other patients wholived less than 6 days after injury. Control sections from patients whodied a few hours after injury contained more fat droplets than this lastgroup of three patients.
Tissues from the 23 patients studied for fat droplet changes were alsostudied for the presence of iron in the tubules. Controls for this studywere provided by producing athrocytosis in rats13 by the injectionof human hemoglobin. Within 2 days after such injections, gran-
244
ules appeared in the epithelial cells of the proximal portion of thenephron which had the same affinity as erythrocytes for the LaPehne andDunn-Okajima stains. In none of the 23 human patients was there similarevidence of demonstrable iron or hemoglobin in the epithelial cells ofthe tubules. However, in sections from patients who had a short intervalbetween injury and death, the pigmented casts in the proximal portion ofthe nephron showed some affinity for the stain. Similar casts in the distalportion of the nephron took the stain less intensely and those in the collectingtubules generally did not react at all. The pigment in the casts was notunequivocally identified as hemoglobin.
Discussion
It has been shown that the changes of acute renal failure are frequentlyfound at autopsy in patients dying after being wounded in battle and thatthis is related, at least in part, to the degree and duration of shock,the region wounded and the severity of the wound. In addition to savingmany lives, modern treatment of battle wounds delayed the death of theseverely wounded and permitted the development and recognition of complicationssuch as acute renal failure. In a certain number the renal changes wereprobably an indication of the failure of one system during the processof slow death after trauma. It was difficult to evaluate the effect ofthe renal changes in some of these cases because death occurred beforeclinical signs and symptoms could be recognized, and because in othersthe process was probably so mild that recovery from renal failure couldhave occurred were it not for the injuries and infection. However, themajority of patients did succumb to the combination of trauma, infectionand acute renal failure. This combination occurs more frequently undermilitary situations than in civilian life because the factors associatedwith the development of renal failure occur more frequently under battleconditions.
The diagnosis of acute renal failure was not always evident but whenclinicians became aware of the frequency of occurrence of this complicationthey were much more successful in recognizing its presence. In 1950, 37per cent of the cases were diagnosed and in 1952, 88 per cent were recognized.With increasing accuracy of diagnosis, more patients were treated morevigorously for renal failure with a saving of some lives. In war the severelywounded will continue to develop acute renal failure and it is likely thatsuccessful methods of prevention and treatment offer an opportunity forfurther reduction in the mortality rate from war wounds.
245
Conclusions
1. Approximately a third of men dying from war wounds in medical installationsshow histologic evidence of the changes associated with acute renal failure.
2. Predisposing factors for the development of these changes includesite and severity of injured and duration and severity of shock.
3. The use of group "O" blood for all transfusions did not appear tobe an important factor in causing acute renal failure.
4. Wounds of the abdomen and extremities were associated with a highincidence of renal changes and wounds of the head, neck, spine or thoraxwith a low incidence of renal changes.
5. Progressive and severe infection was frequently present in patientswho showed renal changes at autopsy.
6. Diagnosis of acute renal failure required a high index of clinicalsuspicion.
References
1. Annual Historical Report, 406 Medical General Laboratory,1950. APO 500, c/o P.M., San Francisco, California.
2. Ibid., 1951.
3. Ibid., 1952.
4. Board for the Study of the Severely Wounded, NorthAfrican-Mediterranean Theater of Operations: Surgery in World War II. Med.Dept., U. S. Army, Washington, D. C., 1952, U. S. Government Printing Office.
5. Beebe, G. W., and DeBakey, M. E.: Battle Casualties,Incidence, Mortality and Logistic Considerations. Charles C. Thomas, Springfield,Illinois, 1952.
6. Meroney, W. H., and Herndon, R. F.: The Managementof Acute Renal Insufficiency. J.A.M.A. 155 (10): 877, 1954.
7. Steer, A., and Hullinghorst, R.: The Blood Programin the Korean War.
8. Crosby, W. H., and Akeroyd, J. H.: Some ImmunohematologicResults of Large Transfusions of Group O Blood in Recipients of Other BloodGroups; a Study of Battle Casualties in Korea. J. Hematology 9:103, 1954.
9. Lucké, B.: Lower Nephron Nephrosis (renal lesionsof the crush syndrome, of burns, transfusions and other conditions affectinglower segment of nephrons). Mil. Surgeon 99: 371, 1946.
10. Mallory, T. B.: Hemoglobinuric Nephrosis in TraumaticShock. Am. J. Clin. Path. 17: 427, 1947.
11. Oliver, J.: Correlations of Structure and Functionand Mechanisms of Recovery in Acute Tubular Necrosis. Am. J. Med. 15:535, 1953.
12. Smith, H.: The Kidney; Structure and Function in Healthand Disease. Oxford University Press, New York, 1951.
13. Rather, L. J.: Renal Athrocytosis and IntracellularDigestion of Intraperitoneally Injected Hemoglobin in Rats. J. Exper. Med.87:163, 1948.
