Contents

AN EARLY CRUSH CASE. This fresco in the Basilica of San Miniato al Monte, Florence, Italy, by Spinello Aretino, 1387 A.D., depicting construction of the Abbey at Monte Cassino, is perhaps the earliest record of treatment of crush injury. The devil has pushed a wall onto a monk who is then revived by St. Benedict. "Our ministrations were less effective."

CHAPTER XI

The Crush Syndrome in Battle Casualties

In preceding chapters examples of the crush syndrome have been included, frequently with little or no distinction from the more usual types of battle casualties. This group of cases, however, presents certain distinctive features which require separate consideration. Moreover, though cases were reported in the German literature among casualties of World War I, and though Bywaters¹ and others^{2 3} have made careful clinical and biochemic studies of many civilian casualties during the "Battle of Britain," few thoroughly studied cases have been recorded among the battle casualties of World War II.

In this chapter the clinical, physiologic, biochemic, and pathologic features and the results of treatment will be discussed on the basis of nine cases, five of which were fatal. All of the patients were crushed within stone houses typical of the Italian countryside. (It is perhaps not a coincidence that the first known description of the syndrome came from Italy in 1908 as a report of casualties of the earthquake at Messina.) Five of the buildings in which the injuries occurred, one of them housing a medical collecting company, were struck by artillery shells. Two of the casualties resulted when a demolition charge exploded and the remaining two were incurred when a house about two hundred yards from our laboratory was struck by a bomb from a German raider at dawn. It was consequently possible to observe the patients virtually from the moment of release from the rubble in which they were partially buried. All nine are listed, together with pertinent data, in Table 95.

¹BYWATERS, E. G. L.: Ischemic muscle necrosis; crushing injury, traumatic edema, crush syndrome, traumatic anuria, compression syndrome: a type of injury seen in air raid casualties following burial beneath débris. J.A.M.A. 124: 1103-1109, April 15, 1944.
²DOUGLAS, J. W. B.: Incidence of signs of renal injury following prolonged burial under débris in an unselected series of 764 air-raid casualties admitted to hospital. Brit. J. Urol. 17: 142-147, December 1945.
³DUNN, J. S.; GILLESPIE, M., and NIVEN, J. S. F.: Renal lesions in 2 cases of crush syndrome. Lancet 2: 549-552, November 8, 1941.

256

TABLE 95.-DATA CONCERNING 9 PATIENTS WITH CRUSH SYNDROME

257

Clinical Features

The clinical features of crush syndrome can be divided into two categories: primary and secondary. The primary features, comprising the direct effects of crushing injury, were common to all nine cases. The secondary features, seen only in the five fatal cases, were the usual signs and symptoms of posttraumatic renal insufficiency described in Chapters IV and V.

Primary Phase

Shock.-Shock, as judged by the usual clinical criteria, was not characteristic of the majority of.our patients with the crush syndrome and was never seen in patients observed within 10 hours of release from the compressing agents. One patient (Case 70), for instance, looked so well when examined 30 minutes after release that he was evacuated to the rear without treatment. He nevertheless died 4½ days later. Only one patient (Case 93) was in shock when first examined and this was 13 hours after release. He had had no previous medical attention. A diagnosis of slight shock was made in Case 78 twenty hours after release, and though diagnoses of shock were never made in Cases 69 and 132, transitory episodes of impalpable pulse and unmeasurable blood pressure were recorded in the case histories (at 19 hours in Case 69 and at 14 and 23 hours in Case 132). In Case 70 and in all the surviving patients, hypotension was never observed.

Nausea and Vomiting.-Nausea or vomiting was recorded in only three cases, all of which terminated fatally. Nausea with hiccuping was observed in one patient (Case 70). This occurred on the first and second days after release when the nonprotein nitrogen concentration was 83 to 104 mg. per 100 cc. of plasma. Vomiting was recorded twice. One patient (Case 93) vomited while he was still pinned down by masonry but not afterward. The second patient (Case 78) had troublesome vomiting on the second and third days after release from compression. On those days the nonprotein nitrogen of the plasma was 92 mg. per 100 cc. and occasionally higher.

Local Changes.-Commonly the appearance of the skin gave no indication of the changes in the underlying muscle. In some patients, however, there were bruises and areas of ecchymosis and lividity of the overlying skin (Fig. 23). Immediately after release from compression the affected parts began to swell. The swelling took place first within the fascial compartments and resembled

258

FIGURE 23. Bruising, ecchymosis, and swelling of extremity (Case 78).

the swelling of the calf muscles following interruption of the popliteal artery. The affected areas became hard, tense, and brawny. Only later, usually after the third day, did soft, pitting subcutaneous edema appear. Distribution of lesions was variable. The extremities were most often involved, particularly the legs, and the muscles of the buttock were frequently included. In one patient (Case 132) the muscles of the shoulder and neck were most affected, and edema of the neck was so great that a tracheotomy proved necessary.

During the first few days of the syndrome, the patients might have no complaints or they might complain of stiffness, soreness, aching, and tenderness in the involved areas. Symptoms were often inversely proportional to the severity of injury. In relatively mild lesions, pain was elicited by pressure on the affected area; more severe lesions were insensitive to pressure. Some patients complained of numbness of the skin of the toes and over the involved muscles. Examination revealed patchy areas of hypesthesia or anesthesia of the skin, a "stocking" type of anesthesia of the limbs, and absent reflexes. Involved muscle groups were paralyzed or sometimes temporarily spastic. It was sometimes impossible to feel pulsations in the arteries of the affected limbs, reminding one of the early phase of Volkmann's contracture. In one patient (Case 78) a fasciotomy was done for that reason. The leg was so insensitive that it was possible to perform the operation without anesthesia.

Fractures.-Rather surprisingly, in view of the nature of the initial trauma,

259

no major fractures were observed. The only fracture recorded (of the zygoma) could be considered only coincidental.

Secondary Phase

One patient (Case 132) died 56 hours after release from compression; the other four lived from 4 to 13 days after release. One died suddenly, apparently of acute cardiac failure, the others more slowly, with progressive lethargy and drowsiness but no typical uremic coma. Drowsiness was particularly marked in one patient (Case 78) who had been treated with magnesium. None of the patients in this group had convulsions at any time during their illness. The mechanism of death appeared to be pulmonary edema, except in one patient (Case 93) in whom lobar pneumonia developed on the thirteenth day shortly after the appearance of a diuresis had led to hope for his recovery.

All of the patients who died showed the usual signs and symptoms of a lower nephron nephrosis previously described (Chapters IV and V). The initial urine specimens were deeply pigmented and reacted positively with the benzidine test; oliguria was marked and, except in Cases 69 and 93, progressive (see Table 97). The urine, with one exception, was persistently acid and the specific gravity was constantly low from the third day onward. The nonprotein nitrogen level in the plasma rose progressively, and on the third or fourth day mild hypertension developed. Details of the chemical abnormalities of the blood and urine follow.

Physiologic and Biochemic Features

Blood

Blood Volume.-The blood volume was determined immediately upon arrival at the hospital in only two patients with crush injury. The first (Case 93) was the only patient of the group thought to be in severe shock. He had lost one-third of his total blood volume, apparently entirely in the form of plasma, for he showed no evidence of diminution of the total hemoglobin or red-cell mass. The hemoconcentration might have been partly responsible for the clinical picture of shock, as severe shock was generally associated with a loss of circulating blood greater than one-third of the normal volume. The second patient

260

CHART 39. HEMATOCRIT VALUES* IN FATAL CRUSH SYNDROME

(Case 124) showed no evidence of shock clinically and no decrease was found in the volume of circulating blood.

In three patients (Cases 69, 78, and 93) the blood volume was determined during their course. It was abnormal in only one of them (Case 78), being increased by about two liters 24 hours after the patient's release from compression. During that period he had received 600 cc. of plasma (2 units), 400 cc. of 4-percent sodium bicarbonate, and 1,000 cc. of 5-percent dextrose in isotonic saline solution intravenously. He had taken 800 cc. of liquids by mouth and had voided 350 cc. of urine. He died 3 days later with severe pulmonary edema. In Case 93, the patient received no more fluid intravenously than was necessary to correct his blood volume deficit. He died of lobar pneumonia 13 days after release. The blood and plasma volumes in Case 69 were within normal limits on the second and ninth days. In retrospect, from rough fluid-balance calculations, this patient was given perhaps 2,000 cc. more parenteral fluid, including large doses of sodium bicarbonate, than optimal. The fact that severe peripheral edema developed indicates that total body water, if not circulating blood volume, was probably increased.

261

CHART 40. HEMATOCRIT VALUES* IN NONFATAL CRUSH SYNDROME
 Hematocrit Value and Plasma Proteins.-The most striking feature of the immediate findings in the blood of patients with the crush syndrome was the hemoconcentration in those who subsequently died. It was chiefly among these patients with the crush syndrome that hemoconcentration was found in our series of seriously wounded. In one patient (Case 70) the hematocrit value was 80 percent 9 hours after his release from the crushing force. It is noteworthy that the clinical evaluation of this man failed to reveal evidence of shock. All of our patients with crush syndrome who died had initial hemoconcentration (Chart 39 and Table 96). It was, however, transitory and the hematocrit value was within normal limits by the end of the second day. None of those who survived had hematocrit levels above normal limits⁴ (Chart 40 and Table 96).

Plasma proteins were within normal limits⁵ in six of the eight patients on whom early determinations were made. One initial determination in a patient who lived and four determinations made in the first 24 hours after release

⁴Normal: 47. Range: 42 to 52 percent.
⁵Normal: 6.5. Range: 6 to 7 Gm. per 100 cc.

262

TABLE 96.-HEMATOCRIT VALUE IN RELATION TO PATIENTS' COURSE

from compression in a fatal case were moderately elevated (Table 97). Again, as was the case in the hematocrit values, the elevated plasma protein concentration soon returned to normal. Although our data are too few to permit of interpretation, it is probable that a consistently high plasma protein concentration in the first 24 hours is indicative of a poor prognosis.

In these crush cases the initial concentration of red blood cells was relatively greater than the concentration of plasma protein, as may be seen in Chart 41. These findings suggest that with release of compression and re-establishment of circulation to the injured areas, there is prompt outpouring of plasma into

263

TABLE 97.-PLASMA PROTEIN CONCENTRATION IN RELATION TO PATIENTS' COURSE

the affected tissues, manifested clinically by the rapid development of swelling in the affected parts. With the loss of plasma from the circulating blood, the concentration of the red blood cells is increased. If, as is likely, the plasma lost into the tissues contains less protein than normal plasma, an explanation of the early increases in the concentration of plasma proteins in some severe cases is afforded. The rapid disappearance of the hemoconcentration (and of the elevated plasma protein level in two instances) is in part accounted for by the intravenous administration of plasma and crystalloids, but this is not the only factor, for the dilution occurred in the absence of intravenous therapy. The effect could be explained by the reabsorption of extravasated plasma.

During the later stages of the syndrome in the fatal cases, the hematocrit

264

CHART 41. RELATIONSHIP BETWEEN PLASMA PROTEIN AND HEMATOCRIT LEVELS IN CRUSH SYNDROME

value remained within low normal limits except in Cases 78 and 93 in which it fell below 40, eventually dropping to 21.9 on the day of death in the latter case. There was greater variability in the concentration of the plasma proteins during the course of the disease. Two determinations in patients who survived and seven determinations in patients who died dropped below 6 Gm. per 100 cc., three of them being below 5.5 (Table 97).

Plasma Nonprotein Nitrogen.-Among the patients who survived, concentration of the nonprotein nitrogen in the plasma was essentially normal throughout the course of the illness. Among those who died, it was elevated from the start and rose progressively at an almost linear rate (Chart 42). No determinations were made in one fatal case. The concentration of phosphorus in the plasma rose more slowly and appeared to reach its asymptote earlier.

Plasma Chlorides.-The plasma chloride concentrations were essentially normal among the patients who survived. Among those in whom fatal crush

265

CHART 42. PLASMA NONPROTEIN NITROGEN CONCENTRATION IN CRUSH SYNDROME*
 syndrome developed, however, there was a progressive drop in the concentration of plasma chlorides except in one patient (Case 93) who developed a diuresis and showed evidence of improving renal function although he subsequently died (Chart 43). At necropsy lower nephron nephrosis and lobar pneumonia were demonstrated. No determinations were made in one fatal case. Concomitant with the fall in plasma chloride level, there was generally a fall in the alkali reserve of the blood as suggested by the lowering of its CO₂ combining power. These changes in plasma carbon-dioxide combining power and chloride levels were entirely comparable to those seen in renal insufficiency following other types of trauma (Chapter V).

Plasma Pigment.-Blood specimens taken shortly after release from compression revealed abnormal quantities of a free pigment in the plasma which reacted positively to the benzidine test. Blood samples were collected in carefully dried, but not oiled, syringes and needles. With this technique, a free plasma "hemoglobin" concentration up to 15 mg. per 100 cc. was considered within the normal range. Chart 44 shows the values obtained in the first 5 days after injury both in the patients who died and in those who survived. All but

266

CHART 43. PLASMA CHLORIDE CONCENTRATION* IN CRUSH SYNDROME
two of the determinations we were able to obtain during the first 24 hours were above the limits accepted as normal; the exceptions were determinations made in Case 99, nine and sixteen hours after release. Only one of the patients showed a heme concentration of over 40 mg. per 100 cubic centimeters. Data were available on eight cases. Of particular interest, from those data, is that, like the return of the hematocrit level and the plasma protein concentration to normal, the concentration of the free benzidine-positive substance in the plasma tended to be within normal limits by the end of the second day. As in the case of the hemoconcentration, then, the presence of free benzidine-positive substance in the plasma might have been missed if specimens had not been collected soon after the patients' release from the crushing force.

The nature of this substance, free in the plasma and giving the reaction of a heme, could not be settled. The quantities present were not large enough to make spectroscopic analysis possible in the instrument available to us. Chemical examination of the urine of these patients (Chapter VIII) showed that both hemoglobin and myoglobin were excreted in large amounts, the proportions of each varying widely from case to case. It is reasonable to assume, therefore, that both pigments were represented in the plasma as well.

267

CHART 44. FREE BENZIDINE-POSITIVE SUBSTANCE IN PLASMA IN CRUSH SYNDROME*
 Urine

Gross abnormalities were found only in the urine specimens collected soon after release from the compression force. Table 98 indicates the duration of some of the abnormalities in the urine. It will be observed that myoglobin was not found in the urine more than a day and a half after release from the crushing force. Abnormality in the color of the urine lasted somewhat longer (up to 3 days), and albuminuria to a significant degree was detected for as long as 9 days after release from the crushing influence.

Output.-None of the patients who survived the injury had any diminution in volume of urinary output below the limits of normal. In all of those who died (Table 99) urinary output was diminished from the start. All had oliguria. In one instance (Case 70) anuria (an output of less than 100 cc.) was recorded during the 24 hours beginning the day after release. Two other patients were oliguric until the second 24-hour period, when the output dropped below 100 cc. a day. The remaining two patients (Cases 93 and 78) were oliguric but never became anuric. In one (Case 69) diuresis had just begun to develop when he suddenly died; in another (Case 93, Table 99) an excellent diuresis was estab-

268

lished and the patient might have survived the injury had he not died of lobar pneumonia.

TABLE 98.-DURATION (DAYS) OF ABNORMALITIES IN THE URINE OF 9 PATIENTS WITH CRUSH SYNDROME
 TABLE 99.-URINARY OUTPUT IN 5 FATAL CASES OF CRUSH SYNDROME

269

Color.-Yellow and straw color were considered to be the normal colors of urine. An amber color was considered on the borderline of abnormality. All of the patients who died had abnormal urinary color varying from red to red-brown, or mahogany to dark amber; this lasted 2 or 3 days. Among the patients who survived, the color varied from "ruddy" to amber; but this did not persist longer than 1 day.

Benzidine-Positive Pigment.-The grossly abnormal color of the urine in patients with the crush syndrome was found to be caused by the presence of a pigment which reacted positively to the benzidine test. By the technique described in Appendix C, this pigment was thought to consist of myoglobin as well as hemoglobin in three of the five patients who died of the crush injury. In the fourth patient no attempt was made to distinguish between the two pigments, and in the fifth patient the benzidine test was not made on the dark-amber urine. Results of the tests for myoglobin were never positive in urine specimens obtained later than 24 to 36 hours after release from the crushing force. Benzidine-positive substance was found later than the second day and up to the eighth day after release, but in such small amounts that it was impossible to employ the test to distinguish between myoglobin and hemoglobin. Qualitative spectroscopic examinations suggested the presence of myoglobin in the specimens obtained in fatal Cases 69, 70, and 93.

Albumin.-Two of the patients who survived showed some albuminuria. In neither case did it last more than 1 day, and in only one (Case 56) of the two was it of any degree of severity. All of the patients who died had marked albuminuria. Up to 4 or 6 days it persisted unchanged or with increased severity. If the patient lived longer, the degree of albuminuria then decreased, as in Cases 69 and 93. In the latter the urine contained only a faint trace of albumin by the time the patient died of lobar pneumonia 13 days after release from the crushing force.

Sediment.-The urinary sediment was remarkable because findings were so scarce. In one patient who survived (Case 124), no record was made of the examination of the organized sediment. Of the other three patients who lived, only one (Case 73) showed casts in the urinary sediment. On the first day after injury his urine showed occasional coarse and finely granular casts; 2 days later the sediment was normal. Two of the five patients who died had no casts in the urine, although a rare red blood cell and a few white blood cells were found; three did show coarse and finely granular casts in the urinary

270

sediment, but these casts were not abundant, even in centrifuged specimens.

Specific Gravity.-For the first day or two of the illness, the specific gravity of the urine was normal in all cases. After the second to fourth days, however, the specific gravity of the urine fell and appeared to become fixed at 1.012 to 1.013. Urine concentration tests were not done in these patients, but as in other cases of posttraumatic renal insufficiency early impairment of ability to reabsorb water by the renal tubules is surely a characteristic feature (Chapter V).

Reaction.-With one exception, the patients passed acid urine at all times throughout the illness. This one exception (Case 69) was a patient whose urine was on the alkaline side of neutral (pH 7.2) on two occasions on the seventh and eighth days of the illness. At all other times the urine of this patient was acid, despite administration of sodium bicarbonate in doses thought adequate for alkalinization of the urine (16 to 36 Gm. per 24 hours). As had been found in the study of patients who had renal insufficiency from other causes, alkalinization of the urine by means of administration of sodium bicarbonate was inconsistent with safety.

CHART 45. BROMSULFALEIN EXCRETION IN PATIENTS WITH FATAL CRUSH SYNDROME

271

CHART 46. PLASMA BILIRUBIN CONCENTRATION IN CRUSH SYNDROME
(FIRST 6 DAYS)

Liver Function

The function of the liver was tested in one of the four patients who survived and in four of the five patients who died. The examination consisted of a test of the ability of the liver to excrete bromsulfalein injected intravenously. Dosage was calculated on the basis of 5 milligrams of bromsulfalein per kilogram of body weight (Chapter II). By this method, retention of the dye up to 3 percent of the injected dose after 45 minutes was considered the upper limit of normal. In the patient who survived the function of the liver was found to be normal as tested by this method. In all of the patients who died liver function was found to be impaired (Chart 45). The average dye retention in the four fatal cases was 12.77 percent, exclusive of one determination made in Case 93, which showed 26-percent retention on the day the patient died of lobar pneumonia.

272

The concentration of bilirubin in the plasma was increased in those patients in whom fatal crush syndrome developed and in one patient (Case 56) who survived (Chart 46). Data were obtained on eight cases. Sufficient information was not available to distinguish between increased production of bilirubin and decrease in its elimination.

Pathology

The important findings at postmortem examination were confined to the lungs, the liver, the skeletal muscles, and the kidneys.

Lungs.-All of the five fatal cases showed slight to extensive pulmonary edema, both grossly and microscopically. The weight of the two lungs together was recorded in four of the five fatal cases. The weights varied from 1,075 to 1,600 grams. Blood as well as edema fluid appeared on cut surfaces of the lungs in Case 132 and, microscopically, large numbers of alveoli were packed with red blood cells. This patient had had a tracheotomy and the blood may have been aspirated. Another patient (Case 78) was unusual in that even the visceral pleura was edematous. Fluid poured from the cut surfaces without the necessity of compression, and the tracheobronchial tree was filled with nonbloody fluid. In Case 93 death was due primarily to lobar pneumonia involving the left upper lobe. Microscopically, in addition to the typical pneumonia process, some areas showed edema with spreading pneumonic involvement.

Liver.-In view of the decreased liver function, as measured by ability to excrete bromsulfalein and as indicated by the elevated plasma bilirubin and the abnormally low ratio of urea nitrogen to total nonprotein nitrogen, some histologic changes were expected in the liver. Surprisingly few were found. The liver was microscopically normal in Case 70, and showed only scattered vacuoles in Case 132. In Case 69 occasional mitotic figures were observed in hepatic cells, and in Case 93, in which there had been terminally a rising bilirubin concentration in the blood and a terminal bromsulfalein retention of 26 percent, the only histologic changes found were mild nuclear atypicality in the cells at the centers of the lobules.

Skeletal Muscles.-The gross changes in the involved muscles were conspicuous. In the involved areas the normal pigment had disappeared, leaving muscle bundles with the appearance of fish or rabbit flesh. Sometimes an entire muscle was decolorized, more commonly only segmental areas, particularly re-

273

FIGURE 24. Decolorization in the gastrocnemius-soleus muscle group. From necropsy in Case 78 (see Fig. 23).

FIGURE 25. Decolorization in the left brachioradialis muscle in Case 93. Note the "fish flesh" appearance, the sharp line of demarcation, and the distinct white band at the junction of the depigmented and normally pigmented muscle.

gions close to major bony structures (Fig. 24). In Figure 25 (Case 93) complete depigmentation of the distal half of the forearm muscles is apparent, whereas the proximal half is normally colored. The transition from normally colored to depigmented areas was sharp, particularly in the cases of longer survival, and in Case 93 (13 days' survival) the junction was marked by a narrow white band 2 to 3 mm. broad (Fig. 25). In spite of the loss of color, the

274

tissue remained moist, capillary oozing appeared from cut surfaces, and no thrombi could be discovered in any grossly visible vessels. Hemorrhage and edema were extremely irregular in extent and intensity, absent from some lesions, and marked in others. The gross depigmentation correlated with chemical assays showing almost complete loss of myoglobin in the affected areas (Table 100).

Microscopic sections were available from four cases 4, 5, 10, and 13 days respectively after injury. In the acute cases hematoxylin and eosin stains showed relatively inconspicuous changes. Evidences of muscle-cell degeneration were, however, present. Some cells were swollen, hyaline, and more strongly acidophilic than normal. More commonly the cells stained less intensely than normal, cross-striations were exaggerated, occasional fractures and clefts were present, and the sarcolemma nuclei appeared decreased in number (Fig. 26). Though edema and hemorrhage were apparent in some areas, there was no inflammatory reaction, either polymorphonuclear or histiocytic. With the Masson and phosphotungstic acid hematoxylin stains, the changes were far more conspicuous. Some of the swollen hyaline fibers stained intensely red with the Masson, or blue with the phosphotungstic stain, but the great majority responded feebly or not at all to the specific staining reactions. In sections across the border of a depigmented area it was evident that the loss of specific staining reaction corresponded closely with the margin of depigmentation (Fig. 27 a). It was of interest that the cells with exaggerated cross-striations usually failed to stain specifically.

TABLE 100.-MYOGLOBIN CONTENT OF MUSCLE AND KIDNEY IN 4 PATIENTS WHO DIED FROM CRUSH SYNDROME
 

275

FIGURE 26. Decolorized muscle from Case 70. The patient died 4 days after injury. The muscle cells, which should stain intensely with the phosphotungstic acid hematoxylin, employed stain weakly. Cross-striations are exaggerated and numerous transverse clefts have appeared. An occasional cell has become hyalinized and lost all evidence of striation. There is much hemorrhage into the interstitial connective tissue.

In the cases of longer duration (10 and 13 days), changes were conspicuous with all staining methods but particularly with hematoxylin and eosin. Many swollen, necrotic muscle cells had become intensely basophilic owing to the precipitate of innumerable fine granules staining like calium throughout the sarcoplasm (Fig. 27 b). In other cells the precipitate was intense in the central portion but absent at the periphery (Fig. 28 a). Another common type of degenerative change was a coarse hydropic vacuolation which frequently produced an appearance suggestive of vegetable cells (Fig. 27 c). Other degenerating cells showed varying degrees of infiltration with histiocytes. (Figs. 27 b and c). The sarcolemma nuclei in many cells showed amitotic division, with chains of touching nuclei 5 to 10 elements in length (Fig. 28 b). About many of these, viable cytoplasm was present in long, slender masses suggesting regeneration of muscle cells. The interstitial connective tissue was slightly edematous and densely infiltrated with histiocytes (Fig. 28 b). No definite evidence of fibroblastic proliferation could be made out. Blood vessels still remained free from thrombi or inflammatory reaction.

Kidneys.-In all but one patient (Case 70) the kidneys were enlarged at necropsy. In three of the five fatal cases the combined weights of the two kidneys were recorded; they were respectively 500, 550, and 625 grams. The capsules stripped easily. In some there were ecchymoses beneath the capsule. Focal

276

FIGURE 27, a. Muscle from Case 70 (4 days' survival). Phosphotungstic acid hematoxylin stain. Section taken from a partially
depigmented area. The muscle cells stain irregularly and striations are difficult to see. The most severely involved cells are stained a faint brown. The interstitial tissue is edematous but shows little inflammatory infiltrations.
b. Muscle from Case 69 (10 days' survival). Hematoxylin and eosin stain. This low-power view shows focal calcification as well
as fibrosis of the interstitial tissue and a chronic inflammatory infiltrate. A cell just below center manifests marked vacuolar degeneration.
c. A higher magnification from the same section shown in b. Masson trichrome stain. Coarse vacuolar degeneration of a muscle cell in the lower portion of the photomicrograph is evident. Several muscle cells in the center no longer stain red in normal fashion and have been invaded by histiocytes.

277

FIGURE 28, a. A focus of intense calcification involving almost every muscle cell. There is marked increase in interstitial fibrous tissue with an occasional focus of lymphocytic infiltration. Section from Case 93 (13 days' survival).
b. Section from the same case. Hematoxylin and eosin stain. A regenerating muscle cell shows long chains of touching sarcolemma nuclei. The fibrosis and the patency of the blood vessels are noteworthy. Partial calcification of some muscle cells at the periphery of the field is evident.

hemorrhages and fibrin deposits were commonly found beneath and on the epithelium of the renal pelvis (see Fig. 20 in Chapter IX). When the kidneys were cut, with one exception (Case 70), the renal parenchyma pouted through the cut, rolling outward and forming a rounded instead of a sharp edge.

Characteristically, the cut surface of the parenchyma showed pallor of the cortex and darkening of the pyramids to a dark brown or mahogany color (Figs. 19 to 22 in Chapter IX). The thickness of the cortex was normal or increased, never decreased.

The microscopic appearance of the kidneys from patients who died from crush syndrome have been described and illustrated in Chapter IX as a "lower nephron nephrosis," and it will be sufficient merely to list the salient features here.

Microscopic Features of the Kidney.-Most conspicuous was the presence of densely pigmented casts in the distal convoluted and the collecting tubules. These could not be distinguished in stained or unstained sections from the pigmented casts in Case 9, a known transfusion reaction. Proximal to the pigmented casts, usually in the ascending limbs of Henle's loops, hyaline nonpigmented casts were numerous. In cases with survival periods of 4 days, these hyaline casts were frequently extruded into the stroma and formed foci for granulomatous reactions. Degenerative changes in the renal parenchyma were limited to the lower nephron segments, usually the ascending limbs, less often the distal convoluted tubules.

278

Glomeruli were uniformly normal except for the presence of granular precipitate in the capsular space. The proximal convoluted tubules were sometimes moderately dilated and their lumens contained granular precipitate, but the epithelial cells were normal and brush borders were well preserved. Interstitial inflammatory reaction was always present at the corticomedullary junction, and in two cases of 10 and 13 days' duration, a narrower zone of interstitial inflammatory reaction was seen beneath the capsule. In the same two cases nonocclusive thrombi were present in some of the veins at the corticomedullary junction.

Treatment of Crush Syndrome

The results of treatment of patients with crush syndrome were disappointing. (Our ministrations were certainly less effective than those of Saint Benedict, whose successful treatment of a monk crushed beneath the masonry of a wall at Monte Cassino is depicted on page 254.) In general, efforts during the first 48 hours were directed toward correction of the hemoconcentration and attempts at alkalinization of the urine. Subsequently the treatment was that for renal insufficiency as described for patients with renal insufficiency from other causes (Chapter VI).

Correction of Hemoconcentration.-Hemoconcentration was treated by administration of plasma and crystalloid solutions during the first 48 hours. Beyond this time, the degree of hemoconcentration was no longer a serious problem, for reabsorption had probably begun to exceed transudation. Only two of the nine patients received whole blood. One of the two (Case 69) received only 250 cc. of blood, administered before the hemoconcentration had been recognized. The other (Case 132) was given 1,500 cc. of whole blood when he became pulseless and no blood pressure could be determined in his extremities.

Alkalinization of the Urine.-According to Bywaters'⁶ hypothesis, alkalinization should be started at the time the casualty is found, immediately after or even before release from compression. This was not possible in any of our patients. The seriousness of the injury was not recognized by the aid-men. As a matter of fact, the diagnosis was not made by the medical officers until, through an educational campaign, they had been made aware of the possibility of the syndrome in any crush injury.

⁶See footnote 1.

279

CHART 47. EFFECT OF ALKALI THERAPY IN A PATIENT WITH CRUSH SYNDROME
(CASE 69)
After admission to the hospital, however, all but two of the nine patients received alkali therapy. Four of the five patients who died were treated with alkali, one starting promptly upon admission to the hospital, one starting during the first 24-hour period after release from the crushing force, and two during the second 24-hour period. No beneficial effects were observed that could be attributed to the alkali. As a matter of fact, the most striking feature was the failure, in the fatal cases, to succeed in alkalinizing the urine. A pH of 7.2 was attained in only two specimens of urine in Case 69 (Chart 47). This patient had received larger doses of sodium bicarbonate than the other patients. One patient in this group (Case 124) did have an alkaline urine (pH 8.4) 4 hours after intravenous administration of sodium bicarbonate, and it was kept alkaline by the administration of 3 Gm. of the bicarbonate every 4 hours. This patient had a minimal renal lesion and survived. Administration of sodium bicarbonate was done cautiously or abandoned altogether, because of unfortunate experiences with vigorous attempts at alkalinization in patients with renal insufficiency from other causes (Chapter VII).

Diuretics.-Alcohol was used with the hope that it would promote blood

280

flow through the kidney and serve as a diuretic. It was used intravenously (total dose: 50 to 100 cc. of 95-percent ethyl alcohol as a 5-percent solution in dextrose or isotonic saline solution), and as whiskey by mouth (60 to 180 cc. per day). No beneficial effects were detected that could be attributed directly to the alcohol. Three of the patients were given 10-percent dextrose in distilled water intravenously, and one received 50-percent dextrose intravenously as a diuretic, but no effect was noted in any of them that could be attributed to the medication. The failure of both alcohol and hypertonic solutions of dextrose to promote urine flow or improve kidney function confirmed the findings in patients with renal insufficiency from other causes (Chapter VI). Not only did concentrated solutions fail to induce a diuresis, but in three cases (Cases 70, 78, and 132) the hypertonic dextrose solution may have hastened the onset of pulmonary edema. Mercurial diuretics were not tried in patients with the crush syndrome.

Dehydrating Agents.-On the hypothesis that swelling and edema of the kidney may be in part responsible for failure of function, the use of potassium and magnesium solutions was recommended by Barker⁷ to reduce the tissue swelling and to promote diuresis. Potassium chloride was not available to us for clinical use, but magnesium sulfate was used intravenously and intramuscularly. It was employed in three cases (Cases 69, 78, and 93) in doses of 4 to 8 Gm. daily for 3 to 5 days, which was sufficient to elevate the concentration of magnesium in the blood as indicated in the two patients in whom such analyses were made (Cases 78 and 93). In these three patients there was no definite effect upon urinary output, and it is impossible to attribute any beneficial effect to the use of magnesium. Undue drowsiness in Case 78 could well have been caused by the magnesium therapy, for his serum concentration of magnesium was 5.8 milliequivalents per liter at necropsy. The serum magnesium reached a level of 6.8 milliequivalents per liter at one stage during the course of treatment in Case 93. For the same reasons stated in Chapter VI, it is our opinion that neither magnesium nor potassium solutions should be used in the treatment of this syndrome.

Other Measures.-Study of these nine patients taught us little as to possible means of preventing anuria or inducing diuresis. Generally accepted methods failed. On the hypothesis that injury to the kidney may be produced when reabsorption of extravasated fluid takes place, pressure bandages were considered worthy of trial and were applied in one case (Case 78) without apparent effect.

⁷BARKER, MARION H.: Personal communication.

281

In this case, however, serious extravasation into the tissues had already occurred before the patient reached the hospital. Cold applications might prevent rapid flooding of the kidney with large quantities of muscle-breakdown products, inasmuch as cold is known to inhibit the flow of lymph from the periphery into the general circulation. This therapy was not tried. Again, however, to be effective it would have to be started immediately after release from compression. Therapeutic amputation of the involved extremities was thought to be out of the question for the same reason.

What was learned from the patients with crush syndrome, as well as from those with renal insufficiency from other causes, was that it is very easy to bring about an increase in plasma volume in these patients by too vigorous attempts at inducing a diuresis with intravenous crystalloids. Once oliguria or anuria is established, it is best to withhold fluid, allowing, preferably by mouth, just enough fluid to compensate for loss from the gastrointestinal tract, and for insensible loss through the skin, from the lungs, and from the sweat glands. Under ordinary circumstances, these requirements can be met by 1,000 cc. or less of fluid daily; more must be allowed if excessive extra-renal losses occur. In general, patients with renal insufficiency have been found to be over-hydrated rather than under-hydrated, and this accounts for the fact that pulmonary edema is a common cause of death.

SUMMARY AND CONCLUSIONS

A laboratory and clinical study was made of nine patients who received crushing injuries; four of the patients survived, five died. It is emphasized that the crush syndrome develops insidiously in patients who appear well when first seen. All patients who have suffered compression interfering with the circulation of muscle masses for periods of an hour, and perhaps even less, should be observed carefully for development of features of the syndrome; namely, tense swelling of the affected parts, hemoconcentration, benzidine-positive pigment (presumably myoglobin) free in the plasma and urine, and oliguria or anuria. If examinations are not made within 24 or 48 hours, the hemoconcentration and the myoglobinuria may be missed. The volume of urinary output should be watched closely, for significant oliguria may easily be overlooked.

The principal pathologic findings are in the compressed muscles and in the kidneys. In the former the normal pigment, myoglobin, was rapidly mobilized

282

and absorbed, and extensive necrosis of muscle developed which tended to heal with calcification if the patient survived a week or longer. In the kidney a pigment nephropathy was found, which was indistinguishable from that which develops following other forms of severe trauma. A common terminal finding was severe pulmonary edema.

The treatment applied to these nine cases has been described. In the presence of oliguria or anuria, it was impossible to alkalinize the urine with doses of sodium bicarbonate which could be administered without danger of producing severe metabolic alkalosis. It is particularly important not to flood the circulatory system with intravenous solutions in an attempt to induce diuresis, once oliguria or anuria have become established. When excessive fluid is administered, the patients may die of pulmonary edema before fatal uremia develops.

That a recovery diuresis may develop in these cases is evidenced by two patients in this series in whom recovery diuresis had begun before death; in one of them it was well developed.

CASES OF SPECIAL INTEREST IN THIS CHAPTER

Fatal
69 70 78 93 132

Nonfatal
56 73 99 124

CASES OF SPECIAL INTEREST IN CHAPTER IX

Nonfatal
13 27 29 37 44 54 60 71 72 81 87 
104 109 112 125 130 133 138 150

Fatal
9 12 22 24 25 26 31 41 47 49 
52 55 65 66 74 80 85 86 88 
95 97 98 108 116 117 118 120 
122 123 129 131 136 A-30

Crush Syndrome
69 70 78 93 132

RETURN TO TABLE OF CONTENTS