PHYSIOLOGICAL INTRAVASCULAR HEMOLYSIS OF EXERCISE. HEMOGLOBINEMIA AND FOLLOWING CROSS-COUNTRY RUNS

D. Rourke Gilligan, … , M. D. Altschule, E. M. Katersky

J Clin Invest. 1943;22(6):859-869. https://doi.org/10.1172/JCI101460.

Research Article

Find the latest version: https://jci.me/101460/pdf PHYSIOLOGICAL INTRAVASCULAR HEMOLYSIS OF EXERCISE. HEMOGLOBINEMIA AND HEMOGLOBINURIA FOLLOWING CROSS-COUNTRY RUNS By D. ROURKE GILLIGAN, M. D. ALTSCHULE, AND E. M. KATERSKY (From the Medical Service and the Medical Research Laboratories, Beth Israel Hospital, and the Department of Medicine, Harvard Medical School, Boston) (Received for publication May 19, 1943)

Studies on 3 cases of march hemoglobinuria were introduced into a tube containing 0.5 cc. of 3.0 per were recently reported from this laboratory and cent sodium citrate solution. The tube was stoppered and gently inverted several times to secure mixing. The the existing literature concerning this syndrome plasma specimen was removed from the citrated blood was reviewed (1). In 1 patient of our series and samples from one-half to one and one-half hours after in some other reported instances, frank hemo- the blood was drawn. In many instances, 5.0 cc. of ad- globinuria has been observed following walks of ditional blood were placed over oxalate mixture for only 20 minutes' duration. Such occurrences are studies of hematocrit and of erythrocyte fragility. A sample of was collected shortly after blood was obviously unusual and are considered to be due to taken. Examination of the abdomen for the presence of some as yet unrecognized and apparently benign hepatic or splenic enlargement was made at the end of abnormality (1). There are, however, other re- exercise. ported cases in which hemoglobinuria occurred Chemical methods. The chemical methods utilized were only after very long, strenuous walks or after long described in detail in a previous publication (3). The runs. concentration of in the plasma and urine was The question arises whether intravascular measured by the benzidine method as described by Bing hemolysis, giving rise to hemoglobinemia and and Baker (4, 5), with slight modification (3). Tests sometimes hemoglobinuria, occurs with sufficient showed that no appreciable change in the plasma hemo- frequency in certain types of strenuous exercise of globin value occurred when the citrated blood sample re- long duration to be considered physiological. To mained at room temperature for 2 hours before separat- in a ing the plasma; these tests were made in order to make gain information this regard, study was made certain that the necessary standing of the blood for from of the incidence and degree of intravascular he- 30 to 90 minutes, between the time it was drawn at the molysis in normal individuals performing long place of termination of the exercise and the removal of strenuous runs; the results of these studies and the plasma in the laboratory, resulted in no measurable of other urinary and blood studies pertinent to the hemolysis. It was also demonstrated that the amount of problem are reported and hemoglobin in the plasma was not affected by transport- discussed herein. ing the blood from the field to the laboratory. The plasma hemoglobin values of some 50 normal subjects, who had MATERIAL AND METHODS not undertaken any strenuous exercise on the day of Material. Studies were made in groups of young sub- study, revealed values usually of 1.0 to 6.0 mgm. per 100 jects performing runs as follows: (1) 11 men, aged 17 to cc., with a rare value between 6.0 and 9.0 mgm. per 100 18 years, after a cross-country run of 2.6 to 2.8 miles, in cc. and an average value of 4.0 mgm. per cent. We be- 14 to 18 minutes; (2) 11 men, aged 19 to 21 years, after lieve that this variability between 1.0 and 6.0 mgm. of a cross-country run of 4.5 to 5.1 miles, in 23 to 31 min- hemoglobin per 100 cc. of plasma is not a true variability utes; and (3) 26 men, aged 18 to 65 years, after a mara- in the plasma concentration of the circulating blood, but is thon run of 10 to 262 miles. In addition, comparative more likely occasioned by slight in titro changes. This studies were made in 1 subject after the 26.2 mile mara- impression is gained from the fact that 2 blood samples, thon run, after a 5-hour period of exercise on the bicycle drawn within a period of a few minutes from a given ergometer, after a 262 mile walk with a 52-pound pack subject at rest, not infrequently vary by 2.0 to 4.0 mgm. on his back, and after an additional run of 112 miles in per 100 cc. Although it is perhaps true that minimal 76 minutes and 19 seconds. extravascular hemolysis cannot be avoided, no plasma Methods of study. A blood sample was taken from an hemoglobin value after exercise has been considered in- antecubital vein, without stasis, within 15 minutes after creased unless it was higher than the uppermost limit completion of exercise. The blood for plasma hemo- encountered in normal resting individuals, namely, 9.0 globin measurement was drawn as suggested by Ham mgm. per 100 cc. (2): a sterilized syringe and needle, previously rinsed The concentration of the plasma was meas- with sterile saline solution, were used; 5.0 cc. of blood ured by the method of Malloy and Evelyn (6). The to- 859 860 D. ROURKE GILLIGAN, M. D. ALTSCHULE, AND E. M. KATERSKY

0

CaC Ca '0 e' . 00 . - . Cd oo bO b Cd .00, 4,, Css 4_l >,0 4° >, 0 u >.4- (00 o0 - eq °= = ox S. x ~~~~~~~Ca a.. 3.. a) Cd 0 NO 0o 4 in 4 00 4C4 C 02 2 2 00 0 00 0 4i 4- 4i 4i -W 4-) 4. 4J 4.) adCadc) o oeq o o o o -14 e14 C 0 lq040 0 coa4 0CV000- c..o -a) 4 4 .J 4 4 a.) 0~ o oou2222 1.3_ " 8 m a4() a-) a ) a) 00 0 a 0 .. .0 2 o 2 o~ 22 3 .. a) cocc; 000e000 C a Cd la U.) ed4i4i Cd 4- C Cd *0 a. a) 0c 0000000 5 - .0 -00 . . 0. 0. co8 0 0 Q 10 00 00 000000 00000 0 a) 0 0 10 C a a.5 0 Cl 4._ a) a) eq '0 0004 Cx b4 00 00 0000000 0 (2 000000o 0 Cl '0-

U) r) 0 0SOoe ~o '0 '0' 'UV r'''a) r 0'a * '0'0 '0'0'V '0 a. .0 *-u * < t.c* C.) C.) C. C.) C. C. C.) C.C)- Ca Cd CaCa CaCaCaCaCaCaCa C Ca Ca Ca Ca Cd cd 4.0 CO Ei 0 COb-.0 a) NO000 Z- 00 V%0qC4 0 cO 0 t' 00 00 00 in 00 00 V- C4 C4 C4 CN N z eq - - eq - -- - 0~0* 0* 0 0 z ** .*.. *00 ..0 Ca -,0e Cx: to000ON00O 0 toNo0t-C)C-Cl ld C.) Cx. ad ) eoo te a)o0o0too (NUCn-14 %0 000 %00 00 a) mmU)0 Q U)%W4c ei - Lq-- -M" w- v-4 v -4 W-|- z Cx co "I) U) C.4 3-. cr Ca CO Cd N U) U) 0 00 04 go " 0.blo z P. O.. a)co eas j% ~. ZA) z co - Cx

OD zV) a) CO bo V CoC 0 cn 0 0 A 11. .- - r. *2d4_ld4Ut-UI CaC >4i 0 O o 'V.14 :3 9 z bO 04&..o 6. C CV%4i in %0 CV4cs E)M ,4eqeq..4 bOCa 4n

I 0U) N 0 0.o-~~~~~~~~~~~~~W4-00toE) OON eq on 0-0U)U) eqeeq) U)eq_ eqC) eqeqe4q~e_ e qeqe" 4 14. a ;Ei 0 :9 4J ' v 93 Ca 0 '04 V-4 %0' V-U)'00%0 4-4%0 E w rACOa E-4 !i A u) ++u) 00U)+u) r. 'IO %ONoso NO O %O Naw U) ei U)U)UI)U) c- in% 00V4to t -V)LO 0 00%OL 0%0% o._ v v r. Cd 93 -W :3 -a .m W r Ca - 0 o 0% .cn C) -w8.9W .0 C: 0000qC 0 50F.00o00o.0 eqC)C)~ U) +%Oab000%0'- ac ) cn O uuuuu ououesoo CO 4-. 4)4* *i*) -U*.*.- t. .) U) in Uin In - - t- - in eq4 eqeqC4v4C4V- 4~~4.h _~~+4

Cd C.)C.) OC.OC)C.C.C. a)

0000 ZOZOOQO a)i4J4- J J4 4 - 00000 0o0oooo U 0 U U U 0i UU Z Z oz INTRAVASCULAR HEMOLYSIS OF EXERCISE 861 tal protein concentration of the urine was measured gravi- passed a red urine after a race in 1939, i.e., a year metrically after precipitation with trichloroacetic acid before these statistical studies were made, and (3). When the total protein was less than approximately studies. In both 30 mgm. per 100 cc., the concentration was estimated by again in 1941, a year after these comparison of the turbidity of the urine with that of instances, the urine became normal in color and standards after the addition of 5.0 per cent solution of showed no hemoglobin or albumin on the morning sulfosalicylic acid. Spectrophotometric examination of following the race. There were no symptoms at plasma and urine specimens were made for us by Dr. the time of the hemoglobinuria. 0. H. Lowry of the Department of Biochemistry of the The plasma hemoglobin values Harvard Medical School; a Ko5nig-Martens type of Marathon runs. spectrophotometer was used. The fragility of the red after the marathon run were elevated above the blood cells in hypotonic saline was measured quantita- upper limit of normal in 18 of the 22 marathon tively, essentially as described by Hunter (7), with the runners who ran the entire 26.2 miles (Table II). following modifications: oxalate mixture was used as an The plasma hemoglobin values of these 18 sub- anticoagulant, the blood was oxygenated for 3 minutes in varied from 10 to 44 mgm. per 100 cc. and air, and after transferring the samples of blood to the jects saline solutions, the mixtures were allowed to stand for averaged 18 mgm. per 100 cc. In 11 of these 18 1 hour. In most instances, a phosphate buffer solution, cases (Cases 1 to 11, inclusive), the plasma hemo- giving a pH of 7.4, was added to the saline solution, ac- globin was greater than 15 mgm. per 100 cc. In cording to the suggestion of Dr. A. Baird Hastings of these instances, there was a definite reddish dis- the Department of Biochemistry of the Harvard Medical of the plasma. Four of these subjects School; the final salt concentrations were calculated in coloration osmolar concentration and expressed in equivalent per showed hemoglobinuria, with concentrations from cent of pure sodium chloride solutions. a trace to as high as 0.4 gram of dissolved hemo- globin per 100 cc. of urine; 3 of these were RESULTS discolored from light brownish to very dark bur- Plasma and urinary hemoglobin findings gundy red. The hemoglobinuria occurred in the 4 subjects who showed the higher plasma hemo- Cross-country runs. Of 11 Harvard freshmen globin values at the end of the run (Table II). who ran 2.6 to 2.8 miles in 14 to 18 minutes, 5 Two of the subjects who showed hemoglobinuria had plasma of 10.0 to 18.0 mgm. per at the end of the race were studied for several 100 cc. (i.e., above the upper limit of normal, 9.0 hours after the race (Table III). Hemoglobi- mgm. per 100 cc.) at the end of the race (Table nemia and hemoglobinuria disappeared in both I). One of these 5 ran again and had an in- cases within 3 and 4 hours after the end of the creased plasma hemoglobin on the second run also race. (Table I). Urines collected in 10 of these sub- Four of the marathon runners had histories of jects, following the run, were of normal color and passing red urines on one or more previous occa- showed negative benzidine reactions. sions following long runs; 1 of these subjects Of 11 Harvard varsity track men who ran 4.5 (Case 6, Table II) showed hemoglobinuria after to 5.1 miles in 23 to 31 minutes, 5 had definitely the race of the present study, and again, 11 months reddish plasma with plasma hemoglobin values later after another long run (Table II). elevated to 17 to 47 mgm. per 100 cc. (Table I). There was no correlation between the age of Studies made after another run in 2 of these 5 the runners or the number of years or state of subjects also showed increased plasma hemoglobin training and the occurrence of hemoglobinemia or concentration. Urines were collected after the the who ran the the hemoglobinuria, either in subjects runs in 8 of the 11 subjects, including subject marathon (Table II) or in those who ran the a of 47 per who showed plasma hemoglobin mgm. shorter races (Table I). 100 cc.; the urines were all of normal color but was detected in one hemoglobinuria specimen by Further observations during various exercises on a test. the benzidine marathon runner wtith Two of these subjects who showed hemoglobi- hemoglobinuria nemia, following these cross-country runs, also Further studies on the production of intravas- gave histories of having passed red urines on other cular hemolysis during walking, bicycling, and occasions following runs. One of these subjects running a shorter distance were made in the case 862 D. ROURKE GILLIGAN, M. D. ALTSCHULE, AND E. M. KATERSKY

Cd( cd 4)70 (4~~~~d (4 c1 ( cC .-004 *do.t .oo.0.0 a- - o a o AS ~~o HYoA oA oAc Y HS co U, bl m C4 to-b Ed NOz= .0 o 0 0 0 C 0 0 0 0 0 000 0 -'0 0 a) 0Cd0 00000 0 . (4- 0 e q 44 Ci) o 2202 V" ~0 - a. .0 0 (4 C4 C4 C4 in (4 .4 0 0 0 0 a) ° °4 0 ° °22° ooE 4._ . _. U- t) 0 000 C4 0 C - 0 a) 0 .4 .bQa- 1..U)4 114 ) 0 8 0 X o 0 0 ' a o% o. o. e o. 00- o._ o . oo. eq00-u go B oo - oo o o o o ooo ooo 8 U a)Cd "4 %a. a-.. 4)b

-4 (4

0. i

0- a)~~~~~~~~~~~~~~~~~~~~~~~~~e .0 .0

04 -- - -v - eq _q i4) (4) eq eqeq(4) 0 eq -4 oq oeq_e a) w F oo. o. o. oooo. I0 - - -o-.-.-.-.-.oo.o.o.--- -.

v-4 0 - - - eq - - co CO 01 0 - CE)5 4.1~~~~~~~~~~~~i4. i J4 '0.0 4J O~~~~~~Wg la IWO L r Cd A-Q-0 0- 0 -) 0 0 - 0o

a)~~~~~~~~~~~~~~~. - '0 a a. 00 W) o. 0oo. Wc0 WW0.m 0 o. W,) 0s m 0 '00 tg0 4 Mo I A~~~~~~~~~C W) m C4 8 t08S Wda-~~~~~c 1)% - ' ~4tf t'.U) ' 4'. 4'I!1) -- f)l)4i)41 1) i~I)0 -Z tt '1)0%if) ~~~0 0 00 eq 0 0 A ] = 0 0 0ee 00 00 OeOf 4) z w pO ~ ~ ~'0>8a ~ 4eq>00 0 00 00' 80eq -- 0 ('' '0 §

.0 eq eq eq eq eq ------g ---g - 5 I

q ) o'o00o 'o 0o o o 4a g a" o I a (E) ie(f -X > > o44(4 I 4 .9 .C 6

cI I UW I ! 0 '0 Wm 0 % 0 0 0 0 0 00'0'c '

aa a eqe seq eqe eq e eqe q qeq eq eqqeqneeq - eq ~eqeq *

"I ~~~~~~~~~~~~~~~~~~~~~~~~~~C I Ok tU4 oo 0 eq ( 0% -to %o'0 07% o0 eq. 0 w =%0% m m < Iq'0eq -eq e eq eq e q eq in,a 00 e ) _ eq (44 U)'NO ON - 4 U) 'IOt- 00 0%0 - eq U) '0 - - - - -_- - - e eq eq +4i INTRAVASCULAR IIEMOLYSIS OF EXERCISE 863

TABLE III Further observations in 2 subjects showing hemoglobinuria at end of the marathon race

Plasma Urine

Subject number* Time Hemoglobin Bilirubin Volume Hemoglobin Total protein mgm. per 100 cc. mgm. per 100 cc. cc. mgm. per 100 cc. mgm. per 100 cc. 4 End of race 24 1.36 190 403 1100 2 hours after 10 1.29 105 6 254 3} hours after 6 1.29 115 trace 22 6 End of race 20 1.07 120 12 416 3 hours after 6 1.18 90 0 78 4j hours after 6 1.12 75 0 0 * The subject numbers are the same as in Table II. showing the largest degree of hemoglobinuria fol- , since the point of maximum absorption lowing the marathon (Case 4, Table II). A run of the alpha band of oxymyoglobin is 581 miu. of 11.2 miles in 76 minutes resulted in a rise of There were no methemalbumin bands observed in plasma hemoglobin to 13 mgm. per cent from the the plasma and the Schumm test was negative in control level of 2.0 mgm. per cent. Neither hemo- 3 of the marathon runners with plasma hemo- globinuria nor occurred; a few leuko- globin values of 20 to 44 mgm. per cent (Table II, cytes appeared in the urine. The plasma bilirubin Cases 1, 4, and 6). rose from 0.46 to 0.60 mgm. per cent and re- mained at that level for another hour. The hema- Plasma bilirubin studies tocrit was unnchanged. A walk of 26.2 miles The plasma bilirubin concentration, measured in with a 52-pound pack caused no change in plasma 12 of the marathon runners, 11 of whom ran the hemoglobin; the urine also showed no changes ex- whole 26.2 miles, ranged from 0.58 to 1.72 mgm. cept for the appearance of a few cells. Plasma per 100 cc., with 10 of the values 0.8 mgm. per 100 bilirubin, hematocrit, and erythrocyte fragility cc. or higher (Table II). These values are dis- were the same before and after. Work on a sta- tinctly elevated; an average value found in 36 nor- tionary bicycle for 5 hours, against a resistance of mal subjects, using the same method, was 0.32 5.0 kgm., during which time the speedometer mgm. per 100 cc. and the range was from 0.1 to showed a change equivalent to 73 miles, likewise 0.8 mgm. per 100 cc. There was no correlation resulted in no increase in plasma hemoglobin. between the degree of hemoglobinemia and the The urine showed no hemoglobinuria or albumi- degree of bilirubinemia at the end of exercise. In nuria; a very few cells appeared in the urine. 2 subjects, studied 4 hours after the end of the Plasma bilirubin and erythrocyte fragility were race, the bilirubin was still elevated to levels of 1.3 the same before and after the work. The hema- (Case 4) and 1.1 (Case 6) mgm. per 100 cc. tocrit showed a questionable increase. (Table III). Subsequent bilirubin values found Spectrophotometric studies at rest in Case 4 were 0.3 and 0.4 mgm. per 100 cc. Enlargement of the liver or spleen was not de- Spectrophotometric examinations of the plasma of Cases 1, 4, and 6 in the marathon series, and of tected in any of the runners. the urine in Cases 1 and 4, and examination of the studies plasma of 1 runner (Case 2, Table I) of the 5- Erythrocyte fragility mile cross-country race revealed in each instance Quantitative studies of the fragility of the red a point of maximum absorption of 576 mp which blood cells in hypotonic saline solution were made corresponds with the point of maximum absorp- in 6 of the subjects after the marathon run. Five tion of the alpha band of human oxyhemoglobin. of these subjects had hemoglobinemia and 1 of These findings show conclusively that the plasma these 5 had hemoglobinuria (Case 1, Table II). and urinary pigments were hemoglobin and not In all instances, the fragility curves fell within the 864 D. ROURKE GILLIGAN, M. D. ALTSCHULE, AND E. M. KATERSKY

PERCENT SODIUM CHLORIDE SOlUTION FIG. 1. ERYTHROCYTE FRAGILITY CURVES ATM THE MARATHON RUN AND IN NORMAL RESTING SUBJECTS range of those observed in normal subjects (Fig- Eight cases in which hemoglobinuria occurred ure 1). after long cross-country or marathon runs have been reported (1, 10 to 13). In 1 of these cases DISCUSSION (11), hemoglobinuria also developed after strenu- ous running for a long time in a game of "Fuss- The present study is, to our knowledge, the first ball." It is of interest also that Hastings (14) has quantitative statistical study of hemoglobinuria observed hemoglobinemia in untrained dogs after and hemoglobinemia occurring in subjects follow- they ran for 20 miles on the treadmill. ing long strenuous runs. The fact that only 13 cases of hemoglobinuria There are now available in the literature reports following runs have been reported in the literature of a total of some 13 cases in which hemoglobi- affords no accurate information concerning the nuria occurred in young men during runs. Dick- incidence of occurrence of intravascular hemolysis inson (8), in 1894, reported 1 case of hemoglobi- during this exercise. As was stressed recently nuria which occurred after a 3-mile run and (1), hemoglobinuria following exercise occurs another case which occurred after a strenuous only when the plasma hemoglobin is increased con- tennis match. This author (8) also stated that it siderably; an amount of exercise just short of that was recognized in English sport circles that the required to produce hemoglobinuria in subjects urine was sometimes red after a strenuous run with march hemoglobinuria does nevertheless re- during the beginning of training. Three cases sult in a definite elevation of the plasma hemo- have teen reported in which hemoglobinuria was globin ( 1 ). Hemoglobinuria is therefore a second- noted after runs as short as 1 or 2 miles (1, 9). ary manifestation of intravascular hemolysis and INTRAVASCULAR HEMOLYSIS OF EXERCISE 8865 occurs only in instances with the higher concen- If the urine alone had been studied after the trations of plasma hemoglobin. race in the 11 freshmen running races of 2.6 to The single most pertinent published observation 2.8 miles in approximately 15 minutes (Table concerning the incidence of occurrence of hemo- I), no evidence of any intravascular hemolysis globinuria during strenuous runs is that of Jun- would have been detected; however, the plasma dell and Fries (13) in 1911. During a statistical hemoglobin studies revealed slight but definite in- study of the albuminuria of exercise, these authors travascular hemolysis in 5 of the 11 men. Simi- noted 4 cases of hemoglobinuria among 39 sub- larly, during the 4.5 to 5.1 mile runs, performed jects who ran distances of 10,000 meters (6.2 by the varsity team in approximately 23 to 31 miles) to 42,194 meters (26.2 miles). In these 4 minutes, hemoglobinuria occurred in only 1 of the cases the urine was described as bright red, brown- 11 subjects, whereas a considerable amount of red, or dark blood-red, and all gave strongly posi- hemoglobinemia, at times sufficient to give a defi- tive guaiac reactions; there were no red cells or nite reddish tinge to the plasma and to be detect- very few in the urinary sediments. The urines able with the Zeiss hand spectroscope, was present in 2 of these cases were examined spectroscopically in 5 of the 11 subjects at the end of the race. The and showed oxyhemoglobin bands; in 1 instance, incidence of detectable hemoglobinuria after the the plasma of a blood sample, drawn 1 hour after 2.6 to 5.1 mile runs in college men, namely, 1 case the run, was reddish brown and showed oxyhemo- in 22 subjects, would certainly be too low to dem- globin bands on spectroscopic examination. Ex- onstrate the benign physiological nature of the in- aminations of the urine for hemoglobin were not travascular hemolysis of exercise; however, the made following runs in any of the 39 competitors incidence of hemoglobinemia, namely, 10 cases in except those 4 cases in which the urines were 22 subjects, is sufficiently high to suggest that obviously red, nor were plasma examinations intravascular hemolysis during such exercises is made in any but the 1 case noted (13). It is im- quite normal. portant that the incidence of frank hemoglobinuria, A quantitative study of plasma hemoglobin re- i.e., reddish discoloration of the urine due to dis- vealed a strikingly high incidence of increased in- solved hemoglobin, after long-distance runs, i travascular hemolysis in those subjects who ran Jundell and Fries' series (13) and in our study 26.2 miles in the marathon race; 18 of the 22 men (Table II) is quite similar. Jundell and Fries studied had suspiciously reddish to definitely red- observed 4 cases of frank hemoglobinuria in a dish plasmas and quantitative hemoglobin meas- series of 39 subjects who ran from 6.2 to 26.2 urements revealed plasma hemoglobin values dis- miles, while 3 cases of frank hemoglobinuria and tinctly above normal (Table II). Four of the 1 case in which a trace of hemoglobin was de- subjects running in the marathon showed free tected by chemical analysis were found in our hemoglobin in the urine; these 4 were the subjects series of 26 subjects running from 10.0 to 26.2 with the highest plasma hemoglobin values, namely, miles.' 20 mgm. to 44 mgm. per 100 cc. at the end of the race (Table II). 1 Mr. William T. Wright of the John Hancock Life Insurance Company recognized an additional instance in Although hemoglobinuria in exercise occurs a marathon runner, during the course of a study on the only in the cases with higher plasma hemoglobin urine of runners in the North Medford Road Race (20 values, the plasma level at which hemoglobinuria miles) on March 14, 1942. A sample of urine was ob- tained from J. M. which gave positive tests for albumin and granular casts, and rare erythrocytes. This urine and hemoglobin, although it contained no erythrocytes; contained 15 mgm. of hemoglobin per 100 cc. An hour his urine was normal immediately before the race and and one-half later his urine still contained a small amount again 24 hours after it. Through the cooperation of Mr. of albumin but no hemoglobin. He was a 32-year-old Wright, we were able to study J. M. after the B. A. A. white mechanic who had noted the occurrence of "black" Marathon of 1942 and obtained the following data: J. M. urine 5 years ago, when he began to run distances over 5 ran the 26 miles and 385 yards in 3 hours, 3 minutes, and miles. There were no associated symptoms of any sort; 40 seconds. Thirty minutes later he voided 37 cc. of hemoglobinuria did not occur after exposure to cold. dark amber urine with a specific gravity of 1.032 and Physical examination was entirely normal; the blood containing a large amount of albumin, a few leukocytes Hinton and Kahn reactions were negative. 866 D. ROURKE GILLIGAN, M. D. ALTSCHULE, AND E. M. KATERSKY occurs following exercise appears to be consider- no evidence of intravascular hemolysis after a 26- ably lower than the level required to produce mile walk with a 52-pound pack. hemoglobinuria in normal man when hemoglobin is It has been questioned by several authors (18 injected intravenously (3, 15). In the marathon to 20) whether the red pigment of the urine in runners, a somewhat higher level of plasma hemo- march hemoglobinuria is not the muscle pigment globin than that found at the end of the race may myoglobin rather than hemoglobin. Our own have obtained earlier in the run, for work is per- studies in 3 previously reported cases of march formed most rapidly earlier in the race, especially hemoglobinuria (1), and a similar spectrophoto- in those subjects who take some 4 or more hours metric observation by Witts (21) in his case, to finish the course of 26.2 miles. Whereas showed, however, that the plasma and urinary pig- hemoglobinuria was observed in 4 of 5 urines from ments in these cases were hemoglobin and not subjects with plasma hemoglobin levels of 20 to myoglobin. Similarly, in several of the cases of 44 mgm. per cent at the end of the 26 mile mara- the present study, the red pigment of the plasma thon race, hemoglobinuria was detected in only 1 and urine was identified spectrophotometrically as of the 4 urines from subjects with plasma hemo- hemoglobin and not myoglobin. globin levels of 22 to 47 mgm. per cent after the The mechanism of the intravascular hemolysis 5-mile cross-country race. The plasma hemo- of exercise is obscure (1). Its occurrence has globin values were probably at their highest point been recorded in man only following the exercise at the end of these shorter cross-country races. of walking or running, and in dogs, only after run- There are not available sufficient statistical data ning. Equal or more strenuous exercise of other to demonstrate to what extent exercises other than types in man does not cause it (1). There are running lead to physiological intravascular hemol- reported in the literature 3 cases of hemoglobinuria ysis. However, there are reports indicating that following walks in the usual upright position, in very strenuous marching may lead to hemoglobi- which cases hemoglobinuria did not occur follow- nemia and hemoglobinuria in normal individuals. ing similar or longer walks in a slightly kyphotic Albu and Caspari (16) reported the occurrence of position (1, 22 to 24). Most subjects with march reddish yellow and dark red urines, showing oxy- hemoglobinuria show no lordosis of the lumbar hemoglobin bands by spectroscopic analysis, in 2 spine (1), and in the present study, there appeared of 3 subjects after a most strenuous march of 202 to be no correlation between body build or stand- kilometers (125 miles) in an endurance walking ing posture and the occurrence of intravascular contest. Feigl (17) examined the plasma spec- hemolysis during runs. troscopically and the urine by the qualitative benzi- Although these evidences of hemolysis, namely, dine test for hemoglobin in 27 men after an army the appearance of red plasma and less frequently pack march of 35 kilometers (21.7 miles); he re- red urine, occur after strenuous runs, the total ported finding oxyhemoglobin in the serum of 6 amount of blood destroyed intravascularly is un- and a positive benzidine test in the urine of 19 of doubtedly very small. It may be calculated that the 27 cases after the march. Of the urines con- the hemoglobin from less than 10 cc. of normal taining no erythrocytes which showed a positive blood is sufficient when released into the plasma benzidine test, some were voided in cases in which to cause a hemoglobinemia of approximately 50 there was no oxyhemoglobin or hematin detected mgm. per 100 cc. of plasma in a subject with a in the serum. Because of the qualitative nature of blood volume of 5 liters and a normal hematocrit. the study, these findings are difficult to interpret, The findings of a previous study (3), in which especially since there is no mention of any reddish hemoglobin in solution was injected intravenously discoloration of the serum or urine in any of the in normal subjects, accord with this statement. cases. Further statistical studies of the incidence These injection studies also demonstrated that the of intravascular hemolysis in normal individuals plasma hemoglobin level does not become normal during marching would be of interest. In the again for about 5 hours after the injection of suffi- present study, the marathon runner (Case 4, Table cient hemoglobin to cause a plasma level of 50 II) with the most marked hemoglobinuria showed mgm. per 100 cc. Since the plasma hemoglobin INTRAVASCULAR HEMOLYSIS OF EXERCISE 867 level was not greater than 50 mgm. per 100 cc. in Tables I and II, the urinary findings after after the runs in any of the subjects of the present strenuous exercise may be quite the same as ob- report, and since increased levels were not main- served in ; in subjects with normal kid- tained long, it must be concluded that the hemo- neys, however, albuminuria disappears and the globin of not more than approximately 10 cc. of urinary sediment becomes normal in one to a few blood was destroyed. This is approximately true hours after the cessation of the exercise. In cer- even when there was some hemoglobinuria. The tain instances, the specific gravity of a urine col- greatest urinary hemoglobin output occurred in lected following strenuous exercise may be suffi- Case 4 after the marathon; the first urinary speci- ciently high to allow differentiation of this urine men voided after the race contained a total of from the urine in chronic glomerular nephritis. 0.77 gram hemoglobin and the following 2 speci- Sometimes, however, the specific gravities follow- mens contained a total of less than 0.01 gram. ing exercise may be within the range usually ob- The total hemoglobin excreted in the urine there- served in advanced chronic glomerular nephritis. fore would be contained in about 5.0 cc. of normal In 2 of the cases reported above (Table III) which blood. These findings are in marked contrast to were studied for some hours following the end of those in dogs in which anemia, reticulocytosis, exercise, the urine became completely normal 3 and increased excretion of bilirubin may occur hours after the termination of runs which pro- after long runs every day for several days (25 to duced albuminuria, hemoglobinuria, and the ex- 28). Whether there was blood destruction in our cretion of formed elements. subjects other than that which occurred intra- The frequent occurrence' of these temporary vascularly cannot be answered by the data of this renal changes manifested by albuminuria and the paper. Addition of erythrocytes to the circulating appearance of abnormal elements in the urinary blood during prolonged exercise has been re- sediment, and by decreased glomerular filtration ported (29 to 31). The hematocrits following ex- (38) during repeated strenuous exercise over ercise in our subjects after the marathon were many years, has never been shown to result in any entirely normal; no control values were obtained permanent damage. Similarly, the re- before the run. peated occurrence of the relatively mild hemoglo- The increased concentrations of bilirubin in the binemia and hemoglobinuria following exercise plasma after the marathon race may have been due has never been known to result in damage to the to decreased liver function consequent to hepatic kidneys.2 ischemia during the exercise. The fact that the The importance of differentiating the physiolog- hyperbilirubinemia occurred in the absence as well ical intravascular hemolysis of exercise from as in the presence of hemoglobinemia demonstrates pathological conditions leading to the passage of that it did not result from intravascular hemolysis; red urine is obvious. The benign nature of the the degree and duration of hemoglobinemia were albuminuria of exercise has been recognized for in no instance sufficient to give rise to such high a long time. That the passage of a red urine fol- levels of bilirubinemia (3). Although we have not lowing strenuous runs or walks may be equally as been able to find reference in the literature to benign has not been generally recognized, how- hyperbilirubinemia after exhausting exercise, there ever. The findings of this study are of particular are reports of hyperbilirubinemia in man during importance at this time, when large numbers of anoxia at high altitudes (32) and of impaired young men in the armed forces are accomplishing bilirubin excretion in dogs during anoxia from low extremely strenuous physical feats. oxygen tensions (33). It has been repeatedly demonstrated that normal SUMMARY AND CONCLUSIONS individuals excrete albumin and formed elements 1. Studies of intravascular hemolysis, as meas- in the urine after strenuous exercise (13, 34 to ured by the appearance of hemoglobinemia and The incidence of occurrence of 38). albuminuria, 2As suggested previously (1), alkalinization of the and the excretion of formed elements in the urine urine during the period of hemoglobinuria provides a pre- following the runs reported here, accord with the cautionary measure against precipitation of hemoglobin findings of Jundell and Fries (13). As observed in the renal tubules. 868 D. ROURKE GILLIGAN, M. D. ALTSCHULE, AND E. M. KATERSKY hemoglobinuria, have been made in groups of Wright, in arranging for the collection of the material athletes following 2.6 to 26.2-mile cross-country of study, and the assistance in the field and laboratory of Drs. Howard A. Frank, Paul M. Zoll, and Norman runs. The plasma bilirubin, the red cell fragility, Zamcheck. the hematocrit, and the urinary albumin and sedi- ment were also studied. BIBLIOGRAPHY 2. Hemoglobinemia was observed in 5 of 11 young athletes who ran 2.6 to 2.8 miles, in 5 of 11 1. Gilligan, D. R., and Blumgart, H. L., March hemo- globinuria. Studies of the clinical characteristics, athletes who ran 4.5 to 5.1 miles, and in 18 of 22 blood metabolism and mechanism; with observa- men who ran 26.2 miles. tions on three new cases and review of the litera- 3. Hemoglobinuria was observed in 1 of these ture. Medicine, 1941, 20, 341. athletes on 3 occasions after 5-mile runs. Hemo- 2. Ham, T. H., Studies on destruction of red blood cells. globinuria occurred in 4 of the 22 men who ran I. Chronic with paroxysmal noc- turnal hemoglobinuria: An investigation of the 26.2 miles; these urines were brownish to very mechanism of hemolysis, with observations on five dark burgundy red. Observations on 1 of these 4 cases. Arch. Int. Med., 1939, 64, 1271. athletes showed the occurrence of hemoglobinuria 3. Gilligan, D. R., Altschule, M. D., and Katersky, again after a 20-mile run. E. M., Studies of hemoglobinemia and hemo- 4. The occurrence of intravascular hemolysis globinuria produced in man by intravenous injec- tion of hemoglobin solutions. J. Clin. Invest., appeared to be unrelated to the age, body build, or 1941, 20, 177. standing position of the runners, or to the number 4. Bing, F. C., and Baker, R. W., Determination of of years or state of their training. hemoglobin in minute amounts of blood by Wu's 5. Hemoglobinemia and hemoglobinuria disap- method. J. Biol. Chem., 1931, 92, 589. peared in a few hours after the end of the run. It 5. Bing, F. C., Purification of benzidine, and an im- proved reagent, for estimating hemoglobin in blood. has been calculated that the amount of blood de- J. Biol. Chem., 1932, 95, 387. stroyed intravascularly is very small. 6. Malloy, H. T., and Evelyn, K. A., The determination 6. The plasma bilirubin was distinctly elevated of bilirubin with the photoelectric colorimeter. J. after the marathon run. The significance of this Biol. Chem., 1937, 119, 481. hyperbilirubinemia is discussed. 7. Hunter, F. T,, A photoelectric method for quantita- tive determination of erythrocyte fragility. J. 7. Quantitative studies revealed normal erythro- Clin. Invest., 1940, 19, 691. cyte fragility after the marathon runs. 8. Dickinson, L., Haemoglobinuria from muscular ex- 8. Transient albuminuria was observed fre- ertion. Tr. Clin. Soc. London, 1894, 27, 230. quently after the 2.6 to 5.1-mile runs and in every 9. Attlee, W. H. W., Haemoglobinuria following exer- instance after the marathon run. Transient uri- tion. Lancet, 1937, 1, 1400. 10. Finny, C. M., Functional hemoglobinuria. Brit. M. J., nary excretion of formed elements occurred fre- 1926, 2, 685. quently after the shorter runs and in almost every 11. Hantzschel, K., Ein Fall von echter Marschhimo- instance after the marathon runs. globinurie. Med. Klin., 1939, 35, 976. 9. Hemoglobinemia sometimes accompanied by 12. Fisher, A. M., and Bernstein, A., March hemoglobin- hemoglobinuria occurs in man with sufficient fre- uria. A case report. Bull. Johns Hopkins Hosp., 1940, 67, 457. quency after strenuous runs to be considered 13. Jundell, I., and Fries, K. A. E., Die Anstrengungs- physiological under these conditions. This in- albuminurie. Eine Studie iiber die Einwirkung travascular hemolysis after strenuous runs is maximaler Korperanstrengungen (des Sports und comparable in its apparently benign nature to the des Trainings) auf die Nieren. Nord. med. Arkiv., albuminuria of exercise. The dif- 1911, 44, Afd. II. N:r 2, 1. importance of 14. Hastings, A. B., The physiology of fatigue. Physico- ferentiating this condition from pathological states chemical manifestations of fatigue in the blood. leading to the passage of red urine is obvious. Public Health Bull., 117, 1921. 15. Ottenberg, R., and Fox, C. L., Jr., The rate of re- We wish to acknowledge the cooperation of Dr. An- moval of hemoglobin from the circulation and its drew W. Contratto of the Department of Hygiene, renal threshold in human beings. Am. J. Physiol., Harvard University, Mr. Jaakko J. Mikkola, Track 1938, 123, 516. Coach, Harvard University, Mr. Thomas J. Kanaly of 16. Albu, A., and Caspari, W., Bericht fiber die Unter- the Boston Athletic Association, and Mr. William T. suchungen an den Dauergehern beim Distanzmarsch INTRAVASCULAR HEMOLYSIS OF EXERCISE 869

Dresden-Berlin am 18/19 Mai 1902. Deutsche med. struction and regeneration after a period of train- Wchnschr., 1903, 29, 252. ing. J. Exper. Med., 1923, 37, 207. 17. Feigl, J., Chemische Blutuntersuchungen an der 28. McMaster, P. D., Broun, G. O., and Rous, P., Studies Teilnehmern eines Armee-Gepackmarsches. I. on the total bile. I. The effect of operation, exer- tJber Umsatz und Ausscheidung von Blutfarbstoff. cise, hot weather, relief of obstruction, intercur- Hamoglobinamie, Hamatinamie und Hamoglobin- rent disease and other normal and pathological in- urie. Biochem. Ztschr., 1916, 76, 88. fluences. J. Exper. Med., 1923, 37, 395. 18. Foerster, A., Ueber Marschhamoglobinurie. Miinchen. 29. Chang, H. C., and Harrop, G. A., Jr., The deter- med. Wchnschr., 1919, 66, 554. mination of the circulating blood volume with car- 19. Watson, E. M., and Fischer, L. C., Paroxysmal bon monoxide. J. Clin. Invest., 1928, 5, 393. "march" hemoglobinuria. With a report of a case. 30. Keys, A., and Taylor, H., The behavior of the plasma Am. J. Clin. Path., 1935, 5, 151. colloids in recovery from brief severe work and 20. Hamburger, L. P., and Bernstein, A., Chronic he- the question as to the permeability of the capil- molytic anemia with paroxysmal nocturnal hemo- laries to proteins. J. Biol. Chem., 1935, 109, 55. globinuria. Am. J. M. Sc., 1936, 192, 301. 31. Kaltreider, N. L., and Meneely, G. R., The effect of 21. Witts, L. J., The paroxysmal haemoglobinurias. exercise on the volume of the blood. J. Clin. Lancet, 1936, 2, 115. Invest., 1940, 19, 627. 22. Porges, O., and Strisower, R., Offizielles Protokoll 32. Hurtado, A., Studies at high altitude. Blood ob- der k.k. Gesellschaft Aerzte in Wien. Ein Fall servations on the Indian natives of the Peruvian von Marschhamoglobinurie. Wien. klin. Wchnschr., Andes. Am. J. Physiol., 1932, 100, 487. 1913, 26, 193. 33. Barron, E. S. G., Bilirubinemia. Medicine, 1931, 10, 23. Porges, O., and Strisower, R., tVber Marschhamo- 77. globinurie. Deutsches Arch. f. klin. Med., 1914, 34. Leube, W., Ueber die Aussheidung von Eiweiss im 117, 13. Harn des gesunden Menschen. Arch. f. path. Anat. 24. Schellong, F., Untersuchungen uiber Marschhamoglo- u. Physiol. u. f. klin. Med., 1878, 72, 145. binurie; ihre Beziehung zur Kaltehamoglobinurie 35. McFarlane, A., The presence of albumin and casts in und orthostatischen Albuminurie. Ztschr. f. d. ges. the urine of football players. M. Rec., 1894, 46, exper. Med., 1923, 34, 82. 769. 25. Broun, G. O., Blood destruction during exercise. II. 36. Frenkel-Tissot, H. C., Zur Frage der sportlichen Demonstration of blood destruction in animals ex- Albuminurie, besonders bei Skifahrern. Ztschr. f. ercised after prolonged confinement. J. Exper. klin. Med., 1921, 90, 54. Med., 1923, 37, 113. 37. Edwards, H. T., Richards, T. K., and Dill, D. B., 26. Broun, G. O., Blood destruction during exercise. III. Blood sugar, urine sugar and urine protein in ex- Exercise as a bone marrow stimulus. J. Exper. ercise. Am. J. Physiol., 1931, 98, 352. Med., 1923, 37, 187. 38. Covian, F. G., and Rehberg, P. B., Uber die Nieren- 27. Broun, G. O., Blood destruction during exercise. IV. funktion wahrend schwerer Muskelarbeit. Skan- The development of equilibrium between blood de- dinav. Arch. f. Physiol., 1936, 75, 21.