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







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






Chapter XV



Diseases of the Blood and Blood-Forming Organs

Maurice B. Strauss, M.D.

Disorders of the hematopoietic and lymphatic systems did not constitute a major problem of the Medical Department during World War II. Nevertheless, over 7 percent of all accessions to the Army Institute of Pathology, Washington, D.C., received between 7 December 1941 and 2 September 1945, numbering almost 5,000 specimens, were classified under this heading.1 For many of the diseases under consideration, the problems of diagnosis, treatment, and ultimate prognosis differed in no significant way from the same problems in civilian life. However, the experience of the members of the Medical Department revealed eight noteworthy observations in this field, as follows:

1. Severe anemia (p. 398) was rarely encountered among soldiers who had had repeated attacks of malaria in contrast to its frequent occurrence in infected civilian populations.

2. Although hookworm infection (p. 398) was common in certain areas where U.S. troops were stationed, severe anemia-the outstanding cause of disability in hookworm disease in civilians-was rare.

3. Eosinophilia (p. 399) as high as from 67 to 83 percent, with total leukocytes numbering between 50,000 and 60,000, were encountered in both hookworm infection and schistosomiasis and sometimes created a diagnostic problem in differentiation from eosinophilic leukemia.

4. In a small number of cases, Atabrine (quinacrine hydrochloride) was implicated as a causative factor in apalstic anemia (pp. 401-404).

5. Infectious mononucleosis (p. 405) was recognized in troops more often than in civilian populations and at times reached minor epidemic proportions. Spontaneous rupture of the spleen was observed at least seven times in infectious mononucleosis.

6. Despite the liberal and widespread use of sulfonamide drugs, agranulocytosis (p. 406) was uncommon and rarely fatal.

7. Nitrogen mustard (p. 406), a chemical warfare agent, was shown to have much promise in the treatment of lymphomatous disease.

1Custer, R. P., and Miller, M. H.: Lymphatic and Hematopoietic Disease in the United States Army During World War II: General Survey and Consideration of Histopathologic Diagnosis. [Professional paper.]


8. Aplastic anemia (p. 412) was produced in a goodly number of individuals exposed to irradiation from atomic bomb explosions at Hiroshima and at Nagasaki, Japan.


Anemia due to the destruction of parasitized erythrocytes was present to some degree in all cases of malaria. A single severe paroxysm of falciparum malaria might reduce the number of erythrocytes by 1 million cells per cubic millimeter. Although such severe hemolysis was uncommon in vivax infections, a significant degree of anemia frequently occurred. It is therefore of interest that anemia from malaria was so rarely encountered among the troops of the United States. In part, this may be explained by the fact that treatment was generally prompt so that repeated paroxysms did not occur in any single attack, although recurrent attacks were general. Of 435 soldiers evacuated from islands in the South Pacific Area and admitted to Harmon General Hospital, Longview, Tex., with recurrent malaria, the average number of attacks before admission was 5.7.2 Of these 435 soldiers, there were 287 whose records contained adequate data on the hematopoietic status. Only three patients had less than 13.0 gm. and only 26 had less than 14.0 gm. of hemoglobin per 100 cc. of blood. The erythrocytes of seven men numbered less than 4.0 million per cubic millimeter and in 72 men the number was under 4.5 million per cubic millimeter. In connection with this conspicuous absence of anemia, it may be mentioned that the spleen was palpable in only 23 percent of the acute attacks, and then only transiently. This also may be related to the prompt institution of therapy in each attack.


Hookworm infection3 was acquired by a goodly number of troops in the combat areas of the Pacific. However, signs or symptoms (other than eosinophilia) which could be attributed to the infection alone were almost never found. Among civilian populations with hookworm infection, hypochromic anemia is the outstanding feature and, in a majority of patients, is the chief cause of disability. Rhoads, Castle, Payne, and Lawson4 clearly showed (1) that this anemia is due to a deficiency of iron brought about by the chronic loss of blood, by defective diets, and by gastrointestinal changes; (2) that it may be corrected by therapy with iron without removing the

2Gordon, H. H., Lippincott, S. W., Marble, A., Ball, A. L., Ellerbrook, L. D., and Glass, E. W., Jr: Clinical Features of Relapsing Plasmodium Vivax Malaria in Soldiers Evacuated From the South Pacific Area. Arch. Int. Med. 75: 159-167, March 1945.
3See also chapter V, pp. 146-153.
4Rhoads, C. P., Castle, W. B., Payne, G. C., and Lawson, H. A.: Hookworm Anemia; Etiology and Treatment With Especial Reference to Iron. Am. J. Hyg. 20: 291-306, 1934.


parasites; and (3) that the removal of the hookworms does not cause a prompt remission of the anemia. Although a great many writers indicate a belief that the presence or absence of anemia in hookworm infection depends on the number of worms, Scott5 points out that it is the host's reserves of iron that determine the onset and degree of the anemia. Two studies are available among troops infected with hookworm-hemoglobin and determination of the number of erythrocytes. In 100 malarial and surgical patients infected with hookworm, not a single case had signs or symptoms which could be attributed to the infection alone, and only 10 patients had fewer than 4.0 million erythrocytes per cubic millimeter. In these, the anemia could be related to other causes.6 Examinations of the blood of 1,000 soldiers, 93 of whom had hookworm infection and 907 of whom did not, showed that 19 men (20 percent) of the infected patients had fewer than 4.0 million erythrocytes per cubic millimeter and that 34 percent had less than 80 percent hemoglobin. In the nonparasitized group, 15 percent had fewer than 4.0 million erythrocytes per cubic millimeter and 37 percent less than 80 percent hemoglobin. The smallest number of erythrocytes encountered was 3.1 million per cubic millimeter and the lowest hemoglobin was 60 percent.7 Although it is probable that the infection of troops was not heavy, the general absence of signs and symptoms-and of anemia-can best be ascribed to the excellent nutritional state of the men and the adequacy of their diet.


Eosinophilia of moderate degree in both hookworm infection and schistosomiasis has long been recognized. Extreme eosinophilia is mentioned in many textbooks on these disorders. Ashford, Payne, and Payne, in 1933,8 reported a case of uncinariasis with 87 percent eosinophils and a total count of 41,200 leukocytes per cubic millimeter and another case with a total count of 78,400 leukocytes per cubic millimeter, 68 percent of which were eosinophils.

Among troops, it was noted that both eosinophilia and leukocytosis generally reached a maximum between 3 and 4 months after hookworm infection.9 The highest percentage of eosinophils was 67 percent of 28,000 leukocytes; the highest number was 47,000 in a patient with 48 percent eosinophils. Eosinophilia may persist for a year or more, although the

5Scott, J. A.: Iron-Deficiency Anaemias. Lancet 1: 357, 549, 1938; also Am. J. Hyg. 26: 455-505, 1937.
6Most, H., Hayman, J. M., Jr., and Wilson, T. B.: Hookworm Infections in Troops Returning from the Pacific. Am. J.M. Sc. 212: 347-350, September 1946.
7Essential Technical Medical Data, U.S. Army Forces, South Pacific Area, dated 1 Feb. 1944.
8Ashford, B. D., Payne, G. C., and Payne, F. K.: Acute Uncinariasis From Massive Infestation and its Implications. J.A.M.A. 101: 843-847, 1933.
9(1) Liebow, A. A., and Hannum, C. A.: Eosinophilia, Ancylostomiasis, and Strongyloidosis in the South Pacific Area. Yale J. Biol. & Med. 18: 381-403, May 1946. (2) Whitehouse, F. R.: Massive Eosinophilia in Uncinariasis; Report of Two Cases. Gastroenterology 9: 303-306, September 1947.


leukocytosis generally disappears sooner. It should be pointed out that both leukocytosis and eosinophilia of marked degree may occur before hookworm ova can be demonstrated in the stools, thereby creating a diagnostic problem. L÷ffler's syndrome, tropical eosinophilia, and eosinophilic lung, which were reported, may well represent instances of hookworm infection or of schistosomiasis. Wright and Gold10 have reported from Camp Blanding, Fla., nine cases of creeping eruption (caused by Ancylostoma braziliense) with transient migratory pulmonary infiltration and eosinophilia. In schistosomiasis, Billings and his coworkers11 reported 83 percent eosinophils in a patient whose total leukocytes numbered 53,000 per cubic millimeter. Mason and his coworkers12 noted that 14 percent of their 300 cases of schistosomiasis had over 70 percent eosinophils and that 32 percent had between 50 and 70 percent eosinophils. Of the 300 patients, 8 percent had leukocytes numbering between 30,000 and 60,000 and 25 percent had between 20,000 and 30,000. Thus, it is apparent that both hookworm infection and schistosomiasis may present a picture of the blood suggestive of eosinophilic leukemia.


It was pointed out by Lt. Col. Richard P. Custer, MC, of the Army Institute of Pathology,13 that, during the first year of the war in the Pacific, bitter experience proved that malaria could neutralize U.S. combat forces more rapidly and more effectively than enemy fire. Guadalcanal, New Guinea, and certain areas of the China-Burma-India theater were among the most highly malarious regions of the world. Control measures for malaria could not be put into effect so long as the military situation remained unstabilized and U.S. soldiers were living and fighting in the jungle. Although mosquito repellents and protective clothing afforded a small measure of protection, it was not until control of malaria by suppressive-drug therapy was introduced in November 1942 in Guadalcanal, and subsequently employed in New Guinea, that the inroads of the disease began to diminish.14 Within a year after the beginning of suppressive-drug therapy, its value was so thoroughly established that it was intensively used in all malarious regions of the Pacific. This intensive use began in November 1942.

During the next year, medical officers in the Southwest Pacific Area called attention to a characteristic cutaneous lesion which was occurring in

10Wright, D. O., and Gold, E. M.: Loeffler's Syndrome Associated With Creeping Eruption (Cutaneous Helminthiasis). J.A.M.A. 128: 1082-1083, 11 Aug. 1945.
11Billings, F. T., Winkenwerder, W. L., and Hunninen, A. V.: Studies on Acute Schistosomiasis in Philippine Islands; Clinical Study of 337 Cases With a Preliminary Report on Results of Treatment With Fuadin in 110 Cases. Bull. Johns Hopkins Hosp. 78: 21-56, January 1946.
12Mason, P. K., Daniels, W. B., Paddock, F. K., and Gordon, H. H.: Latent Phases of Asiatic Schistosomiasis. Arch. Int. Med. 78: 662 -678, December 1946.
13Custer, R. P.: Aplastic Anemia in Soldiers Treated With Atabrine (Quinacrine). Am. J.M. Sc. 212: 211-224, August 1946.
14Dieuaide, F. R.: Personal communication.


soldiers who had been evacuated from New Guinea and adjacent islands.15 This lesion, termed "atypical lichen planus," but which also included eczematoid reactions, was associated with Atabrine therapy so frequently that an etiological relation between the drug and the skin lesion seemed probable.

During 1943 and 1944, it was noted that the incidence of aplastic anemia in the Southwest Pacific and the South Pacific Areas and in the China-Burma-India theater was considerably higher than in all the other foreign theaters and also higher than in the continental United States.16 A study of aplastic anemia made at the Army Institute of Pathology is quite revealing (chart 4). Throughout the entire period from January 1942

CHART 4.-Incidence of aplastic anemia among U.S. Army troops, 1942-45

to June 1945, the incidence of aplastic anemia in the entire Army-exclusive of the South Pacific, Southwest Pacific, and China-Burma-India-remained below 0.18 cases per 100,000 troops; whereas, in the areas just cited, it rose steadily from early 1943 to the end of 1944, reaching a peak of 2.84 cases per 100,000 troops in the period from July to December 1944, at a time when the rate for the rest of the Army was 0.04 cases per 100,000.

15Bull. U.S. Army M. Dept. 4: 653-659, December 1945.
16See footnote 13, p. 400.


The beginning of this increased incidence in the South Pacific and Southwest Pacific Areas and in the China-Burma-India theater in 1943 coincided with the period of the increased use of Atabrine in these areas. The peak in 1944 coincided with the period when suppressive-drug therapy was in extensive and widespread use. The incomplete figures for the first half of 1945 are of dubious significance. If borne out by complete data, they might reflect the diminished use of Atabrine as other measures to control malaria became effective and might reflect as well a smaller number of persons taking Atabrine as a result of troops being moved out of the more malarious areas in these theaters.

The cases of aplastic anemia occurring in 1943 and 1944 were further analyzed.17 Among the 20 cases from nonmalarious regions, 2 followed arsenical treatment for syphilis, 3 occurred after irradiation of malignant tumors, 4 were ascribed to sulfonamide medication, and in 11 patients no cause could be demonstrated. The 47 cases that occurred in the South Pacific and Southwest Pacific Areas and in the China-Burma-India theater had all received suppressive doses of Atabrine over periods ranging from 1 to 34 months; the majority ranged between 4 and 9 months. The 34-month case was unique, however, and the patient apparently had not adhered to the prescribed regimen, as his record showed repeated attacks of tertian malaria which required active treatment.

There was specific mention of overdosage before the onset of illness in six cases. Four men had increased the dose to 0.2 gm. daily, one for 3 weeks, another for 6 months, a third for 8 months, and the last for an unspecified time. Another soldier took between 20 and 30 tablets during the 4 days preceding onset of symptoms; still another was said to have ingested massive doses for 3 weeks before. Other instances of overdosage probably occurred, but were either not known or not recorded. For example, men on patrol or detached service were given supplies of Atabrine adequate for estimated time away from their units and, although instructions in its use were given, its administration was unsupervised.

Seven of the group treated with Atabrine had also taken sulfathiazole or sulfadiazine. In two instances, it was not possible to exclude sulfonamide drugs as a precipitating factor, but in the others it was obvious that the drugs were administered only after the onset of the anemia and in two cases aggravated the purpuric manifestations.

Atabrine dermatitis complex, the term proposed to designate atypical lichen planus, preceded aplastic anemia in 20 patients. In these, topical applications of various sorts were used-and light roentgenologic treatment was given in one instance-but were not regarded as significant causal factors.

A few patients had taken aspirin for headache, which was frequently a feature of the prodromal period. The others had received no drugs, nor could a history of exposure to other etiological agents be elicited.

The study of case records of soldiers who developed aplastic anemia after taking Atabrine revealed no significant differences in the manifestations from those encountered in aplastic anemias of varied etiology.

17See footnote 13, p. 400.


The most common complaint of the prodromal period was weakness, often associated with headache, vertigo, and dyspnea. A hemorrhagic tendency was frequently noted early, evidenced by bleeding gums after brushing teeth, nosebleed, "bruising easily," or the spontaneous appearance of ecchymoses in the skin. Dimness of vision, noted first by two of the patients, was found to be the result of intraocular hemorrhage. In the majority of cases, the onset was gradual, several weeks to a month or even more elapsing before the man reported to sick call. The prodrome was of brief duration in six cases and the onset precipitous in three in which an acute febrile state was the first indication of illness.

The early manifestations of aplastic anemia were often unnoticed in the group of 25 patients who were being treated for skin lesions until individuals among them were observed to be rather pallid and a blood count disclosed a reduction in the formed elements of the blood. In no instance did the anemia precede the dermatitis.

Almost invariably the course was marked by fever and hemorrhagic phenomena, the location and extent of the hemorrhages accounting for differences in symptoms. For example, hemiplegia, convulsions, and coma were noted in patients with intracranial bleeding; in contrast, hematuria and sometimes pain were associated with hemorrhage of the urinary tract. Of four patients, one was known to have survived at least 2 months; a second, 4 months; and the other two, 7 months.

Hematological findings-Rather complete series of blood counts were submitted in most cases, and the diagnosis of aplastic anemia was confirmed by sternal biopsy in 25.

In four patients, blood counts were performed very soon after the onset and showed that the number of erythrocytes was fairly well maintained (4.0 to 5.0 million per cubic millimeter), whereas the numbers of leukocytes and platelets had fallen sharply, the lowest being 1,500 and 40,000 per cubic millimeter, respectively. This is to be expected when one recalls the relatively long life (120 days or more for erythrocytes as opposed to from 3 to 5 days for neutrophils and platelets) of the mature erythrocytes and that most of the residual cells found in the bone marrow at necropsy are their progenitors.

The other 21 patients were already anemic when first examined, and initially the erythrocytes numbered as few as 600,000 per cubic millimeter, although the average range was from 1.5 to 3.0 million per cubic millimeter. The anemia was generally of the normocytic, normochromic type. In some cases, it seemed to be macrocytic, but this is questionable since it was reported only in instances where the anemia was severe; that is, in the range where hemoglobin readings are apt to be very inaccurate. Furthermore, no hematocrit values were available in these cases as a check on the presence of macrocytosis. Nucleated erythrocytes almost never appeared in the peripheral blood. Reticulocyte counts over 1 percent were rarely recorded and they were frequently reported as either none or less than 0.1 percent. Anisocytosis and poikilocytosis were seldom marked except in the few cases showing evidence of regeneration. It was possible with frequently repeated blood transfusions to restore the number of erythrocytes and the amount of hemoglobin to normal or nearly so in 16 patients and to effect


an improvement in nearly all. A few patients became progressively more anemic despite transfusions, and the majority did not maintain their improved status.

The initial leukocyte count was seldom over 3,000 per cubic millimeter and usually fell between 1,000 and 3,000 being normal (7,200) in only one case. Values below 1,000 per cubic millimeter were noted in five instances. The counts fluctuated somewhat, and occasionally rose following transfusions, but the general trend was toward lower levels as the disease progressed, with little tendency to increase significantly even during clinical remissions. The leukopenia was due primarily to failure of the bone marrow to produce granulocytes, but when the total count was greatly reduced the relative lymphocytosis actually represented a lymphocytopenia. This was also the case with respect to the monocytes. Absolute agranulocytosis was frequently recorded, but a general scan of the blood films usually disclosed a few neutrophils, most of them mature forms. Immature granulocytes were occasionally seen early in the course of the disease, but rarely thereafter.

The number of platelets was significantly low in all but 4 of the 49 cases in which counts were done. In three of these exceptions, the platelets were counted once and it was then early in the course of the disease. In the fourth, two counts were recorded as 263,000 and 154,000 per cubic millimeter. Necropsy disclosed extensive visceral hemorrhages in every one of these cases, however, and the marrows were devoid of megakaryocytes, indicating that platelets must have been subsequently much reduced. In general, the counts ranged between zero and 100,000 per cubic millimeter, with most of them below 50,000 at some time during the disease.

The coagulation time of the blood was but little altered, but clot retraction was either slow or absent. Bleeding time was sometimes normal even though the tourniquet test was positive. In other cases, the bleeding time was prolonged; in one patient, blood oozed for several days from the prick caused by the lancet.

Examination of the bone marrow in all cases showed the marrow to be markedly depleted of all normal hematopoietic elements. Usually the residual cells of the marrow belonged to the erythropoietic series and were in the late erythroblastic and normoblastic stages. The few granulocytes were generally stab or segmented forms, although occasionally small foci of myelocytes or even younger elements were encountered. Megakaryocytes were either absent or very sparsely distributed and when present often showed evidence of degeneration.

Although the evidence now makes it appear that Atabrine was etiologically significant in the production of bone marrow aplasia, it must be pointed out most emphatically that even the highest rate, 2.84 cases of aplastic anemia per 100,000 troops, was of trivial significance in comparison with both malarial morbidity and mortality before the introduction of suppressive-drug therapy.



Infectious mononucleosis has been a commonly diagnosed disorder among troops both in the continental United States and in theaters overseas. During 1944, the rate of admission for this disease was 97 per 100,000 troop strength. This high incidence of the disease may be explained by a number of factors. First, a routine examination of the blood was made on all Army patients admitted to hospitals, while in civilian life many of these patients would have been treated at home without the benefit of a hematological study and, therefore, their disease would have gone undiagnosed. Secondly, infectious mononucleosis is a disease which has the highest incidence in youth.

Contratto18 found that 1.5 percent of the total medical admissions to Stillman Infirmary, Harvard University, Cambridge, Mass., were due to mononucleosis, whereas only 0.36 percent of the total medical admission to the Peter Bent Brigham Hospital, Boston, Mass., were so diagnosed. The disease was definitely epidemic in certain areas. In the Caribbean Defense Command during the last 4 months of 1944, 92 cases of infectious mononucleosis were admitted to the 368th Station and the 262d General Hospitals.19 In the United States, 300 cases were observed within a period of 20 months at several large hospitals of the Army Air Forces.20 At the Station Hospital, Camp McCoy, Wis., between December 1943 and May 1944, 340 subclinical cases were discovered while 26 clinical cases were being treated.21

No evidence was accumulated during the war which would indicate that the severity of infectious mononucleosis was increased when it appeared in epidemic form. However, spontaneous rupture of the spleen, a complication which has been reported only three times in civilian life,22 occurred at least seven times in the Army.23 A total of 44 instances of spontaneous rupture of the spleen have been collected by the Army Institute of Pathology from 7 December 1941 to 2 September 1945.24 The cause of rupture could not be determined in 5 patients; in 22, it was due to malaria; in 7, to mononucleosis; in 5, chronic congestive splenomegaly (Banti's syndrome); in 3, to torsion; and in 2, to leukemia.

18Contratto, A. W.: Infectious Mononucleosis; A Study of One Hundred and Ninety-Six Cases. Arch. Int. Med. 73: 449-459, June 1944.
19Essential Technical Medical Data, Caribbean Defense Command, for January 1945, dated 20 Feb. 1945. Inclosure 1, "Epidemic of Infectious Mononucleosis."
20Read, J. T., and Helwig, F. C.: Infectious Mononucleosis; An Analysis of Three Hundred Cases With Three Characterized by Rare Hematologic Features. Arch. Int. Med. 75: 376-380, June 1945.
21Vander Meer, R., Lutterloh, C. H., and Pilot, J.: Infectious Mononucleosis; Analysis of 26 Clinical and 340 Subclinical Cases. Am. J.M. Sc. 210: 765-774, December 1945.
22(1) King, R. B.: Spontaneous Rupture of Spleen in Infectious Mononucleosis; Report of a Case. New England J. Med. 224: 1058-1060, 19 June 1941. (2) Ziegler, E. E.: Infectious Mononucleosis; Report of a Fatal Case With Autopsy. Arch. Path. 37: 196-201, March 1944. (3) Darley, W., Black, W. C., Smith, C., and Good, F. A.: Spontaneous Splenic Rupture in Infectious Mononucleosis; A Case and Pathologic Report. Am. J.M. Sc. 208: 381-384, September 1944.
23Smith, E. B., and Custer, R. P.: Rupture of the Spleen in Infectious Mononucleosis; A Clinicopathologic Report of Seven Cases. Blood 1: 317-333, July 1946.
See footnote 23.



During 1942 and 1943, agranulocytosis was not separately indexed, and therefore data bearing on its incidence are not available. In 1944, a rate of only 1.0 per 100,000 troop strength was reported with a total of 8 deaths. During the entire period of hostilities, only 37 accessions diagnosed as agranulocytosis were received by the Army Institute of Pathology. Although precise figures were not available on how many soldiers received sulfonamide drugs, it has been estimated that over 80 percent of the wounded were so treated and often for considerable periods of time. The 15th Medical Hospital Center, in the European theater, reported on 300 to 400 admitted casualties of whom 75 percent received local sulfonamide and over 80 percent received oral sulfonamide medication. This low incidence of agranulocytosis was therefore rather remarkable. One explanation offered was that repeated examinations of blood were made routinely on patients receiving sulfonamides in hospitals, and medication was stopped at the first indication of leukopenia.


Mustard gas, bis (▀-chloroethyl) sulfide, which was extensively employed as a poison gas in 1918 and which was known to have remote effects on hematopoietic tissues, was studied further after World War I. Its adverse effect on leukopoietic tissues and on the growth of experimental tumors was demonstrated. With the advent of World War II, biological research on gases used in war was resumed and attention directed to the nitrogenous analogs of mustard gas, the bis- and tris- (▀-chloroethyl) amines, known as nitrogen mustards.

It was soon determined that these compounds owed their biological activity to their ability to undergo intramolecular cyclization in a polar solvent to form the highly reactive ethylenimonium cation.25 This cation alkylates the functional groups of many compounds of biological importance, and it is likely that the basic mechanism of the cytotoxic action of the mustards is through such a reaction with a vital cellular constituent.

In general, it may be stated that the susceptibility of cells to the mustards is proportional to the proliferative activity of the cells. Thus, the blood-forming organs and the gastrointestinal mucosa are first and most profoundly affected (as is also the case with irradiation, see p. 412). Following the intravenous administration of the nitrogen mustards to experimental animals and to man, lymphocytopenia, granulocytopenia, and thrombocytopenia occur and, to a less extent, anemia. Marked effects on the blood may be obtained with doses of such a small size that nausea and vomit-

25Gilman, A., and Philips, F. S.: The Biological Actions and Therapeutic Applications of the ▀-chloroethyl Amines and Sulfides. Science 103: 409-415, 5 Apr. 1946.


ing are the only gastrointestinal effects. Larger doses lead to diarrhea which may become hemorrhagic. The outstanding pathological lesions after a mustard has been administered consist of (1) lymphatic fragmentation (as early as 10 hours after injection), leading to atrophy of lymphatic structures, and (2) the early disappearance of all mitotic activity in the bone marrow, leading to depletion. With larger doses or prolonged administration, there was almost complete aplasia, and necrosis and desquamation of the gastrointestinal mucosa.

The marked effects of the mustards on lymphoid tissue coupled with the finding that actively proliferating cells are selectively vulnerable to the cytotoxic action of the mustards suggested the therapeutic use of these compounds in the treatment of neoplasms of lymphoid tissue. Because of its undesirable physical properties and extreme chemical reactivity, sulfur mustard is not suitable for parenteral administration. However, nitrogen mustards in the form of their hydrochloride salts are water-soluble crystalline compounds that can be readily dissolved in sterile saline for intravenous administration. Experiments on transplanted lymphosarcoma in mice revealed that dissolution of such tumors could be effected rapidly although the dose required bordered on the toxic, and the tumor invariably returned.26 The first clinical trial of the nitrogen mustards27 was conducted on a group of six patients in the terminal stages of various neoplastic diseases. In two cases of lymphosarcoma in which roentgenologic therapy had been discontinued, a rapid dissolution of large tumor masses followed a course of injections. The results were sufficiently encouraging to warrant further clinical experimentation. Since then, much clinical experience with nitrogen mustard has accumulated.

Although sporadic temporary remissions of symptoms, sometimes associated with regression of primary or metastatic lesions, have occurred in a variety of malignancies-including some of neural origin-the only statistically significant results were observed in undifferentiated bronchogenic carcinoma and in malignant lymphoma, particularly of the Hodgkin's type. In bronchogenic carcinoma, striking remissions of fever, pain, weakness, cough, and anorexia have occurred. At times, these conditions were accompanied by regression of pulmonary infiltrations and of the signs of obstruction of the superior vena cava. In general, individual remissions were of relatively short duration, measured in weeks rather than months, but often repeated courses of treatment have continued to produce remissions (figs. 42 and 43).

In Hodgkin's disease, rapid improvement in the patient's general condition-disappearance of fever, gain in weight, and increase in the number of erythrocytes and hemoglobin-was accompanied by regression in the

26Dougherty, T., Gilman, A., Goodman, L., and Dougherty, J.: 1942-43. [Unpublished work cited by Gilman and Philips, see footnote 25, p. 406.]
27Gilman, A., Goodman, L., Lindskog, G. E., and Dougherty, J.: 1942-43. [Unpublished work cited by Gilman and Philips, see footnote 25, p. 406.]


FIGURE 42.-Roentgenogram of chest of 24-year-old male with bronchogenic carcinoma before therapy with nitrogen mustard.

FIGURE 43.-Roentgenogram of chest of 24-year-old male (fig. 42), 52 days later, after three courses of nitrogen mustard at intervals of 4 weeks. Each course was followed by a remission of fever and constitutional symptoms. The first two courses consisted of 0.4 mg. of nitrogen mustard per kilogram of body weight; the third, of 0.6 mg. per kilogram. The duration of each remission was from 2 to 3 weeks.


size of palpable lymph nodes, by reabsorption of pleural effusions, and by dramatic disappearance of enlarged mediastinal tumors. Figures 44, 45, 46, and 47 show the roentgenographic appearance in a case of Hodgkin's granuloma and a case of lymphosarcoma before and after a course of treatment consisting of 0.1 mg. per kilogram of body weight daily by intravenous injection on 4 consecutive days. Remissions occurred more rapidly after therapy with mustard than following irradiation, but these remissions seemed to be of shorter duration, although this was not demonstrated in a large enough group to be of statistical significance. The response to repeated courses of treatment appeared to be the same as to the first course, although the duration might be longer, of similar length, or shorter.

Patients who have had repeated courses of irradiation, and who no longer obtain remissions from the largest dose of roentgen rays which the radiologist gives, might respond very well to the mustards and subsequently regain sensitivity to irradiation.

In contrasting mustard to roentgenologic therapy, it should be pointed out that nausea and vomiting seem to be more frequent and severe after the former and that leukopenia is less often seen after irradiation.

Mycosis fungoides responded in dramatic fashion to nitrogen mustard with relief of itching and involution of the skin lesions, leaving pigmented areas as the only clinically detectable residuum.

Although some remissions were observed in lymphosarcoma, mustard therapy has been disappointing in other types of lymphoma, both of the lymphoblastic and lymphocytic type. Acute leukemias and plasmoma (including myeloma) have not been benefited. Clinical results in both chronic myelogenic and lymphatic leukemia have not been satisfactory, although the peripheral blood and the picture of the bone marrow may show improvement.

Results similar to those just presented were reported by other investigators.28

Although some patients receiving nitrogen mustards were observed for a period of several years, the evaluation of the clinical status of this group of compounds will require many more years of careful study. As yet, there is no basis for assuming that the therapeutic efficacy of the nitrogen mustards is any greater than that of irradiation.

28(1) Ap Thomas, M. I. R., and Cullumbine, H.: Nitrogen Mustards in Hodgkin's Disease; Report on 21 Cases and 4 of Other Reticuloses. Lancet 1: 899-901, 28 June 1947. (2) Goodman, L. S., Wintrobe, M. M., Dameshek, W., Good, M. J., Gilman, A., and McLennan, M. T.: Nitrogen Mustard Therapy. Use of Methyl Bis (▀-chloroethyl) Amine Hydrochloride and Tris (▀-chloroethyl) Amine Hydrochloride for Hodgkin's Disease, Lymphosarcoma, Leukemia and Certain Allied and Miscellaneous Disorders. J.A.M.A. 132: 126-132, 21 Sept. 1946. (3) Jacobson, L. O., Spurr, C. L., Barron, E. S. G., Smith, T., Lushbaugh, C., and Dick, G. F.: Nitrogen Mustard Therapy. Studies on the Effect of Methyl Bis (▀-chloroethyl) Amine Hydrochloride on Neoplastic Disease and Allied Disorders of the Hemapoietic System. J.A.M.A. 132: 263-271, 5 Oct. 1946. (4) Osborne, E. D., Jordon, J. W., Hoak, F. C., and Pschierer, F. J.: Nitrogen Mustard Therapy in Cutaneous Blastomatous Disease. J.A.M.A. 135: 1123-1128, 27 Dec. 1947. (5) Philpott, O. S., Woodburne, A. R., and Waldriff, G. A.: Nitrogen Mustard in the Treatment of Mycosis Fungoides. J.A.M.A. 135: 631-633, 8 Nov. 1947. (6) Rhoads, C. P.: Nitrogen Mustards in the Treatment of Neoplastic Disease. J.A.M.A. 131: 656-658, 22 June 1946.


FIGURE 44.-Roentgenogram of chest of a 23-year-old male with Hodgkin's granuloma, before therapy with nitrogen mustard. Note the enlarged mediastinal and hilar nodes.

FIGURE 45.-Roentgenogram of chest of a 23-year-old male (fig. 44), 2 months after a single course of 0.4 mg. of nitrogen mustard per kilogram of body weight.


FIGURE 46.-Roentgenogram of chest of a 29-year-old male with lymphosarcoma, before therapy with nitrogen mustard.

FIGURE 47.-Roentgenogram of chest of a 29-year-old male (fig. 46), 2 weeks after a single course of 0.4 mg. of nitrogen mustard per kilogram of body weight.



The medical complications of exposure to the atomic bombing of Hiroshima and Nagasaki were reported in part by the Joint Commission for the Investigation of the Effect of the Atomic Bombs in Japan and abstracted by LeRoy.29

There were approximately 120,000 casualties in Hiroshima and 65,000 in Nagasaki, about one-sixth of whom were either killed instantly or died under circumstances in which no medical care was possible. These people were burned to death by the direct heat of the bomb, were crushed under demolished buildings, or were burned in the fires started by the bombing. It has been estimated that about one-seventh of the total number of casualties escaped significant mechanical injuries or burns but received sufficient irradiation to produce significant effects on the blood and blood-forming organs as well as on the germinal epithelium and the intestinal epithelium. It appeared that the effect of this extremely brief exposure (probably no longer than 1 second) was similar to that observed following massive roentgen irradiation of animals or the administration of nitrogen mustard gas; that is, the destruction of tissues in which active cell division occurred, such as the bone marrow, the gastrointestinal epithelium, and the germinal epithelium. In addition to ionizing radiation, consisting principally of gamma rays harder than those produced by any known electrical apparatus, alpha particles, beta particles, and neutrons were also produced, but the relation of these to the casualties is not clear. The intensity of the gamma radiation emitted by the bombs decreased as the inverse square of the distance from the source of irradiation. It is accordingly presumed that the most severe casualties were probably those closest to the center of detonation, although other factors, such as shelters, played a role.

Persons who were subjected to intense exposure and who were not immediately killed or who did not die shortly thereafter of burns or other injury began to have nausea and vomiting within a few hours after exposure. Fever, diarrhea, and leukopenia began from the second to the seventh day. Purpura and thrombocytopenia generally did not appear until at least the fourth day. Death usually occurred in from 4 to 10 days. Examination of the blood in some cases was stated to have shown a total absence of leukocytes and platelets. Red urine was also reported. At necropsy, widespread ulcerative, necrotic, and hemorrhagic lesions were found in the gastrointestinal tract, together with degeneration of all bone marrow elements. A photomicrograph of the spleen of a 24-year-old man who died 5 days after exposure is shown in figure 48. This shows disappearance of lymphocytes from the Malpighian corpuscles, cytolysis of lymphocytes, and fibrinoid changes in the subendothelial portion of the central artery. Figure

29LeRoy, G. V.: The Medical Sequelae of the Atomic Bomb Explosion. J.A.M.A. 134: 1143-1148, 2 Aug. 1947.


FIGURE 48.-Photomicrograph of spleen of 24-year-old man who died 5 days after exposure to the atomic bomb. The field shows disappearance of lymphocytes from Malpighian corpuscles, cytolysis of lymphocytes, and fibrinoid changes in subendothelial portion of central artery. X 400. (LeRoy, G. V.: J.A.M.A. 134: 1143-1148, 2 Aug. 1947.)

49 shows a section of the bone marrow of a 39-year-old man who died 7 days after the bombing. It will be noted that few normoblasts remain and that there is proliferation of reticuloendothelium and of many plasma cells and lymphocytes. These observations resemble those of Shouse, Warren, and Whipple30 in dogs receiving large, single doses of roentgen rays.

Individuals who received a less severe dose of irradiation had early nausea and vomiting but then presented no symptoms for at least a week, with the possible exception of diarrhea which occasionally began as early as 4 days after exposure. Leukopenia, anemia, and epilation generally began sometime between the end of the first week and the end of the first month; fever, thrombocytopenia with purpura, and ulceration of the mucous membrane of the mouth occurred in the third or fourth week; and anemia became manifest in from 1 to 4 weeks after the detonation. Death occurred

30Shouse, S. S., Warren, S. L., and Whipple, G. H.: Aplasia of Marrow and Fatal Intoxication in Dogs Produced by Roentgen Radiation of All Bones. J. Exper. Med. 53: 421-445, 1931.


FIGURE 49.-Photomicrograph of bone marrow of 39-year-old man who died 7 days after bombing. The section shows a proliferation of reticuloendothelium and many plasma cells and lymphocytes. There are a few normoblasts remaining. X 654. (LeRoy, G. V.: J.A.M.A. 134: 1143-1148, 2 Aug. 1947.)

in approximately 50 percent of the cases between 10 days and 6 weeks after the bombing. Laryngitis, pharyngitis, tonsillitis, gingivitis, and tracheal and female genital ulcerations developed in many of these patients and presented essentially the same appearance as that seen in malignant neutropenia from any cause. Petechiae and purpura appeared in the skin of almost all the patients, together with epistaxis, melena, metrorrhagia and hematuria in many. The loss of hair was particularly marked in the scalp. Regrowth of hair did not begin for several months in those patients who recovered. Leukocytes numbering as few as 100 cells per cubic millimeter were reported. Recovery was rare in those whose leukocytes numbered less than 600 per cubic millimeter, essentially all of which were lymphocytes. Anemia was present in all such patients and became progressively more pronounced but did not occur until several weeks after the maximum leukopenia was observed. Presumably, the anemia was due, partly, to bleeding. In association with the leukopenia, serious infection and septicemia were common. Apparently, the chief causes of death were pneumonia and infection. The


FIGURE 50.-Photomicrograph of bone marrow of 29-year-old man who died 29 days after the bombing. The section shows hypoplasia, a necrotic area, and bacteria. The majority of the cells are reticulum cells, plasma cells, and lymphocytes. X 230. (LeRoy, G. V.: J.A.M.A. 134: 1143-1148, 2 Aug. 1947.)

bone marrow (fig. 50) of a 29-year-old man who died 29 days after the bombing shows marked hypoplasia, a necrotic area, and bacteria. The cells were mostly reticulum cells, plasma cells, and lymphocytes. The spleen (fig. 51) of a 35-year-old woman who died 19 days after the bombing shows atrophy of the lymphoid tissues with karyolysis of some of the cells in the vicinity of the central arteriole. Hyalinlike material is deposited beneath the arteriolar endothelium.

Individuals with still less exposure may or may not have vomited on the day of bombing. They then were asymptomatic from 1 to 3 weeks. Diarrhea generally began from 2 to 5 weeks after the explosion as did epilation. Leukopenia was noted from the second to the fourth week, ulceration of the mucous membrane in the third and fourth weeks, and anemia from the third to fifth week. Death occurred in the second or third month. It is uncertain what proportion of these individuals succumbed, but those who died appeared to have an aplastic type of anemia. The hemorrhagic tendency in


FIGURE 51.-Photomicrograph of spleen of 35-year-old woman who died 19 days after the bombing. The section shows atrophy of the lymphoid tissue, with karyolysis of some of the cells in the vicinity of the central arteriole. The arterioles show a deposition of hyalinlike material beneath the endothelium. X 135. (LeRoy, G. V.: J.A.M.A. 134: 1143-1148, 2 Aug. 1947.)

these patients was never severe, leukocytes numbering less than 1,500 cells per cubic millimeter were rare, and ulceration of the mucous membrane tended to be transitory and not severe. Figure 52 is of the bone marrow of a 31-year-old man who died of bronchiectasis 14 weeks after the bombing. The section shows gelatinous marrow from the rib with a large focus of regenerating cells replacing the fat.

LeRoy (p. 412) has summarized the important pathological changes as follows: The bone marrow was badly damaged with an almost complete disappearance of all cells of the myelopoietic and erythropoietic series. In all but the most severely irradiated patients, regeneration commenced within a week to 10 days after injury. The type of regeneration of tissue varied considerably, and often the marrow appeared to be producing mainly macrophages, plasma cells, and lymphocytes 3 to 4 weeks after the injury with very few cells of the granulocyte or erythrocyte series, although in


FIGURE 52.-Photomicrograph of bone marrow of 31-year-old man who died of bronchiectasis 14 weeks after the bombing. The section shows gelatinous marrow, from the rib, in which there is a large focus of regenerating cells replacing the fat. X 115. (LeRoy, G. V.: J.A.M.A. 134: 1143-1148, 2 Aug. 1947.)

other cases well-marked hyperplasia of these elements occurred. Tissue from patients dying early showed an almost complete absence of lymphocytes in the lymph nodes, the spleen, and the thymus. Recovery of the lymphatic elements was evident after a few weeks although, in many patients who died several months after irradiation, the lymphatic tissues contained only a fraction of the usual number of adult-type lymphocytes.

It seems probable that, had adequate measures been available at Hiroshima and Nagasaki to combat serious infection by the use of antibiotics and to combat anemia with blood transfusions, the early death rate might have been considerably lessened. The evidence suggests that some of the very early deaths occurred from an overwhelming dose of gamma irradiation without conspicuous morphologic change at necropsy.

The sequence of events in those who survived this immediate effect appears to be: First, a destruction of the marrow and lymphoid elements; then, following the destruction of marrow, the development of leukopenia


and severe infection and the onset of thrombocytopenia and hemorrhagic phenomena; and finally, the development of anemia, with the destruction of the circulating erythrocytes by normal processes. Thus, some of the early deaths resulted from leukopenia and sepsis before conspicuous bleeding occurred. Others died with hemorrhagic manifestations (subarachnoid hemorrhage was commonly reported in one group) after the onset of thrombocytopenia but before the development of anemia. Finally, late deaths were generally due to anemia associated with aplasia of the bone marrow.

A preliminary report was issued by Tullis and Warren31 on the observations of animals exposed to the atomic detonation at Bikini. The earliest lesions to appear were hemorrhages-involving almost any tissue of the body and apparently associated with radiation injury-dilatation and congestion of the small blood vessels of the brain, heart, lungs, mesentery, bowel, and subcutaneous tissue. Hemorrhages into the kidney, often filling the kidney pelves, were occasionally seen. The lymph nodes became extremely hemorrhagic; the lungs in 80 percent of the fatal cases were mottled, dark red, and pink. The cut surface was wet and the bronchi were filled with abundant pink or white froth. The gastrointestinal tract showed ulcerative lesions of the large intestine. The bone marrow sometimes was pale and sometimes hyperemic. The lymphocytes in the peripheral blood were the first to fall, within a matter of a few hours. After a few days, the granulocytes decreased and at the end of from 1 to 2 weeks the erythrocytes declined. Microscopic examinations were not yet available when this paper was written.

31Tullis, J. L., and Warren, S.: Gross Autopsy Observations in the Animals Exposed at Bikini. J.A.M.A. 134: 1155-1158, 2 Aug. 1947.