Chapter 7
Rickettsial Diseases and Leptospirosis
Colonel O` Neill Barrett, Jr., MC, USA (Ret.), and Colonel Fred R. Stark, MC, USA
Rickettsial disease, especially epidemic typhus, has had an important effect on the outcome of military operations for centuries. Zinsser (1935, pp. 153, 283) noted: "Typhus had come to be the inevitable and expected companion of war and revolution; no encampment, no campaigning Army, and no besieged city escaped it ." "Epidemics get the blame for defeat, the generals the credit for victory. It ought to be the other way around."
Beginning in the 16th century and continuing for the next 300 years, typhus was to play a prominent part in all major military campaigns in Europe. Remarkably, however, epidemic typhus has never been a serious problem for the U.S. Army. There was little or no typhus in the American Civil War; in World War I a worldwide total of only 42 cases occurred in American troops, and in World War II only 104 cases of epidemic typhus and 787 cases of murine typhus were recorded (MD-PM7, pp. 178-82). Scrub typhus, however, has had a significant-although geographically localized -impact on American military activities, especially since 1942. During World War II, 6,685 cases were recorded (Smadel 1952, p. 639).
Section I. Scrub Typhus (Tsutsugamushi Fever)
Colonel O` Neill Barrett, Jr., MC, USA (Ret.)
HISTORY AND MILITARY SIGNIFICANCE
Although a "disease caused by the bite of small red insects" was described in China as early as the 3d century, B.C. (Zdrodovskii and Golinevich 1960, p. 354), and a Chinese work from the 16th century discusses a "sand mite referred to as a fever carrier," the first detailed written description of scrub typhus is Japanese, attributed to Hakuju Hashimoto in 1810. The disease was subsequently described in Formosa and Pescadores Islands but, surprisingly, was not reported from China. The first report in English was written in 1878 by Theodore Palm, a medical missionary in Niigata. Occasional cases were observed in Malaya (now West Malaysia) in the early 1920`s, but descriptions of the disease were primarily limited to Japan until the Allied involvement in Southeast Asia in World War II. Identification of the causative organism,
134
Rickettsia tsutsugamushi [orientalis], is attributed to Norio Ogata in the late 1920`s, although claims for its discovery were made by other Japanese investigators. Nagayo, Kawamura, and Hayashi are also well known for their research on rickettsial diseases (Audy 1968, pp. 30-70).
The vector of scrub typhus in Japan was found to be a trombiculid mite (chigger), Leptotrombidium akamushi. The species name is derived from the Japanese terms aka (red) and mushi (mite). The vector is also referred to as the dangerous mite, tsutsuga-mushi; hence the name "tsutsugamushi fever." For at least two centuries, a mushi-yaki (mite burning) ceremony has been held every January at Haguro Shrine, where effigies of the tsutsugamushi mite are burned, the original purpose being protection of farmers from the disease (Audy 1968, pp. 30-40).
During World War II, three separate episodes of scrub typhus in U.S. Army forces were recorded, two of major significance. In the SWPA (Southwest Pacific Area), there were 5,718 cases, including 284 in troops who reoccupied the Philip pines in late November 1944, and in the CBI (China-Burma-India) theater, 967 cases were documented. Cases of murine typhus were undoubtedly included in these figures, however, especially in the 125 cases reported from China. In the SPA (South Pacific Area), 32 cases were observed on Bougainville, New Georgia, and Espiritu Santo (MD-PM7, pp. 278-79).
The two most serious outbreaks in Army experience occurred in New Guinea following landings at Owi-Biak and Sansapor beachheads: by the end of 1944, there had been 2,500 cases in these areas. Although the case fatality rate was low (2 percent), morbidity was high as fever frequently lasted up to 20 days and convalescence was often prolonged (MD-PM7, p. 285). Another virulent outbreak occurred in 1944 on Goodenough Island, north of Papua, where Lt. Gen. (later Gen.) Walter Krueger had established the Sixth U.S. Army Headquarters. Approximately 40 cases of scrub typhus occurred among personnel of the 9th General Hospital who had helped clear an area of kunai grass (Imperata cylindrica) before construction of the hospital buildings. Several died, including one medical officer (MD-IM1, p. 524).
Although the typhus fevers, including scrub typhus, had been recognized in Burma as early as 1932, the first case of scrub typhus reported in the CBI theater was not from that country but from the 100th Station Hospital in Delhi, India. Still, Burma was the main focus of the disease in the theater (MD-IM1, p. 742). The five major peaks of scrub typhus there were primarily related to combat activities and not to a seasonal variation as had originally been supposed. The disease was widely distributed throughout the country, occurring along the entire length of the Ledo Road, most dramatically at Myitkyina, in the Fort Hertz district and around Lashio. The first cases were referred to as "CBI fever," although scrub typhus was suspected; this reluctance to diagnose FUO (fever of undetermined origin) as scrub tyhus occurred even among Japanese medical officers who had greater experience with the disease, as evidenced by captured Japanese medical reports (MD-PM7, pp. 292-97; MD-IM1, pp. 743-44). Audy (1968, p. 110) wrote: "The dead hand of tradition, the clinical and
135
epidemiological mental picture of `classical` tsutsugamushi disease along the rivers in Northwest Honshu" kept these "competent clinicians and epidemiologists from diagnosing a disease familiar at least by repute to every Japanese physician." Early confusion concerning diagnosis was resolved by a special team of the USATC (United States of America Typhus Commission) (MD-PM7, p.297). A detailed report of the clinical and pathologic manifestations of the disease in Assam and Burma is given by Sayen and associates (1946) in their evaluation of 616 cases seen at the 20th General Hospital.
The psychological impact of scrub typhus on the combat effectiveness of American and Chinese troops in Burma was out of proportion to its actual statistical importance as a cause of mortality and morbidity. Still, American physicians knew little about the disease and almost nothing about its epidemiology or prevention, and they lacked a specific therapy for it. Furthermore, scrub typhus was a serious disease among the troops in Burma; mortality was higher there than in any other area, probably because of the frequent occurrence of other diseases, including malaria and dysentery, and the general debility of troops in continuous combat for prolonged periods (Stone 1969, pp. 123n-124n). The annual scrub typhus mortality rate of 14.6 per 100,000 in the CBI theater was the highest for an infectious disease in any World War II theater of operations (MD-PM7, p.6). The mortality rate was approximately 4 percent among 113 American soldiers in Services of Supply units along the Ledo Road who developed the disease sporadically, and it was 16 percent among 105 patients evacuated by air from jungle combat in the mountains north of Myitkyina between March and May 1944 (MD-IM2, p.128). The patient`s pretyphus condition and the quality of early care were vital to the prognosis; patients given careful nursing and vigilant treatment of symptoms had the best chance of recovering.
Between 1945 and the beginning of hostilities in Korea in 1950, scrub typhus was not a significant problem for American military forces. However, one experience in Japan is noteworthy, primarily because of new epidemiological knowledge which resulted. Tsutsugamushi fever was thought to occur in grassy flood lands along river valleys in three prefectures of northwest Honshu. The disease had a definite seasonal occurrence, most cases being seen in July, August, and September, with a mortality rate greater than 20 percent (Smadel 1952, pp. 638, 647). However, during October and November 1948, 23 of 1,769 Army troops exposed to vectors in the military maneuver area at the foot of Mount Fuji were treated for scrub typhus, and it was estimated that an equal number of inapparent infections occurred (NAVY-MN). The vector for this outbreak was not definitely determined, but R. tsutsugamushi had previously been isolated by Wharton (1946) from Leptotrombidium pallida, a mite indigenous to the area.
Another outbreak in American military personnel in Japan occurred on the lower southwest slope of Mount Fuji in October 1953, where two regiments of the 3d Marine Division, Fleet Marine Force, Pacific, were undergoing training. Eleven days after entering the area, the first of 57 patients was admitted to the medical company dispensary. The clinical course was typical in these cases, including prompt response to Terramycin, and no complications or
136
deaths occurred (NAVY-MN). Although the vector was not determined, the organism was isolated from patients` blood.
Tsutsugamushi fever was not a problem for American or allied forces during the Korean war. Only eight confirmed cases occurred among United Nations personnel from 1951 to 1954. This is not surprising since before 1951 there were only occasional unconfirmed reports of the disease (Ley and Markelz 1961). The first two documented cases were reported by Munro-Faure and associates (1951) from the BCOF (British Commonwealth Occupation Forces) 29th General Hospital in Korea. Jackson and associates (1957) later demonstrated Rickettsia sutsugamushi in some of the Korean rodents they examined. The established vectors of scrub typhus are not known to occur in Korea or else are so rare that they have not been collected, although forms closely related to Leptotrombidium akamushi are plentiful. The absence or dearth of primary vectors is thought to be responsible for the low incidence of cases (Traub 1954).
EPIDEMIOLOGY
Although originally reported in Japan, scrub typhus now has been shown to exist throughout a large area of the Far East in a roughly triangular area bounded by Japan, India, and Australia (map 2). The term "scrub" is more of historical interest than of descriptive value, as the disease has a far wider distribution than along grassy islets and banks of fertile silt (as in Japan) or in scrub vegetation (as in West Malaysia); for instance, it was noted in suburban areas of Calcutta and in deep jungle in Burma during World War II (Audy 1968, p. 87). More recently, infective vectors have been found on the edge of sandy beaches and in deep jungle in Malaysia (Traub 1960; Traub, Frick, and Diercks 1950). Cadigan and coworkers (1972) demonstrated that over 73 percent of adult aborigines in West Malaysia`s deep jungle had antibodies to R. tsutsugamushi despite the apparently infrequent occurrence of clinical disease among them; their exposure probably occurred in the deep jungle itself rather than in cleared areas. Although conditions are exacting for the maintenance of an endemic area, areas now devoid of the disease could become new foci if the infective vector were introduced; this seems to have already happened in Queensland, Australia, and the disease probably could be established in other sites in both Australia and Madagascar (Audy 1968, p. 116). Therefore, a more proper name for "scrub" typhus would be mite-typhus or tsutsugamushi disease.
Rickettsia tsutsugamushi is an organism which naturally infects mammals-including mice, rats, and voles-and probably some birds as well. Man is only an accidental host, becoming exposed and infected when he enters an enzootic area where the mammalian reservoir and a suitable ectoparasite are present. Basic requirements for maintenance of an endemic area include suitable cover for the reservoir host and moisture and temperature conditions favorable for growth of the vector (MD-PM7, pp. 302-21).
137
138
Mites acquire the infection either transovarially or by settling on an infected host during the larval stage of development. The former mode is more important for perpetuation of the disease, since larvae settle on a potential host only one time (Zdrodovskii and Golinevich 1960, p. 367).
The disease is transmitted to man by a few species of larval trombiculid mites (chiggers) of the genus Leptotrombidium, the most important of which are L. akamushi and L. deliensis, two closely related chiggers. Other species including L. tosa, L. scutellaris, and Sch?ngastia indica also transmit the disease to humans (Smadel 1952, pp. 638-51; Faust and Russell 1964, pp. 773-76). The varying epidemiological host-vector patterns are shown in table 23.
These vectors represent a relatively small number of the many species of trombiculid mites found in enzootic areas. None of the others has been found to be naturally infected or to transmit the disease under laboratory conditions. A possible additional vector for human infection is L. arenicola, recently studied by Upham et al. (1971). This mite, which resembles L. deliensis, was found infesting rats trapped on a sandy beach during investigation of a scrub typhus outbreak near Singapore. Rickettsia tsutsugamushi was isolated from a significant number of pools of this species; to date it has not been described outside West Malaysia.???
The epidemiology of scrub typhus in Vietnam is not well defined. Before World War II only an occasional, undocumented case was reported. On the basis of geographic considerations of proper terrain and humidity (fig. 40), Le Gac and Arquie (1964) suggested that endemic areas would probably include
TABLE 23.-Areas of known occurrence of scrub typhus with identified hosts and vectors
139
Lang Son, Tan Hoa, and Son La in the North, Quang Tri, Dong Ha and "Cuvette Daden" in Central Vietnam, and Ben Cat in the South. Mountain jungles and areas around plantations were thought to be especially suitable areas; this speculation was partly confirmed by the experience of French forces fighting the Vietminh in Central Vietnam from 1949 to 1952. In 1949 and 1950, most of the fighting was along the coast or on the approaches to mountain areas and only 23 cases of scrub typhus were noted. From 1951 to 1952, however, combat moved into the Chaine Annamitique and 248 cases were encountered. In the last phase of the French-Indochina War, 5,708 cases of scrub typhus were reported, with 158 deaths; these last figures are probably lower than the actual numbers since only cases in hospitalized patients are included. Although Le Gac and Arquie suspected that L. deliensis was the vector, they did not prove it. During studies at Ben Cat in 1969, Upham* identified L. deliensis as a parasite of several small mammal species but was unable to determine whether the mites were infected.
Data concerning the incidence of scrub typhus in American military personnel in Vietnam are approximate at best, although it was first seen as early
_______
* R. W. Upham: Personal communication.
140
as 1962.* As was to happenfrequently, the first case occurred simultaneously with falciparummalaria. Following clearing of the malaria parasitemia, the patientremained febrile anddeveloped a typical rash, eschar, and positive OX-K titer. An estimated6-percent incidence ofscrub typhus occurred among patients hospitalized for malaria(USARV-MB, p. 17). During theearly combat period, the only diagnostic laboratory method availablewas the Weil-Felixreaction, known to be positive in only 50 percent of proven cases(Bozeman and Elisberg 1963).Many clinically recognized cases were treated with tetracycline withoutserologic confirmationand, in some instances, without hospitalization.
Although the total number of cases which occurred in Vietnam is not known, some comparative data are available. For example, three FUO studies, each accomplished with sophisticated bacteriologic, virologic, and serologic support, were reported on. Two of them provide data from U.S. Army hospitals in 1966. At the 8th Field Hospital in Nha Trang, Reiley and Russell (1969) reported a 14-percent incidence of scrub typhus among 94 cases of FUO. Deller and Russell (1967) reported 110 FUO cases from the 93d Evacuation Hospital at Long Binh, with an 8-percent incidence of scrub typhus. Only those cases of scrub typhus which were diagnostic FUO problems were studied; therefore, while one can estimate that 10 percent of all FUO cases were scrub typhus cases, no extrapolation can be made concerning total numbers of cases of the disease. The third study of febrile illness in Vietnam was done by Deaton (1969) at the U.S. Air Force Hospital, Cam Ranh Bay. Of his 306 cases, only 1 percent were scrub typhus. The FUO data from these three studies are summarized in table 24, which also includes data from a study by Cottingham et al. of indigenous and U.S. troops in II and III CTZ`s (Corps Tactical Zones) (discussed later in this chapter).
TABLE 24.-Incidence of scrub typhus among cases of fever of undetermined origin in Vietnam, 1966-67
In a review of serological tests performed by the 9th Medical Laboratory, researchers (Baker, McKinney, and Huxoll) studied 190 cases with diagnostic titers for scrub typhus during a 12-month period. Specimens were received from the entire country, and diagnosis was based on a fourfold rise in specific antibody
____________
0`Neill Barrett, Jr.: Unpublished data.
141
titer. Admitting a possible sampling error in the first 3 months, a significant correspondence between occurrence of disease and the rainy season was found. Further analysis revealed that 93 percent of cases occurred among infantry and artillery troops rather than support troops. This distribution contrasts with that of murine typhus and confirms previous French data (Le Gac and Arquie 1964) indicating presence of disease primarily in mountain jungle terrain. Extrapolating from Baker, McKinney, and Huxoll`s estimated sample source, approximately 2,000 cases of scrub typhus probably occurred during the 12-month period of observation.
ETIOLOGY, PATHOGENESIS, AND PATHOLOGY
The etiology and pathogenesis of scrub typhus are described in detail in several texts (Smadel 1952; Faust and Russell 1964). The etiologic agent, R. tsutsugamushi, is an obligate intracellular parasitic micro-organism appearing as short rods or diplococci, 0.3 to 0.5 I,m long and 0.2 to 0.4 gm wide. In the electron micrograph, a well defined limiting membrane enclosing protoplasmic material that contains dense granules is seen. Although the rickettsiae are rendered noninfectious by 0.1-percent formaldehyde solution, they remain viable for long periods when stored at - 70 0 C in appropriate protective media. They grow well in yolk sac tissue of embryonated eggs and in several types of tissue cultures (Smadel 1952, pp. 641-43).
In the experimental animal, rickettsiae initially multiply at the site of local inoculation and then become disseminated. The basic histologic lesion is a perivasculitis characterized by infiltration of monocytes, plasma cells, and lymphocytes with associated edema. Endothelial changes are uncommon except in the lungs where thromboses are frequently observed. In severe infections, focal necrosis of involved tissue may occur. In both experimental and human disease, lesions may be widespread, but they are most common in the skin, lungs, myocardium, and brain. In brain lesions, the focal perivascular reaction with proliferation of neuroglial cells is seen (Faust and Russell 1964, p. 774).
Recent studies concerning the pathogenesis of scrub typhus infection, especially in terms of latency and recrudescence, were performed by Kundin et al. (1964). They followed the course of infection in suckling and weaning mice, noting the effects of various routes of inoculation on the distribution of antigen, and also attempted to locate rickettsiae in mice at intervals after the initial infection. Their studies confirmed previous reports that susceptibility and length of survival varied with age of host, route of inoculation, and size of inoculum. An impressive finding was the observation that antigenic localization followed a characteristic pattern in all cases and that antigen was deposited in connective tissue of mesenchymal origin, especially loose areolar tissue, adipose tissue, and myeloid hematopoietic tissue. Infectivity titrations, however, showed that rickettsiae were present even in areas where antigen was not demonstrated by immunofluorescence. Tissue showing the highest degrees of infectivity included liver, kidney, brain, muscle, and spleen. In animals sacrificed 1 year after infection, organisms could be found in a similar distribution by infectivity titration
142
but not by immunofluorescence. Irradiated mice showed greater susceptibility to rickettsial infection with an increase in mortality, a short incubation period, and enhanced antigen deposition, supporting the hypothesis that X-radiation inhibits antibody response and renders cells more vulnerable to invasion.
The only two extensive pathological reports of scrub typhus in the literature represent experience from World War II. In the first study, done by Sayen et al. (1946) in conjunction with the clinical evaluation of disease at the 20th General Hospital in Burma, autopsies were performed in 28 of the 29 deaths caused by the disease. The second report, by Allen and Spitz (1945), reviews material collected by the Army Institute of Pathology and the USATC on 78 cases of scrub typhus in New Guinea.
The characteristic lesion, confirming the experimental data discussed above, was a perivasculitis involving arterioles, capillaries, and veins with eccentric infiltration by lymphocytes, plasma cells, and macrophages (Allen and Spitz 1945, p.609).???
Cardiac involvement was characterized by diffuse myocarditis, especially severe in the endocardium of the papillary muscles with the typical mononuclear infiltrate. There was no evidence of necrotizing arteritis, in contrast to a 17-percent incidence in epidemic typhus (Allen and Spitz 1945, pp. 611-13).
In 55 percent of scrub typhus cases, interstitial pneumonitis of varying degrees of intensity was found, characterized by marked dilatation of septal capillaries and interstitial infiltration by mononuclear cells as well as desquamation of alveolar epithelium and intra-alveolar hemorrhage in severe cases (Allen and Spitz 1945, p. 615; Sayen et al. 1946, p. 184).???
Changes in the nervous system were seen in all patients and were characterized by diffuse meningoencephalitis with perivascular cuffing of arteries, degeneration of ganglion cells, and focal hemorrhages in parenchyma and meninges (Allen and Spitz 1945, p. 626).
Evidence of early but definite glomerulonephritis was found in 30 percent of the cases studied. The common finding was moderate to complete ischemia of the glomerular capillaries, often with obliteration by platelet thrombi or enlarged endothelial cells (Allen and Spitz 1945, p. 636).
There are no pathologic data on the cases of scrub typhus in Vietnam since no deaths from the disease were documented there.
CLINICALMANIFESTATIONS
Scrub typhus is an acute febrile illness characterized by temperature elevation to 104 0 or 105 0 F, chills, malaise, and headache. The common but not invariable occurrence of an eschar and characteristic rash makes it more distinctive than other febrile illnesses seen in Vietnam. On the other hand, the disease may be atypical at onset, as evidenced by several FUO studies in which 1 to 14 percent of initially undiagnosed cases were caused by scrub typhus (see table 24). A comparison of the signs, symptoms, and routine laboratory studies from several detailed reports is given in table 25. The data of Sayen and associates are from the Burma experience in World War II; those of Cottingham et al. are
143
TABLE 25.-Comparison of clinical manifestations of scrub typhus from five studies
from indigenous Vietnamese units, including CIDG (Civilian Irregular Defense Group) forces; those of Deller and Russell and of Reiley and Russell from their respective FUO studies; and those of Hazlett from studies of an outbreak of 32 cases in a combat unit conducting search-and-destroy operations around an abandoned rubber plantation near Phan Rang.
Fever is almost invariably present and, by the third day of clinical illness, usually presents a classic spiking or "sawtooth" pattern with a return of temperature to normal or near normal between spikes (figs. 41 and 42). This characteristic pattern was seen in all 32 patients studied by Hazlett (1970). While a similar spiking pattern is seen in leptospirosis, the temperature in that disease rarely drops to 99 0 F between spikes (Reiley and Russell 1969, p. 38). Several authors report a relative bradycardia with the pulse rate disproportionate to the temperature elevations. However, a similar observation has been made both in dengue fever and malaria, and this finding is of little diagnostic value (Sayen et al. 1946; Hazlett 1970; Reiley and Russell 1969).
The eschar, when present, is an extremely valuable diagnostic aid. Classically it is a painless, nonpruritic lesion 1 to 2 cm in diameter with a black ne-
144
crotic center and halo of subacute inflammation (fig. 43). A typical eschar is seen in 33 to 85 percent of cases (Deller and Russell 1967; Hazlett 1970), but in some, because of perspiration and maceration, a crust may not form and the lesion appears only as a denuded papule. Eschars occur mostly on the extremities or trunk but may be found in either the axilla or genital area in about 20 percent of the cases (Hazlett 1970; Sheehy, Hazlett, and Turk 1973). A transient rash, macular or maculopapular, pink to red, nonconfluent and blanching, appears on the trunk after several days in about 50 percent of cases (fig. 44) (Reiley and Russell 1969). Lymphadenopathy occurs in almost all cases (Sayen et al. 1946, p. 171; Reiley and Russell 1969; Deller and Russell 1967; Hazlett 1970) and may be regional or generalized, usually appearing several days after onset of fever. The lymph nodes are usually tender and may be 1 to 3 cm in size. Splenomegaly is seen in about one-third of the cases but is generally mild. Hepatomegaly is not characteristic of the disease.
Headache is the most common symptom. The pain is usually severe and located in either the frontal or the retro-orbital area. Lumbar spinal punctures were not performed in any of the cases, so information concerning possible central nervous system involvement is unavailable. Cough, also a frequent
145
symptom (Sayen et al. 1946, p. 168; Reiley and Russell 1969; Cottingham et al.), is generally mild and nonproductive and not associated with pneumonitis.
Complications of scrub typhus were remarkably uncommon in the cases reported from Vietnam and, as previously noted, there were no deaths from the disease in American troops. This is in striking contrast to the 5-percent mortality rate in Sayen`s series and the 9-percent mortality rate in the CBI theater as a whole. In addition, clinical involvement of several organ systems was noted in the Burma experience: severe myocarditis was found in all cases at autopsy, about 12 percent of all patients had cardiomegaly, and 36 percent had definite electrocardiographic changes, including P-R interval prolongation and ST-T wave changes (Sayen et al. 1946; MD-PM7, p. 292). Only one case of myocarditis associated with scrub typhus was reported among American troops in Vietnam: Ognibene et al. (1971) described an 18-year-old white male who developed myocarditis which was self-limited, although it persisted well beyond the period of acute disease. The case was further complicated by the development of disseminated intravascular coagulation. This phenomenon had been described only once previously in an American in Vietnam. Chernof (1967) reported hypofibrinogenemia, prolonged prothrombin time, and mild thrombocytopenia in association with thrombophlebitis in a 22-year-old man with scrub
146
typhus. He was treated with heparin and recovered. The author properly presumed that the likely cause of this picture was "accelerated intravascular coagulation." This complication is well recognized in Rocky Mountain spotted fever and was probably a common occurrence in the lethal cases in Burma as suggested by the clinical and pathological evidence of widespread hemorrhages at autopsy (Sayen et al. 1946). There is also one case report of acute renal failure complicating scrub typhus in an individual with glucose-6-phosphate dehydrogenase deficiency (Whelton, Donadio, and Elisberg 1968). Mild renal involvement, characterized by a transient albuminuria and hematuria, was noted in 18 percent of the patients in Hazlett`s series (1970). The serious pulmonary and central nervous system complications reported in the Burma study were not seen in Vietnam.
Routine laboratory data are not helpful in the diagnosis of scrub typhus. Sayen et al. (1946, p. 179) described anemia as occurring during the first 2 weeks of the disease but did not cite an incidence. Furthermore, the coexistence of malaria, amebiasis, and hepatitis makes the association between the anemia and typhus unclear. On the other hand, Reiley and Russell (1969) reported anemia (hematocrit below 40) in 6 of 13 cases. In only two was the hematocrit below 35 percent. No apparent cause was found; however, the evaluation was limited.
147
The leukocyte count is generally normal although leukopenia and leukocytosis have been noted. Sayen and associates (1946, p. 179) found that the average total leukocyte count rose slightly during the second week and more def initely after the fourth week. Early leukopenias were observed in a few cases; a somewhat larger number of patients had considerable leukocytosis, with some counts as high as 20,000 in the second or third week. Associated disease and the lack of specific chemotherapy in this group may have affected the data. Reiley and Russell (1969) noted only five cases with leukocyte counts above 10,000 per mm3 and only one of these exceeded 15,000. In Hazlett`s larger series (1970) the leukocyte count varied in some patients from 2,750 to 26,000 per mm3 at different stages of the disease. Leukopenia with relative lymphocytosis was noted in 16 percent of cases between the third and tenth day of illness. Leukocytosis
148
was seen in 20 percent and usually occurred between the 13th and 24th day of illness. Leukocytosis may, therefore, be a late manifestation of untreated scrub typhus, while leukopenia is less characteristic than in dengue and malaria (Deller and Russell 1967; Reiley and Barrett 1971).
LABORATORY DIAGNOSIS
The laboratory diagnosis of scrub typhus can be made either by immunological techniques or by recovery of rickettsiae from laboratory animals inoculated intraperitoneally with blood from infected persons.
Of the immunological tests available, one of the oldest, simplest, and most widely used is the WFR (Weil-Felix reaction). This agglutination test does not directly involve rickettsial antigens. Instead, it depends on the fact that, in the course of some rickettsial diseases, antibodies are formed which "react fortuitously" with the polysaccharide O antigens of the X strains of the Proteus bacillus, which are referred to as Proteus OX strains. The test was originally described by Weil and Felix, who because of this cross reactivity mistakenly suspected that Proteus was the cause of typhus fever (Davis et al. 1973, pp. 898-910).
Three Proteus strains are used in the WFR: Proteus OX-19 for the diagnosis of epidemic and murine typhus and Rocky Mountain spotted fever, OX-2 for the spotted fever group but not the typhus fevers, and OX-K (Kingsbury strain) for tsutsugamushi fever. A significant rise in titer is not often seen before the 10th day of illness; the maximal titer usually occurs by the fourth week and gradually falls to nondiagnostic levels by the eighth week.
Unfortunately, the WFR has three serious limitations. First, infections with Proteus organisms are fairly common and will give rise to elevated titers (Davis et al. 1973, p. 903).
Second, antibodies may also be detected in serums of patients infected with either louseborne relapsing fever or leptospirosis. Diagnostic titers have been reported in both of these spirochetal diseases, and both occur in the known geographic distribution of scrub typhus (Carley et al. 1955). Zarafonetis, Ingraham, and Berry (1946) studied 50 patients with louseborne relapsing fever in Cairo, all of whom had positive blood smears for Borrelia recurrentis; all patients demonstrated an OX-K titer of 1:40, and 60 percent had titers of 1:160 or greater. Therefore, caution is necessary in interpreting titer rises to OX-K in areas where these diseases may occur simultaneously. While louseborne relapsing fever has not been reported from Vietnam, leptospirosis is endemic and accounted for approximately 7 percent of FUO cases seen there (Reiley and Russell 1969; Colwell et al. 1969).
Finally, the Weil-Felix reaction is diagnostic in only 40 to 60 percent of cases of scrub typhus (Sayen et al. 1946; Blake et al. 1945; Bozeman and Elisberg 1963). In the study of Blake and associates in New Guinea, seven patients had classical disease but failed to show significant titer rises; R. tsutsugamushi was recovered from the blood of two of them by inoculation of mice. The same observation was made in Vietnam, in an epidemic setting, by Hazlett (1970), who
149
noted four individuals with clinical disease who failed to develop diagnostic titers. These factors, plus the availability of more specific serologic techniques, render the WFR of historical importance only.
The complement fixation reaction, also used for the diagnosis of scrub typhus, is unsatisfactory because of the high degree of strain specificity of soluble antigens and the marked antigenic heterogenicity among various strains of R. tsutsugamushi (Bozeman and Elisberg 1963). The test is also used for typing various strains of the tsutsugamushi organisms, especially in epidemiological studies, and remains a useful tool although it will likely be replaced by direct immunofluorescence techniques as described by Iida, Kawashima, and Kawamura (1965).
Various other diagnostic immunologic techniques are available but are not frequently used. Serums may be titrated for ability either to neutralize rickettsial suspension toxicity in mice or to protect chick embryos or laboratory animals from infection. Neutralization in both of these tests reflects activity of antibodies to type-specific antigens. The protection test is troublesome to perform because of the difficulty in obtaining standard suspensions and is hazardous because accidental human infection may occur (Davis et al. 1973, p. 902).
Currently, the most effective and specific immunologic means of diagnosing scrub typhus involves an indirect immunofluorescence reaction. The test was originally described by Goldwasser and Shepard (1959) in their studies of epidemic and murine typhus and was adapted by Bozeman and Elisberg (1963) for scrub typhus. Antigen is prepared by smearing dilute suspensions of rickettsial organisms on glass slides. Each antigen smear is ringed with a Vaseline-ether-Sudan black solution to prevent mixing of the applied serums. Fourfold dilutions of serums, prepared in normal yolk sac diluent, are applied to the smears and incubated. Serums are removed by rinsing the slide, and to each slide is added diluted fluorescein isothiocyanate-labeled antihuman horse globulin. The slides are again incubated and rinsed, and the intensity of immunofluorescence is recorded.
In Bozeman and Elisberg`s study, antibody was regularly demonstrated in serums collected after the seventh day of illness; in a few, low levels of antibody were detected before the seventh day. Significant titers noted during the latter part of the second week increased to maximum levels between the third and fourth weeks. Only 6 of the 15 paired serums studied showed a significant rise in OX-K agglutinins, and no correlation was noted between the presence or absence of OX-K antibody and the fluorescent antibody titers. Nonspecific reactions did not occur when the scrub typhus serums were exposed to antigens of other rickettsial species, and no reaction occurred between R. tsutsugamushi smears and serums from other rickettsial infections or serums from patients with Proteus OX-K agglutinins of nonrickettsial origin.
The indirect immunofluorescence test of Bozeman and Elisberg is now used by laboratories throughout the world and is one of the most reliable serologic tests in use. However, the technique is difficult to apply when epidemiological studies or work in small animals is indicated because of the large
150
amounts of serum required. Gan, Cadigan, and Walker (1972) described a filter paper technique, requiring as little as 0.065 ml of blood, used in conjunction with the indirect immunofluorescence test. This technique is reliable and reproducible and works equally well with human, rodent, and primate specimens; it is particularly valuable since no special handling of the filter paper specimens is necessary.
Ideally, scrub typhus should be diagnosed by recovery of rickettsiae in mice inoculated with the blood of infected patients. Unfortunately, some strains of organisms cause disease in man but have only relative degrees of virulence in mice. For example, in the 15 cases of scrub typhus studied by Bozeman and Elisberg, 3 strains consistently killed all inoculated animals, 2 were mildly virulent, causing only occasional deaths, and 10 strains infected mice but did not kill them. In such instances, infection of test mice can be demonstrated only by their ability to survive challenge with a known pathogenic strain; this lengthy procedure requires 5 to 6 weeks to complete and is not applicable as a routine diagnostic study for scrub typhus.
TREATMENT
In 1945, the treatment of scrub typhus, as described by Sayen et al. (1946), consisted of "rest, nutrition, fluids and adequate nursing care." Although no effective specific therapy was available during World War II, two interesting treatment programs were evaluated. First, administration of immune blood and plasma was attempted. Blood was drawn from persons convalescing from the disease, usually during the fourth week and at a time when the Weil-Felix titer was maximal; six patients were treated with it, receiving from 500 to 2,000 cc of either whole blood or plasma. The authors found no evidence that the course of disease was alleviated by immunotherapy and did not believe that further trials of immune blood would be worthwhile.
Then, in the summer of 1944, penicillin became available in the China-Burma theater and was evaluated in the treatment of scrub typhus. Thirty-six patients were treated, receiving from one-half million to 6.5 million units (an average of 1.5 million units). Unfortunately, despite initial optimism, it became clear that "penicillin did not abort the disease, make its course milder, reduce the incidence of signs of major organ damage in severe cases, nor prevent patients from dying" (Sayen et al. 1946, p. 210). No other antibiotics were available for trial during this campaign. The disease remained a serious problem in Southeast Asia, with a significant mortality rate, prolonged clinical course with fever lasting 10 to 28 days, and protracted convalescence of 3 to 4 months before return to duty.
Within 3 years, this once dreaded infection was reduced to a nonfatal illness in which fever could be terminated within 30 hours. This dramatic reversal of the course of the disease resulted from the availability of Chloromycetin (chloramphenicol) and the work of Smadel and associates at the U.S. Army Institute for Medical Research in Kuala Lumpur, Malaya, and the Army Medical Department Research and Graduate School in Washington, D.C. The antibiotic
151
was first shown to have chemotherapeutic activity in a variety of experimental rickettsial and viral infections in 1946. Smadel and associates then undertook clinical trials of the drug in Malaya. In the initial report (Smadel et al. 1948), 25 treated and 12 control patients were compared. In a second study (Smadel, Woodward et al. 1949), results from a series of 30 treated and 19 control patients were reported. The individuals in these groups all had naturally acquired infection. The researchers also studied 37 volunteers who were exposed to scrub typhus in hyperendemic areas near Kuala Lumpur. These studies form the basis for the original chemotherapy of scrub typhus, and the observations of the authors have been reconfirmed by results from treatment of the disease in Vietnam.
The first important observation was that patients with naturally acquired scrub typhus who received chloramphenicol had a rapid resolution of fever, within 6 to 96 hours, and usually by about 30 hours, following institution of therapy. Total length of the febrile phase of the disease was shortened from 17 to 7 days, and the time between onset of symptoms and discharge from the hospital was shortened from 30 to 18 days. No relapses of the disease followed cessation of therapy in this group (Smadel, Woodward et al. 1949).
The experimental study confirmed the efficacy of the drug but raised some pragmatic problems and led to interesting and important immunologic conclusions. Of the 75 volunteers who were exposed, 37 developed clinical disease. Of these, 22 received no chemoprophylaxis and developed symptoms within 2 to 3 weeks of exposure. The remaining 15 received chloramphenicol prophylactically during exposure and for 2 succeeding weeks; they remained well until 2 weeks after the drug was discontinued but then developed clinical disease. The overall attack rate did not vary significantly between the prophylactically treated group and the control group (Smadel, Woodward et al. 1949).
An unusual and unexpected manifestation occurred among the volunteers who developed disease: 20 had relapses after they had apparently been cured by the initial course of the drug. This was especially disconcerting because there had been no relapses in the treated group with naturally occurring disease, nor had relapse been a part of the previously observed natural history of scrub typhus in man. An analysis revealed that the infecting dose in the volunteer group was probably heavier and was repeated over a several-day period, and that treatment in these individuals was invariably begun earlier than in those who acquired disease naturally. The ultimate conclusion was that Chloromycetin is rickettsiostatic rather than rickettsicidal (Smadel, Traub et al. 1949). When treatment was begun early in the course of disease, after a temporary interruption of growth of rickettsiae, the rickettsiostatic effect was dissipated quickly following discontinuance of the drug. At this early stage of infection, the immune mechanism of the host had not had time to respond adequately to the organisms. The patient`s clinical and serologic response appears to depend on the mass of antigen which the pathogen releases in the body (Smadel 1954). Specific therapy may interfere with development of the usual mass of antigen and thus delay or diminish antibody production.
This observation was confirmed by the Vietnam experience. In Hazlett`s
152
series (1970), all five patients who received treatment within 3 days of onset of symptoms had a relapse; among 24 treated 4 days or longer after onset of symptoms, there were no relapses. On the other hand, withholding therapy is unreasonable once the patient presents with symptoms. Furthermore, even with relapse, response to a repeat course of therapy is excellent. Smadel, Bailey, and Diercks (1950) suggested that relapses could be prevented by administering a 3-g dose of the antibiotic on the sixth to eighth day after termination of the original course of therapy. On the basis of these observations and data concerning total dosage, Hazlett recommended one of two possible regimens. If the patient is seen 4 or more days after onset of symptoms, he is given an initial dose of 3 g, followed by 0.5 g every 6 hours, to a total dose of 15-16 g. In treatment begun less than 4 days after onset of symptoms, an initial dose of 3 g is given, followed by 0.5 g to a total dose of 12 g, and an additional dose of 3 to 4 g is given 4 days after completion of the initial therapy for a grand total of 15 to 16 g. Because of the known serious hematologic toxicity of chloramphenicol, tetracycline, which is also effective (Smadel 1951), is the drug of choice for the treatment of scrub typhus at the present time. Recent data from Sheehy, Hazlett, and Turk (1973) suggest that tetracycline actually eliminates fever and other symptoms more rapidly than does chloramphenicol.
PREVENTION
Measures to prevent scrub typhus in endemic areas can be directed against the vector or toward active or passive protection of the human host from clinical infection (Smadel et al. 1952; Philip and Kohls 1945; Smadel et al. 1950). Control of the vector can be accomplished both by treatment of the terrain and by individual use of insecticides. Terrain control may be applicable to some military situations, but it is less satisfactory during peacetime because of cost and the need for treatment of large areas. Although controlled studies are not available, miticidal chemicals such as dimethyl phthalate, dibutyl phthalate, and benzyl benzoate clearly provide effective individual protection. However, M-1960 repellent, despite its availability in Vietnam (Hazlett 1970, p. 33), was used only infrequently by troops in the field because of the belief that the odor of the repellent could be detected by enemy forces.
Because chloramphenicol was extremely effective against clinical disease, attention was quickly directed to its possible use as chemoprophylaxis in individuals exposed to scrub typhus under natural conditions. Field studies by Smadel, Traub et al. (1949) clearly demonstrated that chloramphenicol given during and for 2 weeks after exposure prevented development of clinical disease only throughout the period of prophylaxis and the 5 ensuing days.
Attempts to develop a vaccine against scrub typhus began during, and continued after, World War II. Because of the post infection immunity which apparently developed in man and was confirmed in animal studies, there was every reason to believe that such a vaccine could be developed, as had been done for
153
epidemic typhus. Several groups attempted to produce a noninfectious material and were successful in producing resistance against homologous strains of R. tsutsugamushi in laboratory animals. Preventive immunization against the disease in man was found to be extremely difficult, however, because of variations in the virulence of the organisms and the antigenic structure of various strains (Kekcheyeva 1968). Three field studies using killed scrub typhus vaccines failed to give encouraging results, and this approach has been abandoned (Smadel 1952, p. 644).
Subsequent studies evaluated the effectiveness of combining live vaccine and chemoprophylaxis to suppress clinical disease without interfering with the development of immunity. In their field trials, using both Gilliam and Karp strains, Smadel et al. (1950; 1952) noted that the level of immunity in persons treated in this manner was similar to that in patients following clinical infection. Immunity to the homologous strain was found to be complete and to persist for several years. On the other hand, resistance to heterologous infection was transient, lasting only a few months or less. Therefore, while the combined administration of live vaccine and chemoprophylaxis to produce homologous immunity is effective, it is of limited practical applicability.
Russian animal studies (Kekcheyeva 1968) suggest that live preventive vaccines made from virulent rickettsial cultures treated with tetracycline (antibio vaccine) can produce both homologous and heterologous immunity in mice. However, immunity has been demonstrated only in short term studies, and no data from humans have been reported. Thus, at the present time, no practical vaccination program for the prevention of human scrub typhus is available.
SUMMARY
Placed in proper perspective, scrub typhus or tsutsugamushi fever does not enjoy a place of prominence in the medical history of Vietnam; rather, it will be best remembered as one of the several causes of FUO which always pose a diagnostic problem for physicians in the field. Early diagnosis was important because appropriate therapy allowed for rapid abatement of bothersome symptoms. On the other hand, in contrast to the World War II experience, no mortality was associated with its occurrence, probably because of the availability of adequate treatment and the good health of American troops, and perhaps because the form of the disease seen in Vietnam was less virulent than that in other areas of Asia.
No new therapeutic approaches were developed, although the efficacy of tetracycline as the drug of choice was confirmed. Previous epidemiological observations were extended to include Vietnam, but no new data were collected.
The refusal of combat troops to use the available repellents and the lack of an effective vaccine combined to assure that the exotic tsutsugamushi disease would provide an anamnestic stimulation to the intellectual titer of American medicine for another generation of physicians.
154
Section II. Murine Typhus
Colonel O Neill Barrett, Jr., MC, USA (Ret.)
Although scrub typhus was recognized early during the American involvement in Vietnam, murine typhus was not recognized until July 1967 and, even then, was diagnosed only through an FUO evaluation using Weil-Felix OX-19 titers. It was never a serious clinical or epidemiological problem in Vietnam.
HISTORY AND MILITARY SIGNIFICANCE
In the mid-1920`s, Maxey (1926) suggested the occurrence of a disease similar to typhus fever but with a low mortality rate, milder clinical picture, and probably a different epidemiological pattern; this hypothesis was confirmed by the work of Dyer, Rumreich, and Badger (1931). By 1934, the differences among epidemic typhus, Brill`s disease, and murine typhus were clearly defined (Zinsser 1934).
Statistics on U.S. Army experience with murine typhus before 1940 are not available because until that time all forms of typhus were recorded in a single figure. During World War II, 787 cases of murine typhus were reported in American troops; of these, 497 were acquired in the continental United States, primarily in the Southeast. The great majority of the 290 cases recorded overseas were in the Central and South Pacific Areas; only 34 cases were officially noted from the China-Burma-India theater, although other data suggest that more cases occurred there. A total of 15 deaths were attributed to murine typhus-a fatality rate of 19 per 1,000 cases. Fourteen of the fatal cases were acquired overseas; this comparatively high mortality might be attributed either to greater virulence of the disease in oversea areas or to the presence of associated diseases and the decreased resistance of the troops. Only 104 cases of epidemic typhus occurred in American troops during World War II, and no deaths were noted. The remarkably low incidence of this disease, which reached epidemic proportions in the surrounding civilian population, attests to extremely effective preventive measures (MD-PM7, pp. 178-79, 268-70). No cases of murine typhus were reported in American troops in Korea from 1950 to 1953.
The recognition of murine typhus in American troops in Vietnam was almost accidental. Statistical records from the Office of the Surgeon General on the incidence of rickettsial diseases in Vietnam show only 19 clinical cases of murine typhus from 1965 to 1970: 1 case was recorded in 1965, none in 1966 or 1967, 16 in 1968, 23 in 1969, and 10 in 1970 (PAD). Serologic and epidemiologic data, however, indicated that the occurrence of the disease was much more frequent than was recognized clinically. From July 1967 through 1968, 61 cases of murine typhus were serologically diagnosed, based on a fourfold rise in specific antibody titer, at the 9th Medical Laboratory (Baker, McKinney, and Huxoll). (In comparison, during the same period 190 cases of scrub typhus were
155
similarly diagnosed [Baker, McKinney, and Huxoll].) In the first reported clinical study in Vietnam, Deaton (1969) found 25 cases of murine typhus among 306 patients who presented with FUO at the U.S. Air Force Hospital, Cam Ranh Bay, from July 1967 to June 1968. Only the Weil-Felix reaction was available for the serologic evaluation in this study. As in scrub typhus, this test is less sensitive than other available serologic tests and probably underestimates the actual incidence of disease. Miller and associates (1974) described 58 cases seen at the same installation in the following year. In their study, both the Weil-Felix titers and fluorescent antibody techniques for specific antigens were used.
EPIDEMIOLOGY
Murine or urban typhus is a natural infection of rats and other rodents in many parts of the world. In the United States, this disease occurred initially in port cities and urban areas of the Atlantic seaboard and in the southeastern states, especially Texas, Georgia, Alabama, and Florida (MD-PM7, p. 269). It has since spread to rural areas, and for more than 20 years cases have been reported from California with increasing frequency (Adams, Emmons, and Brooks 1970).???
Classically, the disease is transmitted by the oriental rat flea, Xenopsylla cheopis; the domestic rat, Rattus norvegicus, is the typical mammalian reservoir (fig. 45). However, a wide variety of naturally or experimentally infected animals and ectoparasites has been recorded, suggesting a more widespread distribution. Recent data indicate that the mammalian reservoir affecting man might also include the domestic cat and the opossum. Other species of fleas which readily bite man and domestic animals, especially the cat flea, Ctenocephalides felis, may transmit the infection from animal to animal and to man (Adams, Emmons, and Brooks 1970; Older 1970).
Murine typhus has been recognized in Southeast Asia since Lewthwaite and Savoor (1936) isolated Rickettsia typhi [mooseri] from rats and people in Malaya (now West Malaysia) and showed it to be distinct from other rickettsiae. Most of the animals harboring murine typhus were those closely associated with man in urban or other man-dominated habitats. The most common mammalian reservoir was Rattus rattus diardi, the house rat. Distribution of murine typhus in and around West Malaysia`s urban centers is similar to that in other endemic areas and, while the cycle depends upon house rats in the cities and towns, the disease has been found in Rattus rattus in nonurban areas, indicating that rural foci may also exist (Marchette 1966). Similar experience has been reported from the Philippines (Woodward, Philip, and Loranger 1946).
In Thailand, Rickettsia typhi was first isolated in 1964 from rodents in the northern province of Chiang Rai. Since then, extensive surveys have indicated that murine typhus is widely endemic throughout the country, especially in the north and northeast regions. Rattus exulans, the Thai domestic rat, appears to maintain the enzootic cycle. The clinical significance of murine typhus
156
in Thailand is unknown and to date the disease has been demonstrated there only twice in man (Sankasuwan et al. 1969).
Because the existence of murine typhus in Southeast Asia had been previously established, its discovery as a cause of FUO in American troops in Vietnam was neither surprising nor particularly disconcerting. The casual response to the disease was justified by its benign course and low morbidity. Data concerning the epidemiology of murine typhus in Vietnam are virtually nonexistent. Baker, McKinney, and Huxoll studied the distribution of rickettsial disease there by location and type of military activity. They found that 93 percent of the cases of scrub typhus occurred among combat troops, usually in infantry and artillery units; in contrast, only 45 percent of the cases of murine typhus occurred in this group, while 55 percent involved support troops, advisers, engineers, and U.S. contract civilians. This finding reconfirmed the urban distribution of the disease. Most cases of scrub typhus (60 percent) were noted in the Eastern II CTZ area, whereas most murine typhus cases (54 percent) occurred in the Western II CTZ area. No information regarding either the vector or the mammalian host is available.
CLINICAL MANIFESTATIONS
Although extensive clinical experience with murine typhus was not recorded during World War II, sufficient data were available to provide a summary of the clinical features of the disease, which mimicked epidemic typhus
157
TABLE 26.- Comparison of clinical features of murine typhus from three studies in Texas and Vietnam
but tended to be less severe. Typically, the incubation period was from 8 to 16 days. Prodromal symptoms of nausea and malaise were common, followed by chills, headache, fever, and generalized myalgia. The total febrile course averaged 12 days with temperatures of 103? to 104? F. Rash occurred in 80 percent of the cases, typically appearing on the fifth day of illness. Lesions were first noted over the anterior chest wall, abdomen, lower back, and buttocks, generally sparing the face, scalp, palms, and soles; initially they were macular, becoming papular and confluent. As in epidemic typhus, a dry hacking cough usually developed, suggesting possible interstitial pneumonia. No specific laboratory changes were observed except for transient albuminuria (MD-IM2, pp. 221-23).
Three recent clinical descriptions of the disease confirm this picture. Older (1970) reported an outbreak of 28 cases in Texas in 1969; Deaton (1969) reported the first extensive experience from Vietnam - a total of 25 cases seen in 1 year at Cam Ranh Bay; and Miller et al. (1974) described 58 cases in the only other clinical study from Vietnam (table 26).
In the Texas series, symptoms were typical. The initial diagnosis in 22 of the 28 cases was influenza. The rash was generally mild and localized to the trunk and was observed in only a third of the cases. Rash was more common in Deaton`s series, noted in 68 percent of the cases. In contrast to the experience in Texas and in the World War II study, splenomegaly was not uncommon in Vietnam; Deaton observed it in 56 percent of cases and Miller in 43 percent. Lymphadenopathy was a significant physical finding only in the studies from Cam Ranh Bay.
No consistent laboratory abnormalities have been defined in murine typhus. White blood cell counts have varied from low (4,600/mm3) to high (17,700/mm3); anemia, when noted, was mild. SGOT (serum glutamic-oxaloacetic transaminase) levels were abnormal in 11 of 20 measurements in Miller`s series, but alkaline phosphatase, bilirubin, and BSP (Bromsulphalein) values were normal.
158
The diagnosis of murine typhus is made by standard serologic methods. The traditional Weil-Felix reaction using OX-19 antigen is useful but lacks sensitivity; of greater value are the complement fixation and hemagglutination or hemagglutination-inhibition tests (Wisseman et al. 1962; Zdrodovskii and Golinevich 1960, pp. 266-68). A toxin neutralization test, with high sensitivity and specificity, is available and useful in the diagnosis of epidemic typhus. Unfortunately in murine typhus, the toxin-neutralizing substance is nonspecific and can be found in a significant number (35 percent) of persons who reside in typhus-free areas and have not received typhus vaccine. Therefore, the test is not reliable as an indication of past infection with murine typhus or as a means of differentiating between murine and epidemic typhus fever (Bell et al. 1969). The fluorescent antibody technique is now the standard test for diagnosing murine typhus (Miller et al. 1974).
TREATMENT AND CHEMOPROPHYLACTIC MEASURES
Like other rickettsial diseases, murine typhus can be treated very effectively. While either chloramphenicol or tetracycline may be used, tetracycline is preferred because it does not pose the threat of aplastic anema as does chlor amphenicol (Older 1970; Miller et al. 1974; Wisseman et al. 1962). Both drugs are rickettsiostatic rather than rickettsicidal. The objectives of therapy are therefore to halt the growth of organisms as rapidly as possible and to prevent further proliferation until immune mechanisms are fully developed. When patients with either scrub or murine typhus are treated 7 or more days after onset of the disease, relapse is rare; when treatment is begun early, especially with a short course of chemotherapy, it is common. In murine typhus, the recurrence may be even more severe than the original infection. Furthermore, in murine typhus the interval between withdrawal of drug and return of clinical manifestations is only 2 to 3 days in contrast to 5 to 7 days in scrub typhus. These observations have led to the recommendation by Wisseman and others (1962) that the antibiotic be given until the patient is afebrile for 48 hours and then continued until 12 to 14 days after the onset of disease. Whether or not such a program is actually necessary is not clear, since neither Older (1970) nor Deaton (1969) comments on relapse following treatment of naturally occurring disease.
Although a vaccine against murine typhus was available during World War II, it was not used, in contrast to the vaccine for epidemic typhus, which was widely used and apparently effective (Gear 1969; MD-PM7, pp.188-89, 267). There was no program for vaccination of troops against murine typhus during the Vietnam involvement, and commercial vaccines for the protection of laboratory workers were not available. However, the attenuated E strain of Rickettsia prowazekii, when used as a living vaccine, has been shown to provide cross-protection against the disease (Wisseman et al. 1962, p.751). Prevention of the disease under naturally occurring conditions is best accomplished by control of the vector and the mammalian host.
159
Section III. Leptospirosis
Colonel Fred R. Stark, MC, USA
INCIDENCEAND EPIDEMIOLOGY
The conditions under which American ground forces operated in Vietnam predisposed them to intense local outbreaks of leptospirosis with considerable morbidity among field forces during the dry seasons. The absolute incidence of leptospirosis was unknown, but available data suggest that in certain regions for brief periods, up to 10 percent of combat troops who were inactive because of febrile illness suffered from leptospirosis (Berman, Irving, and Kundin 1968). The overall incidence of the disease, estimated from surveys of serums on medically evacuated wounded soldiers during the 1966-67 period, was approximately 1 percent.* In contrast, the rate of detection in Vietnam was only about one person per 1,000, based on confirmed cases reported from the 9th Medical Laboratory.**
Recognizing that about 1 in 10 febrile soldiers seen in the field at the battalion level reached a hospital, that only 20 percent or less of all soldiers in country were in the field at any one time,*** and that leptospirosis in other out breaks is often a mild self-limited disease (Heath, Alexander, and Galton 1965), a very high annual attack rate may be estimated, perhaps exceeding 100 per 1,000 combat soldiers per year. This is consistent with the experience of other forces operating in Panama (Mackenzie et al. 1966) and Malaysia (McCrumb et al. 1956). Although specific epidemiological research was not undertaken at the battalion level in Vietnam, reports from the 9th Medical Laboratory in 1969 (ML9-AR) indicate the larger number of serums positive for leptospirosis from such units.
There is no reason to believe that the pattern of leptospirosis transmission was unique in the Republic of Vietnam. A history of exposure to fresh water, often involving immersion, was usually elicited from patients there, as else where. Excellent data accumulated under the direction of Lt. Col. (later Col.) Hinton Baker, MC, commander of the 9th Medical Laboratory from 1967 to 1969, revealed that the disease was widespread, seasonal (commencing with the end of the monsoon and peaking early in the dry season), and far more common in field soldiers than in support troops, and that it demonstrated a varied and nonspecific clinical picture.
______
*Personal observation by author, 250 random serums, Walter Reed General Hospital, 1966-67. Three serums possessed reciprocal hemolytic titers greater than 640 against Leptospira biflexa.
** Col. Hinton Baker: Personal communication. This figure inadequately reflects total incidence. In 1969, 169 cases of leptospirosis were serologically confirmed from among 3,300 serums (a 5.1-percent positivity rate) obtained from an estimated 22,000 FUO admissions (ML9-AR, pp. 14-17; Med Stat).
***Estimated by Lt. Col. Andre J. Ognibene, USARV Medical Consultant, 1969.
160
CLINICAL FEATURES
Generally leptospirosis presented as an FUO. As noted below, high fever, usually of 6 to 9 days` duration, persisting with or without antibiotic treatment was the rule. Headache, nausea, vomiting, and backache were common. The clinical features for 19 patients in one study (Allen, Weber, and Russell 1968) were as follows:
Fever - 17
Headache - 12
Chills - 11
Backache - 9
Nausea or vomiting - 2
Lethargy or malaise - 2
Abdominal complaint - 2
Dizziness - 2
Physical findings of conjunctival suffusion and liver tenderness were occasionally present in this group. Rash was uncommon. Mild degrees of liver and renal dysfunction were noted. Following are the physical findings in these patients:
Conjuctival suffusion - 8
Hepatic tenerness - 9
Splenic tenderness - 6
Adenopathy - 4
Muscle tenderness - 3
Periumbilical tenderness - 2
Petechiae ( palate:1; feet:1) - 2
Pharyngitis - 1
Palable spleen - 1
Flush of face - 1
No abnormal findings - 1
Infrequently, more severe renal involvement requiring dialysis was documented. In many cases, high output renal failure occurred. High ambient temperatures contributed to the rapid development of hypovolemia and in some cases may have played a role in the production of oliguric rather than nonoliguric renal failure. In contrast to classical descriptions of the leptospiral disease syndrome, Weil`s disease, jaundice, which is the most striking manifestation, was not characteristic of the leptospirosis seen in Southeast Asia.
In most cases seen by the author, mild proteinuria and many casts with infrequent red cells were present, but these findings were usually not striking compared to those in other patients with high fever. Marked hematuria was in frequent, few cases demonstrated cerebrospinal fluid abnormalities, and hemorrhagic complications were uncommon.
Neither autopsy studies nor antemortem tissue examinations are available from Vietnam. When fatal cases have been examined, myositis, hepatic inflammation or necrosis, marked inflammation of the kidneys, pneumonia, myocarditis, diffuse gastrointestinal hemorrhage, meningitis, transient transverse myelitis, and marked conjunctival suffusion have all been described; these findings are discussed in an excellent review of clinical experience with leptospirosis in the United States (Heath, Alexander, and Galton 1965, pp. 917-19).
The majority of leptospirosis patients in Vietnam had normal or modest elevations of the white blood count, moderate disturbances of liver function, with elevations of the serum glutamic-oxaloacetic transaminase and lactic de hydrogenase, and mild elevations of the alkaline phosphatase and bilirubin. BUN (blood urea nitrogen) elevations to 50 mg percent were common and were
161
associated with an element of prerenal azotemia, as described in studies on the Indian subcontinent (Sitprija 1968).
LABORATORYDIAGNOSIS
The bulk of cases were confirmed by serologic tests. Generally, Leptospira biflexa antigen obtained from Dr. Aaron Alexander, Walter Reed Army Institute of Research, was used to coat red cells for a nonspecific hemolytic assay. Between 1967 and 1969, fourfold titer rises in paired acute and convalescent serums, or single reciprocal titers of 1/400, with an appropriate history were taken as confirmation of leptospirosis. Few attempts at systematic isolation of the organism were made; some data comparing specific agglutination titers with the hemolytic assay were obtained from the I CTZ area laboratories (Berman, Irving, and Kundin 1968). These data from 297 cases suggested the presence of multiple species of leptospires, even in a single outbreak (Lipton and Legters), as follows (listed by serotype and number of cases):
austalis - 66
bataviae - 68
ictrohaemorrahgiae - 47
canicola - 43
grippotyphosa - 35
hebdomidis - 38
Multiple infections, simultaneously or serially, have been described in animals and man, from a single exposure source. Multiple species have been isolated from single samples of fresh water (Heath, Alexander, and Galton 1965).
PREVENTION AND TREATMENT
Few studies of the prevention of leptospirosis were undertaken in Vietnam by U.S. personnel, probably because early data on morbidity did not reflect the countrywide impact of the disease. Studies of polyvalent vaccines in animals have been impressive (Hanson, Tripathy, and Killinger 1972), and some preliminary data in man are available (Benoist and Lataste-Dorolle 1970). Measures such as prophylactic antibiotics, topical protection against immersion in fresh water, and vaccines were not evaluated during the Vietnam campaign. Studies of drug treatment of leptospirosis were not undertaken. Most patients treated received tetracycline because of the difficulty of differentiating the disease from scrub or murine typhus. Antibiotics did not appear to shorten the course of the disease, but the absence of mortality may have been related to the widespread policy of prompt administration of tetracycline at the battalion and hospital levels to severely ill febrile patients in whom leptospirosis or scrub typhus could not be excluded. In addition, the prompt use of parenteral fluids in most febrile patients may have prevented the oliguric renal failure known to be associated with the disease.
162
CONCLUSION
Significant gaps in the knowledge of leptospirosis persist. The relatively benign course in most cases in Vietnam blunted research efforts. In future operations in wet, tropical areas, this disease must be evaluated to determine its importance as a cause of combat man-days lost.
REFERENCES
Activities of medical consultants, Internal Medicine in World War II. See MD-IM1.
Adams, W. H.; Emmons, R. W.; and Brooks, J.E.1970. The changing ecology of murine (endemic) typhus in southern California. Am. J. Trop. Med. 19: 311-18.
Allen, A. C., and Spitz, S. 1945. Acomparative study of the pathology of scrub typhus (tsutsugamushi disease) and other rickettsial diseases. Am. J. Path. 21: 603-82.
Allen, G. L.; Weber, D. R.; and Russell, P.K. 1968. The clinical picture of leptospirosis in American soldiers in Vietnam. Mil. Med. 133: 275-80.
Audy, J. R. 1968. Red mites and typhus. London: Athlone Press.
Baker, Col. H.; McKinney, Lt. Col. R.; and Huxoll, Maj. D. Rickettsial diseases indicated by sero tests at the 9th Medical Laboratory, Vietnam, 1967-68. Report, undated.
Bell, E. J.; Lackman, D. B.; Ormsbee, R. A.;and Peacock, M. 1969. Neutralization of murine typhus toxin by serum of normal human beings and monkeys. Am. J. Trop. Med 18:559-67.
Benoist, F., and Lataste-Dorolle, C. 1970.Preliminary research on the vaccination of sewage workers against occupational leptospirosis. Ann. Med. Intern.(Paris) 121: 489-96.
Berman, S. J.; Irving, G.; and Kundin, W. D.1968. Infectious disease survey of U.S. personnel in I Corps, SouthVietnam, U.S. Naval Medical Research Unit No. 2, Taipei, Taiwan, Mar.68.
Blake, F. G.; Maxey, K. F.; Sadusk, J.F., Jr.; Kohls, G. M.; and Bell, E. J. 1945. Studies on tsutsugamushi disease (scrub typhus, mite-borne typhus) innew Guinea and adjacent islands: Epidemiology, clinical observations and etiology in the Dobadura area. Am. J. Hyg. 41: 243-373.
Bozeman, F. M., and Elisberg, B. L. 1963.Serological diagnosis of scrub typhus by indirect immunofluorescence. Proc. Soc. Exper. Biol. & Med. 112:568-73.
Cadigan, F. C., Jr.; Andre, R. G.; Bolton, M.; Gan, E.; and Walker, J. S. 1972. The effect of habitat on the prevalence of human scrub typhus in Malaysia. Tr. Roy. Soc. Trop. Med. & Hyg. 66: 582-87.
Carley, J. G. et al. 1955. The investigation of fevers in North Queensland by mouse inoculation, with particular reference to scrub typhus. Australasian Ann. Med. 4: 91-99.
Chernof, D. 1967. Hypofibrinogenemia in scrub typhus. Report of a case. New England J. Med. 276: 1195-96.
Colwell, E. J.; Brown, J. D.; Russell, P. K.; Boone, S. C.; Legters, L. J.; and Catino, D. 1969. Investigations on acute febrile illness in American servicemen in the Mekong Delta of Vietnam. Mil .Med. 134:1409-14.
Communicable diseases. Arthropodborne diseases other than malaria. Preventive Medicine in World War II. SeeMD-PM7.
Cottingham, A. J., Jr.; Legters, L. J.; Boone, S. C.; Proctor, R. F.; and Lipton, H. L.; Some clinical and epidemiological observations on scrub typhus incidence among indigenous and U.S. forces during combat operations in II and III Corps Tactical Zone. In Annual Progress Report, U.S. Army Medical Research Team (WRAIR)Vietnam and Institute Pasteur of Vietnam, 1 Sept. 1966-31 Aug. 1967, pp. 238-73.
Deaton, J. G. 1969. Febrile illnesses in the Tropics (Vietnam). Mil. Med. 134: 1403-8.
Deller, J. J., Jr., and Russell, P. K.1967.An analysis of fevers of unknown origin in American soldiers in Vietnam. Ann. Int. Med. 66: 1129-43.
Dyer, R. E.; Rumreich, A.; and Badger, L. F.1931. Typhus fever: A virus of the typhus type derived from fleas collected from wild rats. Pub. Health Rep. 45: 334-38.
Faust, E. C., and Russell, P. F.1964. Craig and Faust`s clinical parasitology. 7th ed. Philadelphia: Lea & Febiger.
Gan, E.; Cadigan, F. C., Jr.; and Walker, J. S. 1972. Filter paper collection for blood for use in a screening and diagnostic test for scrub typhus using the IFAT. Tr. Roy. Soc. Trop. Med. & Hyg. 66: 588-93.
Gear, J. H. S. 1969. Rickettsial vaccines. Brit.M. Bull. 25: 171-76.
Goldwasser, R. A., and Shepard, C. C. 1959.Fluorescent antibody methods in the differentiation of murine and epidemic typhus sera. Specificity changes resulting from previous immunizations. J. Immunol, 82:373-80.
Hanson, L. E.; Tripathy, D. N.; and Killinger, A. H. 1972. Current status of leptospirosis immunization in swine and cattle. J. Am. Vet. M. A. 161: 1235-43.
Hazlett, D. R. 1970. Scrub typhus in Vietnam: Experience at the 8th Field Hospital. Mil. Med. 135: 31-34.
Heath, C. W., Jr.; Alexander, A. D.; and Galton, M. M. 1965. Leptospirosis in the United States. Analysis of 483 cases in man, 1949-1961. New England J. Med. 273: 857-64, 915-22.
Iida, T.; Kawashima, H.; and Kawamura, A.1965. Direct immunofluorescence for typing of tsutsugamushi disease rickettsia. J. Immunol, 95: 1129-33.
Individual Medical Records, Patient Administration Division. See PAD.
Infectious diseases. Internal Medicine in World War II. See MD-IM2.
Jackson, E. B.; Danauskas, J. X.; Smadel, J.E.; Fuller, H. S.; Coale, M. C.; and Bozeman, F. M. 1957. Occurrence of Rickettsia tsutsugamushi in Korean rodents and chiggers. Am. J. Hyg. 66: 309-20.
Kekcheyeva, N. 1968. Preventive immunization against tsutsugamushi fever. J. Hyg. Epidemiol. 12: 14-17.
Kundin, W. D.; Liu, C.; Harmon, P.; andRodina, P. 1964. Pathogenesis of scrub typhus infection (Rickettsiatsutsugamushi) as studied by immunofluorescence. J. Immunol, 93:772-81.
Le Gac, P., and Arqui, E. 1964. Les facteursendmiques du scrub-typhus Indochinois. Bull. soc. path. exot. 57:277-83.
Lewthwaite, R., and Savoor, S. R. 1936. Thetyphus group of diseases in Malaya. Part III. The study of the virus of the urban typhus in laboratory animals. Brit. J. Exper. Path. 17: 23-24.
Ley, H. L., and Markelz, R. A.1961. Scrubtyphus: Occurrence in United Nations personnel in Korea. Mil. Med 126:834-37.
Lipton, H. L., and Legters, L. J. Clinical and epidemiological notes on leptospirosis. In Annual Progress Report, U.S. Army Medical Research Team (WRAIR) Vietnam and Institute Pasteur of Vietnam, 1 Sept. 1966-31 Aug. 1967, pp.224-37.
Mackenzie, R. B.; Reiley, C. G.; Alexander, A. D.; Bruckner, E. A.; Diercks, F. H.; and Beye, H. K. 1966. An outbreak of leptospirosis among U.S. Army troops in the Canal Zone. I.Clinical and epidemiological observations. Am. J. Trop. Med 15: 57-63.
Marchette, N. J. 1966. Rickettsioses (tick fever, Q-fever, urban typhus) in Malaya. J. Med. Ent. 2:339-71.
Maxcy, K. F. 1926. An epidemiological study of endemic typhus (Brill`s disease) in the southeastern United States with special reference to its mode of transmission. Pub. Health Rep. 41: 2967-95.
McCrumb, F. R., et al. 1956. Leptospirosis in Malaya. Walter Reed Army Institute of Research Publication 199: 56.
MD-IMl-Medical Department, United States Army. 1961. Activities of medical consultants. Internal Medicine in World War II, vol. I. Washington: Government Printing Office.
MD-IM2-Medical Department, United States Army. 1963. Infectious diseases. Internal Medicine in World War II, vol. II. Washington: Government Printing Office.
MD-PM7-Medical Department, United States Army. 1964. Communicable diseases. Arthropodborne diseases other than malaria. Preventive Medicine in World War II, vol. VII. Washington: Government Printing Office.
Medical Statistics Agency, Office of the Surgeon General. See Med Stat.
Med Stat-Medical Statistics Agency, Office of the Surgeon General, Department of the Army. Health of the Command, 1965-1970. Report, undated.
Miller, M. B.; Bratton, J. L.; Hunt, J.; Blankenship, R.; Lohr, D. C.; and Reynolds, R. D. 1974. Murine typhus in Vietnam. Mil. Med. 139: 184-86.
ML9-Aft-9th Medical Laboratory. Activities Report, 1969. On file at U.S. Army Center of Military History.
Moulder, J. W. (rev.) 1973. In Microbiology, including immunology and molecular genetics, ed. B. D. Davis, R. Dulbecco, H. N. Eisen, H. W. Ginsberg, and W. B. Wood, Jr., pp. 898-913.2d ed. New York: Harper & Row.
Munro-Faure, A. D.; Andrew, R.; Missen, G. A.K.; and Mackay-Dick, J. 1951. Scrub typhus in Korea. J. Roy. Army M. Corps 97: 227-29.
NAVY-MN-Communicable Disease Control. Outbreak of scrub typhus. Navy M. Newsletter 23: 33-35,1954.
Navy M. Newsletter. See NAVY-MN.
9th Medical Laboratory. See ML9-AR.
Ognibene, A. J.; O`Leary, D. S.; Czarnecki,S. W.; Flannery, E. P.; and Grove, R. B. 1971. Myocarditis and disseminated intravascular coagulation in scrub typhus. Am. J. M.Sc. 262: 233-39.
Older, J. J. 1970. The epidemiology of murine typhus in Texas, 1969. J.A.M.A. 214: 2011-17.
PAD-Patient Administration Division, Health Services Command, Department of the Army. Individual Medical Records (IMR),1965-70.
Parsons, R. E.; McLaurin, B. F.; Do Van Quy; and Legters, L. J. Preliminary observations on scrub typhus ecology in II Corps Tactical Zone. In Annual Progress Report, U.S. Army Medical Research Team (WRAIR) Vietnam and Institute Pasteur of Vietnam, 1 Sept. 1966-31 Aug. 1967, pp. 274-99.
Patient Administration Division, HealthServices Command. See PAD.
Philip, C. B.1966. Scrub typhus. In Amanual of tropical medicine, ed. G. W. Hunter, W. W. Frye, and J.C. Swartzwelder, pp. 105-11. 4th ed. Philadelphia: W. B. Saunders Co.
Philip, C. B., and Kohls, G. M. 1945. Studies on tsutsugamushi disease (scrub typhus, mite-borne typhus) in New Guinea and adjacent islands. Tsutsugamushi disease with high endemicity on a small South Sea island. Am. J. Hyg. 42: 195-203.
Reiley, C. G., and Barrett, 0. 1971. Leukocyte response in acute malaria. Am. J. M. Sc. 262:153-58.
Reiley, C. G., and Russell, P. K. 1969.Observations on fevers of unknown origin in the Republic of Vietnam. Mil. Med 134: 36-42.
Sankasuwan, V.; Pongpradit, P.; Bodhidatta,P.; Thonglongya, K.; and Winter, P. E. 1969. Murine typhus in Thailand, Tr. Roy. Soc. Trop. Med & Hyg. 63: 639-43.
Sayen, J. J.; Pond, H. S.; Forrester, J. S.;and Wood, F. C.1946. Scrub typhus in Assam and Burma: A clinical study of 616 cases. Medicine 25: 155-214.
Sheehy, T. W.; Hazlett, D.; and Turk, R. E.1973. Scrub typhus. A comparison of chloramphenicol and tetracycline in its treatment. Arch. Int. Med. 132: 77-80.
Sitprija, V. 1968. Renal involvement in human leptospirosis. Brit. M. J. 2: 656-58.
Smadel, J. E. 1951. Present status of antibiotic therapy in viral and rickettsial disease. Bull. New York Acad. Med.27: 221-31.
Smadel, J. E. 1952. Scrub typhus. In Viral and rickettsial infections of man, ed. T.M. Rivers. 2ded.Philadelphia: J.B. Lippincott Co.
Smadel, J. E. 1954. Influence of antibiotics on immunologic responses in scrub typhus. Am. J. Med. 17:246-58.
Smadel, J. E.; Bailey, C. A.; and Diercks, F.H. 1950. Chloramphenicol (Chloromycetin) in the chemoprophylaxis of scrub typhus (tsutsugamushi disease). IV. Relapses of scrub typhus intreated volunteers and their prevention. Am. J.Hyg. 51: 229-41.
Smadel, J. E.; Ley, H. L., Jr.; Diercks, F.H.; Paterson, P. Y.; Wisseman, C. L., Jr.; and Traub R. 1952.Immunization against scrub typhus: Duration of immunity in volunteers following combined living vaccine and chemoprophylaxis. Am. J. Trop. Med. 1: 87-99.
Smadel, J. E.; Traub, R.; Frick, L. P.; Diercks, F. H.; and Bailey, C. A. 1950. Chloramphenicol (Chloromycetin)in the chemoprophylaxis of scrub typhus (tsutsugamushi disease). III. Suppression of overt disease by prophylactic regimens of four-week duration. Am. J. Hyg. 51: 216-28.
Smadel, J. E.; Traub, R.; Ley, H. L., Jr.; Philip, C. B.; Woodward, T. E.; and Lewthwaite, R. 1949.Chloramphenicol(Chloromycetin) in the chemoprophylaxis of scrub typhus (tsutsugamushi disease). II. Results with volunteers exposed in hyperendemic areas of scrub typhus. Am. J. Hyg. 50:75-91.
Smadel, J. E.; Woodward, T. E.; Ley, H. L.,Jr.; and Lewthwaite, R. 1949. Chloramphenicol (Chloromycetin) in the treatment of tsutsugamushi disease (scrub typhus). J. Clin. Invest.28: 1196-1215.
Smadel, J. E.; Woodward, T. E.; Ley, H. L., Jr.; Philip, C. B.; Traub, R.; Lewthwaite, R.; and Savoor, S. R. 1948. Chloromycetin in the treatment of scrub typhus. Science 108: 160-61.
Stone, J. H., ed. 1969. Crisis fleeting, Original reports on military medicine in the Second World War. Washington: Government Printing Office.
Traub, R. 1954. Advances in our knowledge of military medical importance of mites and fleas due to postwar experiences in the Pacific area. In Recent advances med &surg. Medical Serial Publication No. 4. Washington: Army Medical Service Graduate School.
Traub, R. 1960. Malaysian parasites. Studies of Institute for Medical Research, Malaya 29: 198.
Traub, R.; Frick, L.P.; and Diercks, F. H. 1950. Observations on the occurrence of Rickettsiatsutsugamushi in rats and mites in the Malayan jungle. Am. J. Hyg. 51: 269-73.
Traub, R.; Hertig, M.; Lawrence, W. H.; and Harriss, T. T. 1954. Potential vectors and reservoirs of hemorrhagic fever in Korea. Am. J. Hyg. 59: 291-305.
Trishnanda, M.; Vasuvat, C.; and Harinasuta, C. 1964. Investigation of scrub typhus in Thailand. J. Trop. Med. 67:215-19.
Upham, R. W., Jr.; Hubert, A. A.; Phang, 0.W.; Mat, Y. bin; and Rapmund, G. 1971. Distribution of Leptotrombidium (Leptotrombidium) arenicola (Acarina: Trombiculidae) on the ground in West Malaysia. J. Med. Ent. 8: 401-6.
USARV-MB-Scrub typhus. USARV M. Bull. (USARV Pam 40-19), Jan.-Feb. 1970. Copy in Joint Medical Library, Office of the Surgeons General.
USARV M. Bull, See USARV-MB.
Wharton, G. W. 1946. The vectors of tsutsugamushi disease. Proc. Ent. Soc. Washington 48: 171-78.
Whelton, A.;Donadio, J. V.; and Elisberg, B. L. 1968. Acute renal failure complicating rickettsial infections in glucose-6-phosphatedehydrogenase-deficient individuals. Ann. Int. Med. 69:323-28.
Wisseman, C. L., Jr.; Wood, W. H., Jr.; Noriega, A. R.; Jordan, M. E.; and Rill, D. J. 1962. Antibodies and clinical relapse of murine typhus fever followingearly chemotherapy. Ann. Int. Med. 57: 743-54.
Woodward, T. E.; Philip, C. B.; and Loranger, G. L. 1946. Endemic typhus in Manila, Philippine Islands; report of cases and identification of murine rickettsial agent in domestic ratsby complement fixation. J. Infect. Dis. 78: 162-72.
Zarafonetis, C. J. D.; Ingraham, H. S.; and Berry, J. F. 1946. Weil-Felix and typhus complement-fixation tests in relapsing fever, with special reference to B. proteus OX-Kagglutination. J. Immunol. 52:189-99.
Zdrodovskii, P. F., and Golinevich, H. M.1960. The rickettsial disease, tr. B. Haigh. New York: Pergamon Press.
Zinsser, H. 1934. Varieties of typhus virus and epidemiology of the American form of European typhus fever (Brill`s disease). Am. J. Hyg. 20: 513-32.
Zinsser, H. 1935. Rats, lice and history, Boston: Little, Brown & Co.
