Eur. J. Epidemiol. 0392-2990 May 1991, p. 237-245

EUROPEAN JOURNAL

Vol. 7, No. 3

Or EPIDEMIOLOGY

EPIDEMIOLOGY OF RICKETTSIAL DISEASES1 D.H. WALKER .2 and D.B. FISHBEIN** *The University o f Texas Medical Branch at Galveston Department o f Pathology - 101 Keiller Bldg., F09 - Galveston, Texas 77550 **Centers for Disease Control - Rickettsial and Epidemiology Sections, Viral and Rickettsial zoonoses Branch - Division o f Viral and Rickettsial Disease - Center for Infections Disease, Atlanta, Georgia 30333

Keywords: H u m a n ehrlichiosis - Rickettsial spotted fevers - Typhus fevers- Scrub typhus - Q fever - Zoonosis Rickettsial diseases have a diversity of epidemiologic characteristics reflective of the variety of ecologic situations in which the obligate intracellular bacteria are transmitted to humans. For the spotted fever group (SFG) rickettsiae, Rickettsia typhL R. tsutsugamushL Coxiella burnetii, and the human ehrlichial agent, humans are a dead-end host who plays no role in the maintenance of the organism in nature. All rickettsioses exist as zoonoses. Moreover, all rickettsiae are found in infected arthopods, which generally serve as the natural hosts and can transmit the infection to the next generation of ticks, mites, chiggers, or fleas. From our anthropocentric viewpoint, Q fever aerosol infection from parturient animals and Brill-Zinsser disease ignited epidemics of louse-borne epidemic typhus are exceptions. However, silent cycles of C. burnetii in ticks and R. prowazekii in the flying squirrel flea may have maintained these agents in transovarial or enzootic cycles for eons before humans and their domestic animals arrived on the scene. Thus, the epidemiology of rickettsial diseases must be recognized as an unfortunate aberration of the rickettsial economy. Several excellent reviews of rickettsial ecology contain a wealth of useful information (2, 8, 55, 70, 84).

Human Ehrlichiosis T h e first case of a h u m a n infection with an organism closely related to Ehrlichia canis was d o c u m e n t e d in 1986 (49). A 51-year-old m a n became ill 12-14 days after tick bites in rural Arkansas. His hospital course was characterized by fever, hypotension, confusion, acute renal failure requiring hemodialysis, coagulopathy, upper gastrointestinal and cutaneous hemorrhages, pancytopenia, and hepatocellular injury. The diagnosis o f ehrlichial infection was established by observation o f m o r u l a inclusions comprising clusters of 0.2 - 0.8 !am gram negative bacteria within cytoplasmic vacuoles in lymphocytes, monocytes, and 1 P r e s e n t e d at t h e 4th I n t e r n a t i o n a l S y m p o s i u m o n Riskettsiae

and Rickettsial Diseases, Piegtany, C.S.F.R., 1-6 October, 1990. 2 Corresponding author. 237

polymorphonuclear leukocytes and by IFA antibodies reactive with E. canis with titers falling from 640 to 40 in late convalescence. Search a m o n g sera from patients initially suspected to have Rocky Mountain spotted fever (RMSF) revealed cases of h u m a n ehrlichiosis in several southeastern and south central states (33, 39, 79). In some areas endemic for RMSF, the incidence o f h u m a n ehrlichiosis and RMSF were approximately equal. One asymptomatic patient seroconverted after being treated prophylactically with tetracycline after tick bite. Currently, m o r e than 160 cases have been reported from 19 states (15, 16, 20). There is a substantial male predominance (74%). Although patients' ages have ranged from 2-82 years, nearly half of patients are 50 years or older. The seasonality o f onset with 50% of cases in May and June and history o f tick exposure in

Walker D.H. and Fishbein D.B.

Eur. J. Epidemiol.

more than 80% of patients support the hypothesis of a tick vector. Most cases have been rural. Cases have been defined principally by demonstration of a significant antibody response to E. canis, particularly among patients with illness resembling RMSF (39, 79). Thus, the clinical case description may be biased in the proportions of patients reporting fever, headache, myalgia, anorexia, nausea, vomiting, and thrombocytopenia. However, patients with human ehrlichiosis are less likely to manifest rash and more likely to have leukopenia than patients with RMSF. Fatal cases have been reported. Evaluation of sera from a group of Army reservists exposed to ticks in New Jersey revealed a milder illness (62). Among 9 subjects diagnosed serologically, all recalled tick bites, none were hospitalized, and 2 remained asymptomatic without any treatment. All 3 with white blood cell counts performed had leukopenia (< 3600/1.ll). Prospective study of febrile patients admitted to a local hospital in southeastern Georgia during the tick seasons of 1987 and 1988 evaluated 75 patients among whom 8 had ehrlichiosis (32). Frequently observed were onset in May and June (75%), history of tick bite (63%), anorexia (88%), rigors (75%), weight loss (75%), headache (63%), nausea (50%), myalgia (50%), thrombocytopenia (75%), and midly elevated AST (100%) and ALT (75%). Less often patients had arthralgia (38%), vomiting (38%), cough (25%), diarrhea (25%), abdominal pain (25%), and leukopenia (38%). Seven patients treated with tetracycline defervesced in an average of 3 days; one patient treated with penicillin was febrile for 19 days. The incidence (5.3/100,000) was much greater than RMSF in that area. The laboratory test upon which the diagnosis of human ehrlichiosis relies is demonstration of antibody reactive with cell culture-propagated E. canis by indirect immunofluorescence (18). Because sera from healthy persons seldom have antibodies reactive at a titer of 80, the diagnostic criteria are a four-fold rise or fall in titer and a peak titer of>_ 80. A GMT peak of 1280 occurs 6 weeks after onset; at 17-31weeks after onset the GMT again falls to < 80. Thus, the assay appears adequate for diagnosis but not for seroprevalence studies. The specificity is less than perfect; 5% of subjects with diagnostic titers were considered not to have ehrlichiosis. Curiously, 36.5% of subjects with diagnostic titers for ehrlichiosis also have concurrent significant titers for C. burnetii, R. rickettsii, or R. typhi. No clinical, microbiologic, or immunologic explanation is apparent at present. Epidemiologic investigations will advance when we can isolate the etiologic ehrlichia, develop additional confirmatory laboratory diagnostic methods, and determine the vector tick and animal reservoir. Spotted fever group rickettsioses

Oriental spotted fever A human disease caused by a newly recognized

species of SFG rickettsia, R. japonica, has been identified on 3 of the 4 major islands of Japan (Shikoku, Kyushu, and Honshu) as well as the smaller Awaji Island (34, 42, 60, 86-88, 96). The prototype strain of R. japonica was isolated from a 62-year-old male woodcutter who in October 1985 suffered a febrile disease with eschar and rash after a tick bite (87). Four additional strains isolated by Uchida from the blood of patients along the southeast coast of Shikoku Island were shown to be distinct from other pathogenic SFG rickettsiae by microimmunofluorescence, species-specific monoclonal antibodies, and the electrophoretic mobility and antigenic reactivity of the surface-exposed proteins (86, 87). Similarly SFG rickettsiae isolated from this area and an island belonging to Hyogo Prefecture by Okada, Tange, and Kobayashi appear to have the same characteristics as R. japonica (60). Twenty-three patients from the Muroto area of Shikoku Island diagnosed between 1983 and 1986 had an incubation period of 4-7 days, onset between April and October, and illness similar to boutonneuse fever with febrile exanthem (1000/0) and eschar (48%) (34). Four patients on Kyushu and Honshu islands had fever, rash, eschar, and seroconversion to SFG rickettsial antigens by IFA (42.96). This interesting newly recognized disease must be investigated to determine its geographic distribution both within and outside of Japan and to identify its reservoir and vector hosts.

Rocky Mountain spotted fever Rocky Mountain spotted fever occurs in the U.S., Canada, Mexico, parts of Central America including Costa Rica and Panama, and parts of South America including Colombia and Brazil. A well documented series of cases representative of the RMSF in the U.S. showed that the disease occurs most often in young patients, particularly the 5-9-year-old age group, males (55%), and whites (85%) (40). The onset of 92% of cases was between April and August overall (45); however, in Oklahoma, Texas, and Arkansas 11% of cases occurred during October-March with 17% of the Texas cases in these months (81). Fatalities have been associated primarily with delay or failure to give antirickettsial treatment, absent or delayed rash, absence of history of tick exposure, older age, black race, glucose-6-phosphate dehydrogenase deficiency, and wintertime onset (5, 14, 16, 17, 29-31, 40, 90). Active surveillance for RMSF during 1979-81 in two high incidence counties in North Carolina confirmed 98 cases with 3 deaths, an incidence of 14.6/ 100,000~as' compared with rates of 0.2-0.3/100,000 for the U.S. overall (94). Among cases reported solely on a clinical basis, 60% were not actually RMSF. Conversely; some actual cases were not reported. Males accounted for 63% of cases. High incidence was observed not only in 5-9 year old children but also in m e n mgre than 60-years-old. Children 0-14-years-old were taken to physicians sooner (1.37 days) than person 15-years and older (2.47 days). Ticks were quite

238

Vol. 7, 1991

Epidemiology of rickettsial diseases

abundant with 85% of cases and 54% of matched controls reporting tick bite or exposure. Moreover 99% of cases had been exposed to woody or brushy areas within 14 days prior to onset. The unanswered question is what factors determine the quantity of ticks that are infected with virulent R. rickettsii. Recent studies reveal that only a small proportion of ticks are infected by rickettsiae, many of these are R. bellii, which is not a member of the SFG, and a minute proportion of ticks contain virulent R. rickettsii (8, 19, 47, 50, 55, 63-65). Nonpathogenic Rickettsia bellii has been demonstrated to be the predominant rickettsia in 23 counties in 7 states including 39% of Dermacentor andersoni isolates in Montana, 82% of isolates in North Carolina, and 83% of isolates in Ohio. In North Carolina, the state with the highest incidence of RMSF, only 3.4% of D. variabilis contained SFG rickettsiae. Among 72 SFG rickettsiae isolated there, only one was R. ricketts#; the rest were R. montana. In western Montana, 8.3% of ticks contained rickettsia-like organisms in hemolymph. Among rickettsiae isolated, 42.6% were R. rhipicephali, 38.7% R. bellii, 9.4% R. rickettsii, and 7.5% R. montana. On Long Island, New York, all isolates of SFG rickettsiae from adult D. variabilis were R. montana. In a large study of rickettsiae in ticks in Ohio, more than 97% of D. variabilis contained no rickettsiae, 2.2% were infected with R. bellii, and only 0.2% contained SFG rickettsiae. Rickettsia rickettsii was detected in only 0.06% of ticks, and even in counties endemic for RMSF only 0.1-0.6% of ticks contained R. rickettsii. It is easier to find cases of RMSF than R. rickettsii-infected ticks. Burgdorfer has reported both that transovarial transmission of R. rickettsii in ticks is highly efficient and that on occasion massive rickettsial proliferation causes tick death, decreased oviposition, or decreased egg development (8, 9). Even if the pathogenic effect of R. rickettsii only killed a small proportion of infected lines of ticks carrying R. rickettsii transovarially, the effect after hundreds of generations would be the disappearance of RMSF. Few animals have been found to contain viable R. rickettsii in nature, and rickettsemia in susceptible wild animals is of relatively short duration. Nevertheless, ground squirrels, chipmunks, snowshoe hares, meadow voles, and cotton rats develop sufficient rickettsemia to infect some of the feeding larval Dermacentor ticks (8, 10, 11, 35, 58). These immature ticks transmit R. rickettsii transovarially to their progeny, thus establishing new infected tick lines. Thus, the loss of infected tick lines would seem to be balanced by establishment of newly infected tick lines. Dogs do not serve as a reservoir of R. rickettsii for ticks although they bring infected ticks to households (59). It is my hypothesis that the reactivation phenomenon is an adaptive response of the rickettsia to control its growth and potential pathogenicity at the precise times required for maintenance of transstadial and transovarial transmission (91). Rickettsiae must proliferate to spread and infect all the ova to assure

transovarial transmission but overgrowth must not be permitted to damage the tick and ova. A signal in the blood meal might turn on rickettsial binary fission and invasiveness precisely when tick cell division begins. At the level of the tick, some nonpathogenic species of SFG rickettsiae (R. montana, R. rhipicephali, and the East side agent) interfere with the invasion of tick ovaries by R. rickettsii acquired in a blood meal (8). This interference prevents the establishment of R. rickettsii in locations such as the east side of the Bitterroot Valley of Montana where RMSF has never been a problem. That ticks have rarely been found naturally infected with more than one species of SFG rickettsia is evidence that such interference is effective in nature. This interference mechanism is less important where more than 95% of ticks contain no rickettsiae than on the East side of the Bitterroot Valley where most of the ticks contain the East side agent in their ovaries. Attempts to understand the conditions necessary for maintenance of ticks and R. rickettsii and explanation of their encountering humans include quantitative statistical analysis of geographic, population, and climatologic data (43, 57). Increasing population by suburbanization, outdoors recreational activities, and mean winter temperatures slightly above 0°C may explain somewhat the increased incidence of RMSF of the 1960's and 1970's; however, to the best of my knowledge, no hypotheses have been advanced to explain the decrease in incidence of RMSF from 1126 cases in 1983 to 592 cases in 1987. The bimodal peaks of RMSF in oak-hickory and Appalachian oak zones may be associated with geographic and climate-determined occurrence of feeding D. variabilis adults. What is perplexing is the asynchronous geographic fluctuations in incidence of RMSF. During 1981-85 the Atlantic Coast states (Georgia, Maryland, North Carolina, South Carolina, and Virginia) and the midsouth states (Alabama, Tennessee, and Mississippi) had a steady decline in RMSF as did the midwestern states (Illinois, Indiana, and Ohio) during 1982-85. Meanwhile, Oklahoma, Texas, and Arkansas underwent an increase in incidence of RMSF of 107% from 1981 to 1983 followed by a decrease of 64% over 1984 and 1985 (81). Other unexplained drastic fluctuations include New Jersey with no cases in 1988 and 26 cases in 1989, Pennsylvania with 2 cases in 1988 and 23 in 1989, and Kansas with 10, 30, 26, and 11 cases in 1986, 1987, 1988, and 1989, respectively (14, 16). While RMSF in endemic areas like the South Atlantic states does not have the characteristics of a focal disease, an outbreak of 4 cases of RMSF in 1987 in the south Bronx area of New York City resembles such a focus (72). There are an abundance of vacant lots and a large park where 8°/0 of the D. variabilis ticks contained SFG rickettsiae. Similar foci of urban cases have been reported in Philadelphia and Ohio. Near the edge of the geographic distribution of elements necessary to result in a case of RMSF, it would seem most difficult to predict what might happen. Under

239

Walker D.H. and Fishbein D.B.

Eur. J. Epidemiol.

these conditions, the setting for focal disease might occur.

Queensland tick typhus After years of silence, Queensland tick typhus is attracting interest and investigation in Australia (76). The first report of a fatal case of this disease by a group of investigators indicated that recent active surveillance is discovering a substantial number of cases that are not reported to health departments. The incidence may be higher, the clinical spectrum broader, and the geographic distribution wider including Victoria and island cases of south of continental Australia (12, 76).

Boutonneuse fever A simultaneous increase in boutonneuse fever in Italy, Spain, France, and Portugal, Israeli spotted fever, and RMSF was observed during the 1970's (51, 75). It was suggested that an unidentified common factor might have been in effect. Increased incidence of boutonneuse fever in northern Spain during 197887 was associated with higher temperatures and less rainfall (4). The incidence of spotted fever rickettsiosis in the Black Sea and Caspian Sea Basins was not reported to have increased (70). Uncertainty exists regarding the precise etiology of the spotted fevers of Southern Europe, Africa, the Middle East, and Southeast Asia. Organisms designated as R. conorii exhibit considerably more antigenic and genetic diversity than human isolates of R. rickettsii. It appears that the etiologic agent of Israeli spotted fever differs substantially from Moroccan, Kenya tick typhus, and Indian strains of R. conorii. Israeli spotted fever differs from boutonneuse fever in that eschars do not occur (37, 38). Monoclonal antibodies to a human Sicilian isolate of SFG rickettsia have a variety of reactivities with 4 Sicilian tick isolates of SFG rickettsiae (89). Spotted fever rickettsiosis in Sicily is associated with a high proportion of SFG rickettsia-infected Rhipicephalus sanguineus ticks on dogs with antibodies to SFG rickettsiae (85). Because dogs and R. sanguineus are found in urban, suburban, and rural environments, boutonneuse fever may be found equally in these settings (4, 83). In Israel the peak of spotted fever rickettsiosis between August and October coincides with the maximal prevalence and activity of R. sanguineus. Rickettsia conorii is introduced periodically into northern Europe where autochthonous cases of boutonneuse fever are diagnosed (44). Rickettsia conorii was isolated from R. sanguineus introduced into Switzerland on a pet dog. The description of severe and fatal boutonneuse fever has been followed by definition of risk factors for severe disease including old age, underlying disease such as diabetes mellitus or cardiac disease, chronic alcoholism, and glucose-6-phosphate dehydrogenase deficiency (66). North Asian tick typhus North Asian tick typhus, which has been studied extensively in the USSR, has recently been documented to occur in the northern provinces of China (Xinjiang Uygur Autonomous Region, Inner Mongolia, and Heilongjiang) (21-24, 70). Rickettsia sibirica has been isolated from ill humans and D. nuttalli ticks in Xinjiang and Inner Mongolia (93). In an endemic area of North Asian tick typhus in Xinjiang, an incidence involving 20/0of the population was observed in April and May, 1978.

Seroprevalence studies for SFG rickettsioses Investigations to determine the prevalence of antibodies to R. conorii, R. sibirica, R. rickettsii, and R. japonica require some caution in interpretation. IFA and ELISA tests have been demonstrated to be able to detect antibodies that persist for years after SFG rickettsial infection. Complement fixation, latex agglutination, and indirect hemagglutination tests measure populations of antibody that do not persist; thus, these assays should not be used for seroprevalence studies. Likewise, it is important to select a threshold value that will not include false positive results. For iFA against SFG rickettsiae, a titer of < 40 is highly suspect. In addition, because the lipopolysaccharide of SFG rickettsiae contains antigens that are shared among the members of the SFG, it is not possible to distinguish which rickettsial species stimulated the humoral immune response. Serosurveys have demonstrated that substantial portions of the population have antibodies reactive with SFG rickettsiae in Spain, France, including Corsica, Sicily, Israel, Central African Republic, Ivory Coast, China (Xinjiang, Inner Mongolia, and Heilongjiang in the north and Hainan Island and Yunnan in the south), Japan (Shikoku Island), and U.S. (Texas) (21-23, 36, 48, 52, 67, 68, 73, 78, 82, 95). It has been suggested that these represent asymptomatic infection with pathogenic members of the group such as R. conorii, R. sibirica, and R. rickettsii. Demonstration of asymptomatic seroconversion has, in fact, been achieved only rarely (73). Retrospective questioning has in some cases indicated that patients had illnesses consistent with SFG rickettsiosis (82). My own particular interest in the high prevalence of antibodies reactive with R. sibirica in Inner Mongolia stems from the demonstration of antibodies in 18-37% of subjects (48). The isolation and specific identification of R. sibirica in the blood of a man with fever of more than two week's duration and no rash, eschar, or history of tick bite shows that symptomatic, undiagnosed North Asian tick typhus occurs. A higher index of suspicion will be required to diagnose patients without a rash and eschar. Problems include physician unawareness of SFG rickettsiosis, failure to consider the diagnosis, unfamiliar symptoms not classically associated with rickettsioses, and lack of diagnostic laboratory tests. SFG rickettsiosis can occur

240

Vol. 7, 1991

Epidemiology of rickettsial diseases

for years as in the Southeastern U.S. prior to 1930 without being recognized. An alternative possible explanation for the prevalence of antibodies is a mild infection with a SFG rickettsia of undetermined pathogenicity such as R. rhipicephali and R. parkeri, both of which cause nonfatal febrile disease in guinea pigs inoculated intraperitoneally. Typhus fevers

Murine typhus Murine typhus is widely distributed, causes severe morbidity, and is frequently misdiagnosed. An outbreak of febrile disease at a refugee camp in a rural area on the Thailand-Kampuchea border 8 months after its construction was demonstrated to be due to murine typhus (6). Infection occurred in the dry season and was associated with Rattus exulans and Xenopsylla cheopis. South Texas continues to be the stronghold of murine typhus in the U.S. (80). Murine typhus there peaks in June and occurs particularly in Hispanics. The incubation period averages 5.5 days. Rash is observed in 58%; there is 1% mortality. In China murine typhus has been diagnosed in both the south (Yunnan and Hainan Island) and the north (Henan, Hebei, Shandong, and Beijing) (22, 92). Azad has continued his laboratory studies of the host relationships of fleas and R. typhi. Some of his data challenge prior concepts. Transovarial transmission of R. 0/phi in a small proportion of X. cheopis may be a factor that enables low level survival of rickettsiae during periods of otherwise limiting conditions (27). Moreover, experimental observations support the transmission of R. typhi by the bite of X. cheopis as well as by its infected feces (26). The more rapid growth ofR. typhi in fleas at 240 or 30o C than at 18o C suggests an explanation for why murine typhus occurs regularly in warm rather than cool climes. A dampening effect on the spread of R. O/phi may be its reduced transmission by fleas feeding on a host with antibodies to R. typhi (25). Among ectoparasites suspected of playing a role in the maintenance and transmission of R. typhi, the cat flea, Ctenocephalides felis, was studied and shown to acquire R. typhi from feeding and to support logarithmic growth of rickettsiae for 9 days (28). These fleas bite humans, opossum, dogs, skunks, and raccoons in addition to cats. They have long been suspected to play a role in murine typhus in a cycle involving opossum.

Epidemic typhus fever Epidemic louse-borne typhus continues to flare up, often only belatedly recognized, in locations where poverty is a major problem (3). Information previously not available from China indicates that there was a very high incidence of epidemic typhus during and after World War II (46, 106 cases with 1,269 deaths registered in 1946) (22). In the 1950's louse-

borne typhus occurred in poor mountainous areas where the provinces Guizhou, Yunnan and Sichuan meet and in the 1960's in Heilongjiang, Jilin, and Liaoning. A focus identified in northeastern Inner Mongolia in 1979 was investigated demonstrating high prevalence of antibodies in the areas with unaffected population indicating immunity from prior infection. Flying squirrel-associated R. prowazekii infection in the U.S. has been diagnosed in 35 persons (13). A case of typhus fever with intense exposure to flying squirrels in southeastern Wisconsin in January was characterized by fever, transient rash, nonproductive cough, and prompt response to tetracycline (1). Cases are sporadic and mostly rural or suburban and with wintertime occurrence. The existence of this zoonotic cycle indicates that typhus fever cannot be eradicated by human immunization and suggests that R. prowazekii might have originated in the Western Hemisphere.

Scrub typhus The epidemiology of scrub typhus is confounded by antigenic and genetic diversity of strains for which no correlations with virulence for humans are known. Moreover, protective immunity is short-lived, particularly for heterologous strains. Thus, the effect of a particular strain on an individual may be affected by undefined rickettsial virulence factors, immune status spanning the entire spectrum between susceptibility and complete protective immunity, and possibly host factors such as G-6-PD deficiency. Whatever the explanation, outbreaks have varied in virulence from 0% to 35% mortality. The ongoing importance of scrub typhus is emphasized in rural Malaysia where scrub typhus is the most frequent diagnosis (19.3%) in febrile hospitalized patients (7). Leptotrombidium larvae are infected transovarially; there is no evidence that uninfected chiggers can acquire R. tsutsugamushi from rats and transmit it transovarially to their progeny. Apparently R. tsutsugamushi has no interference mechanism because several serotypes have been isolated from a single chigger source. In the Pescadores Islands, there was a close correlation between monthly indices of L. deliense for Suncus murinus shrews and Rattus sp. and the number of cases of laboratory-confirmed scrub typhus (61). Weekly indices from S. murinus correlated more closely with weekly numbers of cases. Scrub typhus activity has been mapped out in many provinces of China (Guangdong, Fujian, Zhejiang, Guangxi, Yunnan, Hunan, Sichuan, Tibet, Shandong, Hainan, and Jiangsu) (22, 92). There were more than 10,000 cases of scrub typhus on Hainan Island in 1956-1985, and the seroprevalence is very high. Vectors in China include L. deliense, L. gaohuensis (in Zhejiang), L. scutellaris (in the recent autumn outbreak in the Yimong Mountains of Shandong), L. insulare (in Zhejiang), and L. jishoum (in Hunan). In Toyama Prefecture (Japan), endemic

241

Walker D.H. and Fishbein D.B.

Eur. J. Epidemiol.

a n d non-endemic areas were investigated with isolation of R. tsutsugamushi from 36% of endemic area rodents, while no rickettsiae were present in nonendemic area rodents (41). While few L. pallidum were present in non-endemic areas, endemic areas had two peaks of L. pallidum yearly, in spring and autumn. Rodent isolates of rickettsiae were high in November during the L. pallidum peak. Endemic area rodents had 55% seropositivity in comparison with only 17% for rodents in non-endemic areas. What remains to be determined is why one area has more infected L. pallidum than the other.

Q fever The rich literature of Q fever ecology and epidemiology has been reviewed quite extensively recently (70, 74). Outbreaks of acute Q fever associated with sheep continue to occur (56, 69, 77); sporadic cases of chronic Q fever affect patients with pathologic and prosthetic valves and vessels. Outbreaks of Q fever in Nova Scotia have been traced to exposure to parturient cats and wild rabbits (46, 53). Coxiella burnetii has been isolated from the uteri of two of these cats, and antibodies to C. burnetii were demonstrated in 45% of snowshoe hares in one area of exposure to those animals. A community based longitudinal study in Nova Scotia showed a 2% rate of seroconversion (30 cases/ 1000 population); none had symptoms of Q fever (54). Studies of infants in the Netherlands infected with C. burnetii has shown that many were exposed to ruminants or animal products (71). Q fever has been diagnosed in 12 provinces in China, as well as in Beijing (22). These areas extend across north China from Heilongiiang through Inner Mongolia and Gansu to Xinjiang, involve the western provinces, Qinghai, Sichuan, Yunnan, and Tibet, the southern provinces, Guangxi and Hainan, and Fujian and Anhui in the east. Application of molecular analysis of plasmids and lipopolysaccharide antigens of prospectively isolated C. burnetii from well investigated patients will allow confirmation of the correlates of sources of infection, characteristics of the organisms, and the patient's particular pathologic lesions (e.g., acute pneumonia, acute nonpneumonic flu-like syndrome, asymptomatic infection, granulomatous hepatitis, or chronic endocarditis). A WHO memorandum promulgated in 1983 stated, "Previously unrecognized rickettsial diseases are in fact contributing substantially to the acute febrile disease burden of many populations... Rickettsial diseases are under-reported... National and international funding agencies should be asked to provide assistance for surveillance and research on rickettsioses'. Presently we need greater financial support and effort in investigation of rickettsial diseases and their epidemiology. 242

REFERENCES 1. Agger W.A. and Songsiridej V. (1985): Epidemic typhus acquired in Wisconsin - Wis. Med. J. 84: 2730. 2. Azad A.F. (1988): Relationship of vector biology and epidemiology of louse- and flea-borne rickettsioses In: Biology of Rickettsial Diseases Vol. II, CRC Press, Boca Raton, pp. 51-61. 3. Banegura F.K., Munube G.M.R., Senyakazana A.G. and Nzaro E. (1983): An epidemic of typhus in Ankole, Uganda, 1975-1976 - East African Med. J. 60: 55-59. 4. BarandaJ.M., CarrencejaJ.C. and Errasti C.A. (1985): Q fever in the Basque Country: 1981-1984 - Rev. Infect. Dis. 7: 700-701. 5. BernardK. W., IIelmick C.G., Kaplan J.E. and Winkler W.G. (1982): Surveillance of Rocky Mountain spotted fever in the United States, 1978-1980 - J. Infect. Dis. 146: 297-299. 6. BrownA.E., Meek S.R., Maneechai N. and Lewis G.E. (1988): Murine typhus among Khmers living at an evacuation site on the Thai-Kampuchean border Am. J. Trop. Med. Hyg. 38: 168-171. 7. Brown G.W., Shirai A., Jegathesan M., Burke D.S., Twartz J.C., Saunders J.P. and HuxsoH D.L. (1984): Febrile illness in Malaysia: An analysis of 1,629 hospitalized patients - Am. J. Trop. Med. Hyg. 33: 311-315. 8. Burgdorfer W. (1988): Ecological and epidemiological considerations of Rocky Mountain spotted fever In: Biology of Rickettsial Diseases Vol. II, CRC Press, Boca Raton, pp. 33-50. 9. Burgdorfer W. and Brinton L.P. (1975): Mechanisms of transovarial infection of spotted fever rickettsiae in ticks - Ann. N.Y. Acad. Sci. 266: 61-72. 10. Burgdotfer W., Cooney J.C., Mavros A.J., JeHison W.L. and Maser C. (1980): The role of cottontail rabbits (Sylvilagus spp.) in the ecology of Rickettsia rickettsii in the United States - Am. J. Trop. Med. Hyg. 29: 686-690. 11. Burgdo(er W., Friedhoff K.T. and Lancaster J.L. Jr. (1966): Natural history of tickborne spotted fever in the USA - Bull. W.H.O. 35: 149-153. 12. Campbell R.W., Abeywickrema P. and Fenton C. (1979): Queensland tick typhus in Sydney: a new endemic focus -Med. J. Aust. 1: 350-351. 13. Centersfor Disease Control (1982): Epidemic typhus associated with flying squirrels - United States M.M.W.R. 31: 555-561. 14. Centersfor Disease Control (1988): Rocky Mountain spotted fever - United States, 1987 - M.M.W.R. 37: 388-389. 15. Centers for Disease Control (1988): Human ehrlichiosis - United States - M.M.W.R. 37: 270-277.

Vol. 7, 1991

Epidemiology of rickettsial diseases

16. Centers for Disease Control (1990): Rocky Mountain spotted fever and human ehrlichiosis - United States, 1989 - M.M.W.R. 39: 281-284.

(1990): Fatal cases of Rocky Mountain spotted fever in the United States, 1981-1988 - Ann. N.Y. Acad. Sci. 590: 246-247.

17. D'Angelo L.J., Bregman D.J. and Winkler W.G. (1982): Rocky Mountain spotted fever in the United States: Use of age-specific incidence to determine public health policy for a vector-borne disease - So. Meal. J. 75: 3-5.

30. Eishbein D.B., Kaplan J.E., Bernard K. W. and Winkler W.G. (1984): Surveillance of Rocky Mountain spotted fever in the United States, 1981-1983 - J. Infect. Dis. 150: 609-611. 31. Fishbein D.B., Kaplan J.E., Bernard K. W. and Winkler IV.G. (1984): Surveillance of Rocky Mountain spotted fever - M:M.W.R. 33: 15SS-18SS.

18. Dawson J.E., Fishbein D.B., Eng T.R., Redus M.A. and Greene N.R. (1990): Diagnosis of human ehrlichiosis with the indirect fluorescent antibody test: Kinetics and specificity - J. Infect. Dis. 162: 91-95.

32. Fishbein D.B., Kemp A., Dawson J.E., Greene N.R., Redus M.A. and Fields D.H. (1989): Human ehrlichiosis: Prospective active surveillance in febrile hospitalized patients - J. Infect. Dis. 160: 803-809.

19. Elliot L.B., Fournier P.V. and Teltow G.J. (1990): Rickettsia in Texas - Ann. N.Y. Acad. Sci. 590: 221226. 20. Eng T.R., Fishbein D.B., Dawson J.E., Greene C.R. and Redus M. (1990): Surveillance of human ehrlichiosis in the United States: 1988 - Ann. N.Y. Acad. Sci. 590: 306-307.

33. Fishbein D.B., Sawyer L.A., Holland C.J., Hayes E.B., Okoroanyanwu W., Williams D., Sikes R.K., Ristic M. and McDade J.E. (1987): Unexplained febrile illnesses after exposure to ticks - Infection with an Ehrliehia J.A.M.A. 257: 3100-3104.

21. Fan M.Y., Walker D.H., Liu Q.H., Li tt., Bai H.C., Zhang J.K, Lenz B. and Cai H. (1987): Rickettsial and serologic evidence for prevalent spotted fever rickettsiosis in Inner Mongolia - Am. J. Trop. Med. Hyg. 36: 615-620.

34. Funato T., Kitamura Y., Kawamura A. and Uchida T. (1988): Rickettsiosis of spotted fever group encountered in Muroto area of Shikoku, Japan Clinical and epidemiological features of 23 cases - J. Jpn. Assoc. Infect. Dis. 62: 783-791.

22. Fan M.Y., Walker D.H., Yu S.R. and Liu Q.H. (1987): Epidemiology and ecology of rickettsial diseases in the People's Republic of China - Rev. Infect. Dis. 9: 823 -840.

35. Gage K.L., Burgdolfer W. and Hopla C.E. (1990): Hispid cotton rats (Sigmodon hispidus) as a source for infecting immature Dermacentor variabilis (Acari: Ixodidae) with Rickettsia rickettsii - J. Med. Entomol. 27: 615-619.

23. Fan M.Y., Wang J.G., Jiang EX., Zong D.G., Lenz B. and Walker D.H. (1987): Isolation of a spotted fever group rickettsia from a patient and related ecologic investigations in Xinjiang Uygur Autonomous Region of China - J. Clin. Microbiol. 25: 628-632.

36. Gonzales J.P., Fiset P., Georges A.J., Saluzzo J.R. and Wisseman C.L. Jr. (1985): Approche serologique sur l'incidence des rickettsioses en Republique Centrafricaine - Bull. Soc. Path. Ex. 78: 153-156.

24. Fan M.Y., Yu X.J. and Walker D.tt. (1988): Antigenic analysis of Chinese strains of spotted fever group rickettsiae by protein immunoblotting - Am. J. Trop. Med. Hyg. 39: 497-501.

37. Gross E.M., Arbefi Y., Bearman J.E., Cohar K., Torok V. and Goldwasser Spotted fever and murine typhus in the region of Israel, 1981 - Bull. W.H.O.

25. Farhang-AzadA. and Emala M.A. (1987): Suppression of Rickettsia typhii transmission in fleas maintained on murine typhus-immune rats - Am. J. Trop. Med. Hyg. 37: 629-635.

38. Gross E.M. and Yagupsky P. (1987): Israeli rickettsial spotted fever in children - Acta Trop. 44: 91-96.

26. Farhang-Azad A. and Traub R. (1985): Transmission of murine typhus rickettsiae by Xenopsylla cheopis, with notes on experimental infection and effects of temperature - Am. J. Trop. Med. Hyg. 34: 555-563. 27. Farhang-Azad A., Traub R. and Baqar S. (1985): Transovarial transmission of murine typhus rickettsiae in Xenopsylla cheopis fleas - Science 227: 543-545. 28. Farhang-Azad A., Traub R., Soft M. and Wisseman C.L. Jr. (1984): Experimental murine typhus infection in the cat flea, Ctenocephalides felis (Siphonaptera: Pulicidae) - J. Med. Entomol. 21: 675-680. 29. Fishbein D.B., Frontini M.G., Giles R. and Vernon L.L. 243

Yagupsky P., R.A. (1984): Negev desert 62: 301-306.

39. Harkess J.R., Ewing S.A., Crutcher J.M., Kudlac J., McKee G. and Istre G.R. (1989): Human ehrlichiosis in Oklahoma - J. Infect. Dis. 159: 576-579. 40. Helm ick C.G., Bernard K. W. and D'A ngelo L.J. (1984): Rocky Mountain spotted fever: Clinical, laboratory, and epidemiological features of 262 cases - J. InfeCt. Dis. 150: 480-488. 41. Ishikura M., Watanabe M., Morita O. and Uetake H. (1985): Epidemiological studies on the background of the endemic occurrence of tsutsugamushi disease in Toyama Prefecture - Microbiol. Immunol. 29: 859-872. 42. Kaiho L, Tokieda M., Ohtawara M., Uchiyama T. and Uchida T. (1988): Occurrence of rickettsiosis of spotted fever group in Chiba Prefecture of Japan Jpn. J. Meal. Sci. Biol. 41: 69-71.

Walker D.H. and Fishbein D.B.

Eur. J. Epidemiol.

43. Kaplan J.E. and Newhouse V.F. (1984): Occurrence of Rocky Mountain spotted fever in relation to climatic, geophysical, and ecologic variables - Am. J. Trop. Med. Hyg. 33: 1281-1282.

D'Angelo L.J. and Smith ZD. (1986): Rocky Mountain spotted fever in Georgia, 1961-75: Analysis of social and environmental factors affecting occurrence Public Health Rep. 101: 419-428.

44. Lambert M., Dugernier T., Bigaignon G., Rahier J. and Piot P. (1984): Mediterranean spotted fever in Belgium - Lancet 2: 1038.

58. Norment B.R. and BurgdooCer W. (1985): Susceptibility of small mammals to nonvirulent spotted fever group rickettsiae - J. Med. Entomol. 22: 200-203.

45. Lange Z V., Walker D.H. and Wester T.B. (1982): Documented Rocky Mountain spotted fever in wintertime - J.A.M.A. 247: 2403-2404.

59. Norment B.R. and Burgdorfer W. (1984): Susceptibility and reservoir potential of the dog to spotted fevergroup rickettsiae - Am. J. Vet. Res. 45: 1706-1710.

46. Langley J.M., Marrie T.J., Covert A., Waag D.M. and Williams J.C. (1988): An urban outbreak of Q fever following exposure to a parturient cat - N. Engl. J. Med. 319: 354-356.

60. Okada T., Tange Y. and Kobayashi Y. (1990): Causative agent of spotted fever group rickettsiosis in Japan - Infect. Immun. 58: 887-892.

47. Linnemann C.C. Jr., Schaeffer A.E., BurgdooCer W., Hutchinson L. and Philip R.N. (1980): Rocky Mountain spotted fever in Clermont County, Ohio II. Distribution of population and infected ticks in an endemic area - Am. J. Epidemiol. 111: 31-36.

61. Olson ZG., Bourgeois A.L. and Fang R.C.Y. (1982): Population indices of chiggers (Leptotrombidium deliense) and incidence of scrub typhus in Chinese military personnel, Pescadores Islands of Taiwan, 1976-77 - Trans. Royal Soc. Trop. Med. Hyg. 76: 8588.

48. Liu Q.H., Walker D.H. and Zhou G.F. (1990): Serologic survey for antibodies to Rickettsia sibirica in Inner Mongolia, People's Republic of China Ann. N.Y. Acad. Sci. 590: 237-242. 49. Maeda K., Markowitz N., Hawley R.C., Ristic M., Cox D. and McDade J.E. (1987): Human infection with Ehrlichia canis, a leukocytic rickettsia - N. Engl. J. Med. 316: 853-856. 50. Magnarelli L.A., Anderson J.F., Philip R.N., Burgdorfer W. and Chappell W.A. (1983): Rickettsiae-infected ticks (acari: Ixodidae) and seropositive mammals at a focus for Rocky Mountain spotted fever in Connecticut, USA - J. Med. Entomol. 20: 151-156. 51. Mansueto S., Tringali G. and Walker D.H. (1986): Widespread, simultaneous increase in the incidence of spotted fever group rickettsioses - J. Infect. Dis. 154: 539-540. 52. Mansueto S., Vitale G., Miceli M.D., Tringali G., Quartararo P., Pieone D.M. and Oechino C. (1984): A sero-epidemiological survey of asymptomatic cases of boutonneuse fever in Western Sicily - Trans. Royal Soc. Trop. Med. Hyg. 78: 16-18. 53. Marrie T.J., Sehlech W.F. III, Williams J.C. and Yates L. (1986): Q fever pneumonia associated with exposure to wild rabbits - Lancet 1: 427-429. 54. Marrie T.J. and Yates L. (1990): Incidence of Q fever: Pilot studies in two areas in Nova Scotia - Ann. N.Y. Acad. Sci. 590: 275-280. 55. McDade J.E. and Newhouse V.F. (1986): Natural history of Rickettsia riekettsii - Ann. Rev. Microbiol. 40: 287-309. 56. Meiklejohn G., Reimer L.G., Graves P.S. and Helmick C. (1981): Cryptic endemic of Q fever in a medical school - J. Infect. Dis. 144: 107-113. 57. Newhouse V.F., Choi K., Holman R.C., Thacker S.B., 244

62. Petersen L.R., Sawyer L.A., Fishbein D.B., Kelley P. W., Thomas R.J., Magnarelli L.A., Redus M. and Dawson J.E. (1989): An outbreak of ehrlichiosis in members of an army reserve unit exposed to ticks - J. Infect. Dis. 159: 562-568. 63. Philip R.N. and Casper E.A. (1981): Serotypes of spotted fever group rickettsiae isolated from Dermacentor andersoni (Stiles) ticks in Western Montana - Am. J. Trop. Med. Hyg. 30: 230-238. 64. Philip R.N., Casper E.A., Anacker R.L., Cory Z, Hayes S.F., Burgdorfer W. and Yunker C.E. (1983): Rickettsia bellii sp. nov.: a tick-borne rickettsia, widely distributed in the United States, that is distinct from the spotted fever and typhus biogroups - Int. J. Syst. Bacteriol. 33: 94-106. 65. Pretzman C., Daugherty N., Poetter K. and Ralph D. (1990): The distribution and dynamics of rickettsia in the tick population of Ohio - Ann. N.Y. Acad. Sci. 590: 227-236. 66. Raoult D., Lena D., Perrimont H., Gallais H., Walker D.H. and Casanova P. (1986): Haemolysis with Mediterranean spotted fever and glucose-6phosphate dehydrogenase deficiency - Trans. Royal Soc. Trop. Med. Hyg. 80: 961-962. 67. Raoult D., Nicolas D., DeMicco P., Gallais H. and Casanova P. (1985): Aspects epidemiologiques de la fievre boutonneuse Mediterraneenne en Corse du Sud - Bull. Soc. Path. Ex. 78: 446-451. 68. Raoult D., Toga B., Chaudet H. and Chiche-Portiche C. (1987): Rickettsial antibody in southern France: antibodies to Rickettsia conorii and Coxiella burnetii among urban, suburban and semi-rural blood donors Trans. Royal Soc. Trop. Med. Hyg. 81: 80-81. -

69. Rauch A.M., Tanner M., Pacer R.E., Barrett M.J., Brokopp C.D. and Schonberger L.B. (1987): Sheep-

Vol. 7, 1991

Epidemiology of rickettsial diseases

associated outbreak of Q fever, Idaho - Arch. Intern. Med. 147: 341-344.

76 cases in the years 1977-1986 - Eur. J. Epidemiol. 5: 516-520.

70. Rehacek J. and Tarasevich/. V. (1988): In: Acari-Borne Rickettsiae & Rickettsioses in Eurasia, Veda Publishing House of the Slovak Academy of Sciences, Bratislava.

84. Traub R., Wisseman C.L. Jr. and Farhang-Azad A. (1978): The ecology of murine typhus: a critical review - Trop. Dis. Bull. 75: 238-317.

71. Richardus J.H., Dumas A.M., Huisman aT. and Schaap G.J.P. (1985): Q fever in infancy: a review of 18 casesPediatr. Infect. Dis. 4: 369-373. 72. Salgo M.P., Telzak E.E., Currie B., Perlman D.C.,

Litman N., Levi M., Nathenson G., Benach J.L., A1Hafidh R. and Casey J. (1988): A focus of Rocky Mountain spotted fever within New York City - New Engl. J. Med. 318: 1345-1348. 73. Sarov B., Galil A., Sikuler E., Yagupsky P., Saah A., Gilad A., Naggan L. and Sarov L (1990): Prospective study on symptomatic versus asymptomatic infections and serological response to spotted fever group rickettsiae in two rural sites in the Negev (Southern Israel) - Ann. N.Y. Acad. Sci. 590: 243-245.

85. Tringali G., Intonazzo V., Perna A.M., Mansueto S., Vitale and Walker D.H. (1986): Epidemiology of boutonneuse fever in Western Sicily: Distribution and prevalence of spotted fever group rickettsial infection in dog ticks (Rhipicephalus sanguineus) Am. J. Epidemiol. 123: 721-727. 86. Uchida T., Uchiyama T. and Koyama A.H. (1988): Isolation of spotted fever group rickettsiae from humans in Japan - J. Infect. Dis. 158: 664-665. 87. Uchida T., Tashiro F., Funato T. and Kitamura Y. (1986): Isolation of a spotted fever group rickettsia from a patient with febrile exanthematous illness in Shikoku, Japan - Microbiol. Immunol. 30: 13231326. 88. Uchida T., Yu X.J., Uchiyama T. and Walker D.H. (1989): Identification of a unique spotted fever group rickettsia from humans in Japan - J. Infect. Dis. 159: 1122-1126.

74. Sawyer L.A., Fishbein D.B. and McDade J.E. (1987): Q fever: Current concepts - Rev. Infect. Dis. 9: 935-946. 75. Segura-Porta F., Font-Creus B., Espejo-Arenas E. and Bella-Cueto F. (1989): New trends in Mediterranean spotted fever - Eur. J. Epidemiol. 5: 438-443. 76. Sexton D.J., King G. and Dwyer B. (1990): Fatal Queensland tick typhus - J. Infect. Dis. 162: 779-780.

89. Vitale G., Di Stefano R., Damiani G. and Mansueto S. (1989): Characterization of Sicilian strains of spotted fever group rickettsiae by using monoclonal antibodies - J. Clin. Microbiol. 27: 1081-1085.

77. Simor A.E., Brunton J.L., Salit LE., Vellend H., FordJones L. and Spence L.P. (1984): Q fever: hazard from sheep used in research - Can. Med. J. 130: 1013-1016.

90. Walker D.H. (1990): The role of host factors in the severity of spotted fever and typhus rickettsioses Ann. N.Y. Acad. Sci. 590: 10-19.

78. Soliman

91. Walker D.H. (1989): The rickettsia-host interaction In: Intracellular Parasitism - Boca Raton, CRC Press, Inc., pp. 79-92.

A.K., Botros B.A.M., Ksiazek T.G., Hoogstraal H., Helmy I. and Morrill J.C. (1985): Seroprevalence of Rickettsia typhi and Rickettsia conorii infection among rodents and dogs in Egypt J. Trop. Med. Hyg. 92: 345-349.

79. Taylor J.P., Betz T.G., Fishbein D.B., Roberts M.A., Dawson J. and Ristic M. (1988): Serological evidence of possible human infection with Ehrlichia in Texas J. Infect. Dis. 158: 217-220. 80. Taylor J.P., Betz T.G. and Rawlings J.A. (1986): Epidemiology of routine typhus in Texas, 1980 through 1984 - J.A.M.A. 255: 2173-2176. 81. Taylor J.P., Istre G.R. and McChesney T.C. (1988): The epidemiology of Rocky Mountain spotted fever in Arkansas, Oklahoma, and Texas, 1981 through 1985 Am. J. Epidemiol. 127: 1295-1301. 82. Taylor J.P., Tanner W.B., Rawlings J.A., Buck J., Elliott L.B., Dewlett H.J., Taylor B. and Betz T.G. (1985): Serological evidence of subclinical Rocky Mountain spotted fever infections in Texas - J. Infect. Dis. 151: 367-369. 83. Tinelli M., Maccabruni A., Michelone G. and Zambelli A. (1989): Mediterranean spotted fever in Lombardy: an epidemiological, clinical and laboratory study of 245

92. Walker D.H., Liu Q.H. and Feng H.M. (1990): Update on rickettsiae and rickettsial diseases in China - Rev. Infect. Dis. 12: 562-563. 93. Wang J.G. and Walker D.H. (1987): Identification of spotted fever group rickettsiae from human and tick sources in the People's Republic of China - J. Infect. Dis. 156: 665-669. 94. Wilfert C.M., MacCormack J.N., Kleeman K., Philip R.N., Austin E., Dickinson V. and Turner L. (1984): Epidemiology of Rocky Mountain spotted fever as determined by active surveillance - J. Infect. Dis. 150: 469-479. 95. Yagupsky P., Sarov B., Naggan L., Sarov I., Avy K. and Goldwasser R. (1990): The prevalence of IgG antibodies to spotted-fever group rickettsiae among urban and rural dwelling children in Southern Israel Scand. J. Infect. Dis. 22: 19-23. 96. Yamamoto S., Kawabata N., Uchiyama T. and Uchida T. (1987): Evidence for infections caused by spotted fever group rickettsia in Kyushu, Japan - Jpn. J. Med. Sci. Biol. 40: 75-78.

Epidemiology of rickettsial diseases.

Rickettsial diseases have a diversity of epidemiologic characteristics reflective of the variety of ecologic situations in which the obligate intracel...
1MB Sizes 0 Downloads 0 Views