Spotted fever and murine typhus in the Negev desert region of Israel, 1981

Bulletin ofthe World Health Organization, 62(2): 301-306(1984) © World Health Organization 1984 Spotted fever and murine typhus in the Negev desert region of Israel, 1981 ELLIS M. GROSS,' YAEL ARBELI,1 JACOB E. BEARMAN,2 PABLO YAGUPSKY,3 KATHERINE COHAR,4 VIOLA TOROK,' & ROBERT A. GOLDWASSER, During 1981, 52 cases of rickettsiosis caused by the spottedfever group of rickettsiae were reported in a population of250 000 living in the Negev desert region of Israel. Retro- spectively, at least S of these cases were serologically shown to have been caused by Rickettsia typhi, the agent of murine typhus. Most cases of rickettsial disease occurred in the summer months and one half of all cases occurred in children aged 9 years or less. WhileJhespecificity ofa physician's diagnosis of the rickettsial diseases appears to be high (i.e., few cases ofother diseases are misdiagnosed as either spottedfever or murine typhus), the sensitivity of the physician's diagnosis may be lower (i.e., many cases of these diseases presenting as fevers of unknown origin are not diagnosed as rickettsial disease). These findings may also be important for other countries of the region. Spotted fever and murine typhus are two known rickettsial diseases endemic in Israel (1,2) that cause. a febrile disease in man, often followed by a rash several days later. The rash of spotted fever resembles the rash of Rocky Mountain spotted fever which involves the palms, soles, and torso (1). In cases of murine typhus, the rash initially appears on the torso and spreads centrifugally, but rarely involves the palms and soles (3). The spotted fever group of rickettsiae, of which spotted fever occurring in Israel is an example, is believed to be maintained in nature by dogs, small wild animals, and ticks; in addition, the vector trans- mitting the agent to man is an infected tick (4). A tache noire or eschar has been associated with the bite of the tick in many cases of spotted fever caused by Rickettsia conori occurring in countries of the Mediterranean basin (5). This lesion is rarely, if ever, reported in cases of spotted fever occurring in Israel. Rickettsial strains of the spotted fever group isolated in Israel have been found to be antigenically different from strains of R. conori (5). Rickettsia typhi, the agent of murine typhus, is usually transmitted to man from its rodent reservoir by the rat flea, Xenopsylla cheopis (6). 1 Ministry of Health, Negev District, 84170 Beersheba, Israel, and University Center for Health Sciences, Ben Gurion University of the Negev, Beersheba, Israel (WHO Collaborating Centre for Integrated Health Services and Manpower Development). Requests for reprints should be sent to this author. 2 Epidemiology and Health Evaluation Unit, University Center for Health Sciences, Ben Gurion University of the Negev, Beersheba, Israel. 3 Department of Pediatrics, Soroka Medical Center and Uni- versity Center for Health Sciences, Ben Gurion University of the Negev, Beersheba, Israel. 4 Israel Institute for Biological Research, Nes Ziona, Israel. Reporting of all human cases of rickettsial diseases to the District Health Offices of the Ministry of Health is required by law. During 1981, the Beersheba District Health Offices of the Negev desert region of Israel received notifications of 52 cases of spotted fever occurring in a population of approximately 250 000. The cases were initially diagnosed on clinical signs and symptoms including the presence of a rash, irrespective of laboratory findings. However, at least 5 of these 52 reported cases were retrospectively diag- nosed as murine typhus (7). The present report describes some of the epidemiological characteristics of these cases of rickettsial diseases in the Beersheba district in 1981. MATERIALS AND METHODS A follow-up interview, some 2-3 weeks after the onset of illness, was attempted by public health nurses working in family health clinics of the Ministry of Health in all cases initially reported as rickettsial disease to the District Health Office. During the interview, the convalescent patients (or family members) were asked about the types of animals the patient came into contact with and whether the patient had recently been bitten by a tick, i.e., in the 2-3 week period prior to the onset of disease. Persons diagnosed with rickettsial disease, who were initially seen in hospital but not necessarily hos- pitalized, were asked to return to the hospital 2-3 weeks after the onset of illness for the drawing of con- valescent serum. These sera were examined for specific IgM antibodies to the spotted fever group of 4403 -301- E. M. GROSS ET AL. rickettsiae and to Rickettsia typhi at a 1:40 dilution by the immunofluorescent antibody method previously described (7). For purposes of this paper, the demonstration of either of these IgM antibodies was considered to be diagnostic of these diseases. If IgM antibodies to both antigens were demonstrated, one could not be sure of which rickettsial agent was responsible for the patient's most recent infection. RESULTS Of the 52 rickettsial cases reported to the Ministry of Health, 43 had follow-up interviews and conva- lescent sera were available from 23 of them. Fifteen of the sera showed IgM antibodies to the spotted fever group of reckettsiae alone, 5 to Rickettsia typhi alone, and 3 to both antigens (7). Where serum was available from a convalescent patient, in no case was there a failure to demonstrate IgM antibodies to one or both of these rickettsial antigens. Age and sex The ages of the 52 reported cases of rickettsial disease ranged from 1 to 81 years (mean 21.6 years, median 12 years), as shown in Fig. 1. The ages of the serologically proven spotted fever cases ranged from 1 to 79 years (mean, 29.9 years; median, 29 years), and the ages of the serologically proven cases of murine typhus ranged from 1 to 54 years (mean, 13.0 years; median, 3 years). The 3 persons with antibodies to both rickettsial antigens were 4, 9 and 81 years old. There were 28 males and 24 females who were reported ill with rickettsial disease; 8 males and 7 females had serologically proven spotted fever, 2 males and 3 females had serologically proven murine typhus, and 1 male and 2 females had antibodies to both antigens. Seasonal influence and places of residence Cases of rickettsial disease were reported from May to November inclusive. Most of them were diagnosed during the summer months of July, August and September, as shown in Fig. 2. These cases occurred from among the residents of Beersheba (population of 125 000), four smaller towns (populations ranging up to 30 000), and six agricultural settlements, and from among the Bedouin, some of whom are still semi-nomadic. Serologically confirmed cases of spotted fever occurred in Beersheba, two of the towns, and three of the agricultural settlements. Serologically confirmed cases of murine typhus occurred in Beersheba, one of the towns (where also a serologically confirmed case 25 r 20 15 E z 10 5 -I 71 E1 IgM antibodies demonstrated to: | Spotted fever group of rickettsiae Rickettsia typhi Both groups Convalescent serum unavailable r--- Patient hospitalized 11111rr- 1111 1 " 10 20 30 40 50 60 70 80 90 Age group ( years ) Fig. 1. Age groups and serological findings in 52 persons reported ill with rickettsial disease, Negev region of Israel, 1981. of spotted fever was reported), and among the Bedouin. Persons demonstrating antibodies to both antigens *were reported from Beersheba, one of the towns (where no serologically confirmed case of spotted fever or murine typhus was diagnosed), and among the Bedouin. One town and three agricultural settlements re- ported clinically diagnosed, but serologically un- confirmed, rickettsial disease from patients among whom convalescent sera were not obtained. In general, the reports indicated that rickettsial disease was occurring in individuals residing in the north- western area, the most densely populated section of the region. Hospitalization Of the 52 persons reported with rickettsial disease, 25 were and 26 were not hospitalized, and in 1 this was unknown. Of the 15 serologically proven cases of spotted fever, 7 were hospitalized; and of the 5 sero- logically proven cases of murine typhus, 4 were hospi- talized. Of the 3 cases demonstrating antibodies to both antigens, 1 was hospitalized. Of the 20 cases ,from whom convalescent sera were not available, 13 were hospitalized. Ttrrn 302 SPOTTED FEVER AND MURINE TYPHUS IN ISRAEL 15 14 13 12 11 V 3 E z 10 9 8 7 6 5 4 3 2 0 11 II IV V VI VIl Vill lX Month diagnosed x xl xiI IgM antibodies demonstrated to: Spotted fever group of rickettsiae E Rickettsia typhi groups [] Fig. 2. Month of onset of 52 reported cases of rickettsial disease, Negev region of !srael, 1981. Clustering of cases in families There were five families who reported 2 or more cases of rickettsiosis in 1981. One family had 3 persons hospitalized in July; serum from one of them demonstrated antibodies to the spotted fever group of rickettsiae. A second family had 3 persons hospi- talized in August and September; the serum from one of them also showed antibodies to the spotted fever group of rickettsiae. A third family had 4 members ill in September and October and a fourth family had 2 persons ill in September, but they were not hospita- lized and no convalescent serum was available. A fifth family had 2 persons ill in September; one of them was hospitalized and convalescent sera from both demonstrated antibodies to the spotted fever group of rickettsiae. History of tick bites Eight out of 37 persons for whom the information, was available either recalled tick bites occurring with- in a two-week period prior to the onset of disease, or family members gave this information during the interview. Four of these 8 persons demonstrated IgM antibodies to the spotted fever group of rickettsiae in their convalescent sera; no such sera were available from the remaining 4 patients who recalled a tick bite. In none of the serologically confirmed cases of spotted fever was there a tache noire identified at the site of the tick bite. However, in 2 serologically un- confirmed cases of spotted fever, a possible tache noire was noticed. Contact with mammals Forty patients or their families gave information about animal contact and/or ownership, which is summarized in Table 1. Thirty-five admitted contact with dogs, 22 with cats, 10 with sheep, 11 with rodents, 2 with horses, 1 with a cow, and 2 denied any animal contact or ownership. Of the 15 proven cases of spotted fever, 13 admitted contact with dogs, 11 with cats, 6 with rodents, 4 with sheep, 2 with horses, and 1 denied any animal contact. Of the 5 proven cases of murine typhus, 4 admitted contact with dogs, 1 with rodents, and 1 denied any animal ownership or contact. DISCUSSION The reported incidence of rickettsial disease caused by Rickettsia typhi or the spotted fever group of rickettsiae in the Negev for 1981 was 20.8 per 100 000 population. In order to obtain a preliminary estimate of the incidence rate for each disease separately, based on the available data, it may be assumed that the cases who provided convalescent sera were rep- resentative of all the reported rickettsial disease in the region. On this basis, at least 65% (15 of 23) of all reported rickettsial disease in the Negev is spotted fever and at least 22% is murine typhus. This leads to an estimate of an annual incidence rate, for spotted fever, of at least 13.6 per 100 000 population and, for murine typhus, of at least 4.5 per 100 000 population. If the demonstration of IgM antibodies to rickett- siae in the sera is diagnostic of recent rickettsial infec- tion, then, based on the 1981 data, the following may be tentatively concluded. The specificity of the phys- icians' diagnosis of rickettsial disease on the basis of clinical signs including a "typical" rash appears to be high, i.e., there are few cases, of diseases that are not rickettsial, which are misdiagnosed as rickettsial disease. However, there is the inability in some in- stances to differentiate on clinical grounds, including a rash, between spotted fever and murine typhus. This failure to differentiate between these two may indi- cate that the rashes are not pathognomonic or that physicians may have a low index of suspicion for murine typhus in the region and may classify all 11.. 111111 al l 11111.1 .1111] 11111 lm 111111 303 E. M. GROSS ET AL. Table 1. Known animal contacts within 2-3 weeks prior to the onset of illness in persons reported to be suffering from rickettsial disease, Negev Region of Israel, 1981 Contact with Serological situation No. Contact Information of cases Dog Cat Cow Sheep Rodent Horse denied unavailable 4a x x 4b x x x x IgM antibodies to the spotted fever 2c x group of rickettsiae 2 x x x 1 x 1 x IgM antibodies to 3 x Rickettsia typhi 1 x x 1 x 1 x x IgM antibodies to spotted fever 1 x x group and Rickettsia typhi 1 x 6c x 5 x x 2b x x x x Convalescent sera unavailable 2 x 1 x x x x 1 x x x 1 x x x 1 1d x a Includes 2 members of the same family, with different history of animal exposure. b Includes 3 members of the same family, 1 with convalescent sera available. ' Includes 3 members of the same family, 1 with convalescent sera available. d Includes 2 families of 4 and 2 persons, sera unavailable. rickettsial cases as spotted fever. The sensitivity of the diagnosis of rickettsial diseases by physicians on the basis of clinical signs and symptoms may be lower, i.e., there may be a cer- tain number of cases of rickettsial diseases that are not diagnosed as such. Our data do not allow us to give an estimate of this number. However, recently two cases of murine typhus have been diagnosed in Soroka Medical Center, Beersheba, in patients with- out rashes, who were admitted with fever of unknown origin. In addition, a recent retrospective survey of 25 persons admitted to the same hospital for fevers of unknown origin revealed IgM antibodies to the spot- ted fever group of rickettsiae or to Rickettsia typhi in the convalescent sera of 3 of these patients (Gross, E.M. & Goldwasser, R.A., unpublished data). It thus appears that the number of cases of rickett- sial disease occurring in the Negev are under- estimated, because many cases either are not recognized or are not reported. The notification of 52 cases during 1981, compared to reports of 6, 24 and 12 cases of rickettsiosis occurring in the district from 1978 to 1980 inclusive (Ministry of Health, unpub- lished data), coincided with an increased interest by the Beersheba District Health Office in the rickettsial diseases. Therefore, one may postulate that increased interest in these diseases may have caused an increase in notifications or an increase in the diagnosis of these diseases by physicians. On the other hand, it is pos- sible that there was a real increase in incidence of rickettsial disease. The finding of IgG antibodies to the spotted fever group of rickettsiae or to Rickettsia typhi in 28% (108 of 389) of a random sample of women delivering in Soroka Medical Center (9) also suggests that many cases of these diseases are not recognized, or not reported, or are subclinical. The fact that most cases of rickettsial disease occurred in the Negev during the summer months is in agreement with the findings in a review of 63 cases of presumed spotted fever seen in Soroka Medical Center from 1972 to 1978 inclusive (10). This seasonal appearance may be due to: (1) an increase in the prevalence of the rickettsial vectors or reservoirs; or (2) increased activity of the reservoir or vector during the warm, dry months; or (3) increased outdoor activities by man, with a greater chance of contact with an infectious vector. The rickettsiae have a wide distribution in the Negev. The serological evidence of rickettsial anti- bodies (spotted fever group or murine typhus) in the 304 SPOTTED FEVER AND MURINE TYPHUS IN ISRAEL convalescent sera of residents of a number of towns and agricultural settlements and among the Bedouin testifies to this fact. However, rickettsial disease was not reported from among the residents of two towns and a dozen or more scattered agricultural settlements in the southern area of the region. To try to clarify this localization of the rickettsial disease opens some interesting avenues for further investigation. The role of rodents as the reservoir of murine typhus and the rat flea as the most probable vector of the disease is well established (6). The reservoir of the spotted fever group, the various tick vectors, and possible reservoirs are less well understood. In Israel, Rhipicephalus sanguineus ticks taken from dogs and hedgehogs have been found to contain the spotted fever group of rickettsiae (5). However, in various parts of the world other species of ticks have also been found to contain the same group of rickettsiae (11,12), possibly indicating that other species of ticks in the south of Israel may be vectors of the disease. Possible animals necessary for the maintenance of the spotted fever group of rickettsiae in Israel, in addition to ticks, may be dogs, mice, Meriones, hedgehogs and rats inasmuch as there is some sero- logical evidence of exposure in these animals (13). However, the finding of rickettsial antibodies of the spotted fever group in any or all of these animals does not prove that they are reservoirs for the disease; they may be only incidentally infected by ticks feeding on them. Ruminants may also play a role in the mainten- ance of spotted fever in Israel, as higher rates of sero- positivity to the spotted fever group have been found in persons living in proximity to them (Goldwasser, R. A. & Rosner, M., unpublished). The clustering of cases of proven spotted fever in families, within a one or two month period, seems to indicate that there are areas where the infected tick density is high enough to cause simultaneous infec- tion of several family members and/or that infected ticks can feed off several persons in rapid succession. The recall of tick bites by only 4 out of 15 persons with serologically proven spotted fever indicates that ticks may attack man and fall off unnoticed. To incrimi- nate vectors other than ticks in the transmission of spotted fever or direct person-to-person transmission under field conditions is open to speculation. The failure of 4 out of 5 patients or family members of cases of murine typhus to be aware of rodents in the patients' environment may indicate the inability of patients or family members accurately to describe all animal contacts. Thus, our data associating a variety of animal contacts with proven cases of spotted fever are only suggestive of a role the various animals may play in the disease. The reporting of 50o (26 out of 52) of all rickettsial disease in children aged 9 years or less may or may not be a true reflection of the actual age distribution of the attack rates of these diseases. If these data do reflect the actual age distribution, it is possible that children are more likely to nmaintain close contact with animals and rickettsial vectors, and/or are less likely to notice or avoid rickettsial vectors such as ticks and fleas. Data indicating, in general, a failure to show increased seropositivity rates to spotted fever and murine typhus in women during the childbearing years (9) might be considered, at least tentatively, as supporting evidence for this conjecture. On the other hand, the reporting of most cases in the young may reflect a high degree of concern by parents in the health of their children, i.e., their children are taken to physicians on account of a rickettsial illness while persons in the older age groups may be less likely to seek medical aid. The hospitalization of all reported cases of rickett- siosis in persons aged 50 years or greater may indicate that the diseases are more severe in persons of this age group. The fact that 4 out of 5 cases of murine typhus were hospitalized compared with 7 out of 15 cases of spotted fever may indicate that the former is a more severe disease. The reported hospitalization of at least 25 persons with diagnosed rickettsial disease in the Negev during 1981, plus an undetermined (but apparently signifi- cant) number of persons ill with retrospectively con- firmed rickettsial disease indicate the importance of these diseases in the region. Accurate, rapid diagnosis of rickettsial disease in persons not displaying a rash, so that proper treatment can be instituted, would aid control measures. Further investigation into the epi- demiology of spotted fever with a view to its control, the differentiation clinically between murine typhus and spotted fever, and possibly more complete report- ing of these diseases are important so that specific measures can be carried out to prevent additional cases. The failure to differentiate between clinical cases of spotted fever and murine typhus and the failure to recognize many cases of these diseases may also be important for other countries in the region. RESUME FIEVRE POURPREE ET TYPHUS MURIN DANS LE DESERT ISRAELIEN DU NEGUEV, 1981 En 1981, on a signale au Bureau sanitaire du district de Beersheba, qui releve du Ministere de la Sante, 52 cas de fievre pourpree survenus dans la region meridionale deser- tique d'Israel et tous accompagnes d'une forte temperature et d'une eruption. Mais il s'est avere retrospectivement que cinq des 23 sujets chez lesquels on avait preleve du serum 305 306 E. M. GROSS ET AL. pendant la convalescence presentaient des anticorps lgM a l'egard de Rickettsia typhi, agent responsable du typhus murin, et non pas a l'egard du groupe de rickettsies respon- sable de la fievre pourpr6e. Dans tous les cas oui l'on soupqonnait une rickettsiose, il a et possible de mettre en evidence des anticorps IgM a l'egard d'un ou des deux types de rickettsies (lorsqu'on possedait les serums). II semble donc que le diagnostic medical de rickettsiose fonde sur les signes et sympt8mes cliniques, y compris l'eruption, soit tres specifique, c'est-a-dire qu'il y a peu de cas de maladie n'etant pas due a des rickettsies oii l'on pose un diagnostic errone de fievre pourpr&e ou de typhus murin. Neanmoins, il est parfois impossible dans certains cas de distinguer entre les cas cliniques de fievre pourpree et de typhus murin- ce qui pourrait indiquer que les eruptions ne sont pas patho- gnomoniques de l'une ou l'autre maladie ou bien que les medecins soupqonnent rarement la presence de typhus murin dans la region et risquent de classer tous les cas de rickettsiose presentant une &ruption comme des cas de fievre pourpree. La sensibilite du diagnostic de fievre pourpree ou de typhus murin fonde sur les signes et sympt6mes cliniques est plus faible, c'est-a-dire que certains cas de ces maladies ne sont pas diagnostiques comme tels. Reemment, une en- quete retrospective faite avec des serums de convalescents ayant presente une fievre d'origine inconnue a permis de mettre en evidence, dans trois des 25 serums examines, des anticorps IgM a 1'egard des rickettsies responsables de la fievre pourpree ou du typhus murin. Dans la serie de 52 cas en question, il a et signale que la plupart des sujets etaient tombes malades en te. Environ la moitie des cas sont survenus chez des enfants de 9 ans ou moins. L'enquete n'a pas fait apparaitre d'association pre- cise avec des animaux. Huit des 37 personnes pour lesquelles on disposait de renseignements ont evoque une piqure de tique avant l'apparition de la maladie. On possedait des serums pour quatre de ces huit personnes et, dans les quatre cas, la presence d'anticorps IgM a 1'egard du groupe de rickettsies responsable de la fievre pourpree a pu etre mise en evidence. II convient de comparer les 52 cas de rickettsiose signales dans la region en 1981 avec les 6,24 et 12 cas de fievre pour- pr&e notifies leF annees precedentes. Cette augmentation a coincide avec un renouveau d'interet pour les rickettsioses manifeste par les agents du Bureau de sante du district. Le diagnostic retrospectif de typhus murin ou de fievre pour- pree qui pourrait expliquer certaines fievres d'origine in- connue chez des sujets hospitalises donne a penser que de nombreux cas de ces maladies ne sont pas signales ou bien ne sont pas reconnus comme tels dans la region. Cette inca- pacite A distinguer entre les cas cliniques de fievre pourpr6e et de typhus murin et A reconnai-tre ces maladies pourrait etre importante pour d'autres pays de la Region. REFERENCES 1. VALERO, A. Rocky Mountain spotted fever in Palestine. Harefuah, 36: 99-101 (1949) (Hebrew with English summary). 2. KLIGER, I. J. & CAMAROFF R. An epidemic outbreak of murine typhus in a labour group in an inland village in Palestine. Transactions of the Royal Society of Med- icine and Hygiene 30: 363-368 (1936). 3. WISSEMAN, C. L. Murine or flea-borne typhus fever. In: Beeson, P. B. et al., ed., Cecil's Textbook ofMedicine, 15th ed., Philadelphia, W. B. Saunders, 1979, pp. 320-321. 4. WOODWARD, T. E. Other tick-borne rickettsial diseases. In: Isselbacher, K. J. et al., ed., Harrison's Principles ofInternal Medicine, 9th ed., New York, McGraw-Hill, 1980, p. 753. 5. GOLDWASSER, R. A. ET AL. The isolation of strains of rickettsiae of the spotted fever group in Israel and their differentiation from other members of the group by immunofluorescence methods. Scandinavianjournal of infectious diseases, 6: 53-62 (1974). 6. TRAUB, R. ET AL. The ecology of murine typhus - a critical review. Tropical diseases bulletin, 75: 237-317 (1978). 7. GROSS, E. M. ET AL. Resurgence of Mediterranean spotted fever. Lancet, 2: 1107 (1982). 8. GOLDWASSER, R. A. & SHEPARD C. C. Fluorescent antibody methods in the differentiation of murine and epidemic typhus sera: specificity changes resulting from previous immunization. Journal of immunological methods, 82: 373-386 (1959). 9. GROSS, E. M. ET AL. Rickettsial antibody prevalence in southern Israel: IgG antibodies to Coxiella burnetii, Rickettsia typhi, and spotted fever group rickettsiae among urban and rural-dwelling and Bedouin women. American journal of tropical medicine and hygiene, 32: 1387-1391 (1983). 10. SHULCHYNSKA, H. ET AL. Spotted fever in the Negev. Harefuah, 102: 317-319 (1982) (Hebrew with English summary). 11. ROBERTSON, R. D. ET AL. Tick-borne rickettsiae of the spotted fever group in West Pakistan: isolation of strains from ticks in different habitats. American journal of epidemiology, 92: 382-384 (1970). 12. SEXTON, D. J. ET AL. Rocky Mountain spotted fever in Mississippi: survey for spotted fever antibodies in dogs and for spotted fever group rickettsiae in dog ticks. American journal of epidemiology, 103: 192-197 (1976). 13. GOLDWASSER, R. A. ET AL. Laboratory and epi- demiologic studies of rickettsial spotted fever in Israel. In: Frontiers of Internal Medicine, 12th International Congress of Internal Medicine, Tel Aviv, 1974, pp. 270-275.