The pathogenic role of Rickettsia slovaca was first demonstrated in 1997 in a patient who presented with a single inoculation lesion of the scalp and enlarged cervical lymph nodes after receiving a bite from a Dermacentor tick. Subsequently, we evaluated the occurrence of R. slovaca infections among patients living in France and Hungary who presented with these symptoms. R. slovaca infections were confirmed by polymerase chain reaction (PCR) in 17 of 67 enrolled patients. Infections were most likely to occur in patients aged <10 years and in patients who were bitten during the colder months of the year. The median duration of incubation for the disease was 7 days. Fever was present in only 2 patients, and only 1 patient developed a rash. Sequelae included persistent asthenia (3 cases) and localized alopecia (4 cases). Immunofluorescence and/or Western blot analysis detected antibodies in 50% of tested patients. Three Dermacentor ticks obtained from patients revealed R. slovaca by PCR.
In recent years, several new tickborne diseases have been identified, including those caused by the 3 species of Borrelia burgdorferi sensu lato , by Borrelia lonestari , by Ehrlichia chaffeensis , by Ehrlichia ewingii , and by the agent of human granulocytic ehrlichiosis . Of the rickettsial diseases, only 5 that are transmitted by ticks were known before 1991 . However, since this time, 7 new diseases have been reported, including Astrakhan fever , Flinders Island spotted fever , African tick bite fever , Japanese spotted fever , and, in Europe, infections caused by Rickettsia mongolotimonae , Rickettsia slovaca , and Rickettsia helvetica . The latter 2 diseases are diagnostically troublesome, because they have been reported in patients with no cutaneous rash, which is a characteristic clinical feature of other rickettsial diseases.
R. slovaca was first isolated in 1968 from a Dermacentor marginatus tick in Slovakia , but it was not until 1980 that the first suspected case of infection was reported. This case occurred in a patient who presented with hemiparesis, meningitis, and transient skin inflammation at the site of the tick bite . However, the only evidence of infection was serological; thus, because wide cross-reactions occur among tickborne rickettsioses, definite species identification was not made . Shortly after the first description of Lyme borreliosis was reported in Hungary, a center for tickborne diseases was opened, which has been visited by thousands of tick-bitten patients every year since it opened. In 1987, a patient presented with an unusual crustaceous scalp reaction after receiving a tick bite. Enlarged lymph nodes that drained the site of the tick bite were also noted. Since 1987, very similar cases have been seen, presenting a distinct picture of a hitherto unknown infection. Besides the local reaction, the most pronounced symptom was enlarged lymph nodes, which led to the infection being named “tick-borne lymphadenopathy” (TIBOLA) [16, 17]. In 1997, we reported the first proven case of R. slovaca infection in a patient who had been bitten by a D. marginatus tick. The patient developed scalp erythema, cervical lymphadenopathies, and postinfectious persistent asthenia . In the present study, performed from 1996 through 2000, we collected data from French and Hungarian patients who developed an inoculation lesion and enlarged draining lymph nodes after they received tick bites on the scalp, to investigate the relative importance of R. slovaca in cases in which the aforementioned clinical symptoms were present.
Patients and Methods
Patient.. All French and Hungarian patients referred to our center (Unité des Rickettsies, Université de la Méditerranée, Marseille) from January 1996 through April 2000 with a suspected rickettsial infection characterized by a tick bite to the scalp, an inoculation lesion, and enlarged draining lymph nodes were included in the study. For each patient, epidemiological and clinical data were collected by the consulting physician by use of a standardized questionnaire. The questionnaire included inquiries about the site of the tick bite, the geographical location where the bite had occurred, contact with animals, underlying diseases, fever, headache, the presence and location of inoculation eschar, cutaneous local reaction, regional adenopathy, and outcome. Leukocyte and platelet counts were recorded, as were aspartate aminotransferase and alanine aminotransferase levels. Appropriate informed consent was obtained from all patients.
Serologic test.. For each patient, a serum sample was obtained early after the onset of symptoms; when available, a late-phase sample was also obtained. IgG and IgM antibody estimation was performed by use of the microimmunofluorescence (MIF) assay, the reference method for rickettsial disease, as reported elsewhere ; both Rickettsia conorii and R. slovaca antigens were used. Titers of 1 : 64 for IgG and 1 : 16 for IgM, as well as seroconversion or a 4-fold increase in titer to 1 : 32, were considered evidence of recent infection by a Rickettsia species. Western blotting procedures were performed as described elsewhere . Cross-adsorption by MIF was performed when patients exhibited antibody titers of >1 : 64, as described elsewhere . R. slovaca and R. conorii MIF testing and Western blot analysis were performed on the resultant supernatants.
Culture and PC.. When a patient had retained the tick that had bitten them, the species of the tick was determined. Each tick was decontaminated by a 5-min immersion in a solution of 70% ethanol and 0.2% iodine, followed by a 5-min immersion in sterile distilled water. Ticks were subsequently cut longitudinally into 2 equal parts, 1 of which was used for culture and 1 of which was used for DNA extraction. For culture, heparinized blood samples, lymph node aspirates, ticks, and skin biopsy specimens were inoculated on human embryonic lung cells and cultivated as described elsewhere . DNA was extracted from EDTA-blood specimens, ground skin-biopsy specimens, lymph node biopsy specimens, or ground ticks by use of the FastPrep extraction system (BIO 101), according to the manufacturer's recommendations. These extracts were used as templates in a regular PCR assay, with primers derived from the rOmpA-encoding gene . Sequencing of PCR products was performed as described elsewhere .
In an effort to increase the sensitivity of DNA detection, we used a suicide PCR assay  to test, when there was remaining material, the acute-phase serum sample, skin biopsy specimens, and lymph node aspirates that had negative results of initial PCR. Suicide PCR is a nested PCR in which both the target DNA fragment and the primers are used only once. Primers used for suicide PCR are described in table 1. All PCR products with positive results were sequenced with the primers used for PCR, as described above, for precise identification of the infecting Rickettsia species. A control sample was introduced for every 5 patients' samples. Negative controls were skin biopsy or serum samples obtained from patients with a noninfectious disease, and positive controls were skin biopsy or serum samples obtained from patients with proven R. conorii infection.
Diagnostic criteri.. A definite diagnosis was considered for patients whose clinical samples yielded an isolate or a positive PCR result.
Statistical testin.. Fisher's exact test was used to compare proportions and the Wilcoxon-Mann-Whitney test was used to compare median values. Observed differences were considered significant when P was <.05 in 2-tailed tests.
Sixty-seven patients who developed a scalp lesion and occipital and/or cervical lymph node enlargement after receiving a tick bite were enrolled in this study. Thirty-six patients were from France and 31 were from Hungary. Seventeen patients (25%), 9 of whom were from France and 8 of whom were from Hungary, had definitive evidence of R. slovaca infection (table 2). In France, the patients had been bitten during the colder months of the year: 2 cases were recorded in October–January and 7 were recorded in February–May. In Hungary, most patients were seen in March and April. The length of time that the ticks were attached was not available. The incubation time of the disease was available for 13 patients and ranged from 1 to 55 days (median, 7 days). Female patients were more frequently affected (10 of 17 patients were female) than were male patients. The median age of female patients was 15 years (range, 5–58 years). Three of the 9 French patients and 4 of the 8 Hungarian patients were <10 years old.
The appearance of inoculation skin lesions varied from crusted scalp lesions, which ranged in diameter from 1 to 30 mm (figure 1), to erythema migrans–like erythemas, which ranged in diameter from 1 to 8 cm. Both single and multiple enlarged lymph nodes were seen, and they were painful in 10 cases (59%). Fever was present in only 2 patients. A maculopapular rash was observed in 1 patient. Treatment was available for 14 patients; 10 patients were treated with doxycycline and 3 were treated with azithromycin for 10–30 days. One patient received no antibiotics. Ten of the 17 patients fully recovered. Four patients (all of whom had received antibiotics) had persistent alopecia at the site of the bite after healing of the local reaction (figure 1). At follow-up, 3 patients still had alopecia 1 year after the acute episode occurred, and, in 1 patient, alopecia persisted for 5 years after the infection. Although they were treated with doxycycline, 3 patients had persistent asthenia for 4, 7, and 12 weeks.
Three of the ticks that were removed from patients were identified; 2 ticks were identified as D. marginatus and 1 was identified as Dermacentor reticulatus. The epidemiological and clinical features of the 17 patients with proven infection were not statistically different from those of the 50 patients with no definitive evidence of R. slovaca infection (table 2).
The leukocyte count was within the normal range for all patients, except for 1 patient who presented with leukopenia (3.0 × 109 leukocytes/L). Two patients had thrombocytopenia (129 × 109 platelets/L for one and 112 × 109 platelets/L for the other). Elevated aspartate aminotransferase and alanine aminotransferase levels were observed in 2 patients, who had levels that were 1.6 and 2.1 times the highest value considered normal in our laboratory (i.e., 45 IU/L).
The results of the MIF assay were positive for 7 (41.1%) of 17 patients, but the findings were specific for R. slovaca for only 2 (11.8%) of 17 (table 3). The findings of Western blot analysis were specific for R. slovaca in 5 (35.7%) of 14 cases. When cross-adsorption was tested, it resulted in a specific diagnosis in 2 of 2 cases, including 1 case that had not been differentiated by MIF or Western blot assay. Skin biopsy and fine-needle lymph node aspirates were the most useful specimens for standard PCR; PCR analysis of these specimens identified R. slovaca infection in 12 (75%) of 16 R. slovaca–positive patients. Moreover, the 3 skin biopsy specimens with negative PCR results had been obtained either >3 weeks after the onset of symptoms, after 7 days of antibiotic therapy, or after fixation in Bouin fixative, which inhibits DNA polymerase. In 1 patient, the result of PCR of the EDTA-blood sample was positive. The results of suicide PCR were positive in 11 (64.7%)of the 17 early-phase serum samples tested. The sequence obtained from all PCR products matched that of R. slovaca. In all experiments, negative controls remained negative. Culture of 3 skin biopsy specimens and 4 fine-needle lymph node aspirates was attempted without success. R. slovaca was recovered from 1 tick by culture, but all 3 ticks were positive for R. slovaca by PCR.
We describe 17 patients infected with R. slovaca after receiving tick bites in France and Hungary who were observed for a 5-year period. This bacterium has been isolated from D. marginatus and D. reticulatus ticks from Slovakia , France, Armenia, Ukraine, Switzerland, Yugoslavia, Hungary, and Portugal . Although Dermacentor ticks are infected with Rickettsia sibirica in Siberia and western China and by Rickettsia rickettsii in America, R. slovaca is the only Rickettsia species found in these ticks in Europe. The prevalence of R. slovaca in D. marginatus ticks ranges from 1% to 17% . In Nagymaros (Börzsöny area), Hungary, where most of the Hungarian patients were infected, the rate of Rickettsia-harboring Dermacentor ticks was 21.1% . In studies from Spain, Dermacentor ticks were involved in 10%  to 11%  of tick bites to humans, and 30% of the Dermacentor ticks removed from patients were infected with a Rickettsia species . Therefore, although R. conorii has long been considered to be the only tick-transmitted Rickettsia species in several areas, such as central France  and northern and western Spain, R. slovaca may also be prevalent. During the period of our study, 81 cases of Mediterranean spotted fever, 2 cases of R. mongolotimonae infection, and 1 case of R. helvetica infection were also diagnosed in our laboratory. Therefore, documented R. slovaca infections represented 19% of the European tick-transmitted rickettsioses documented in our center during this period.
In Hungary, to the best of our knowledge, TIBOLA is the only recognized human rickettsial infection. TIBOLA seems to be more widely distributed geographically than are other European rickettsioses. At the Center for Tickborne Diseases in Budapest, data on patients with symptoms consistent with TIBOLA have been collected from Austria, Slovakia, the Czech Republic, Romania, Slovenia, Croatia, and Bulgaria. The clinical symptoms of TIBOLA include the presence of a scalp lesion and enlarged cervical lymph nodes after receiving a bite from a Dermacentor tick. In our study, symptoms were mostly mild, but a few patients presented with such sequelae as persistent asthenia for up to 3 months and persistent localized alopecia at the site of the tick bite. Antibiotic therapy cured the patients but did not prevent the development of sequelae. However, the spectrum of the disease is not fully known, because a case of meningoencephalitis has been reported from Slovakia , and also because we observed seroconversion to R. slovaca in a French patient who had neurological symptoms (authors' unpublished data).
In our study, children and women had a higher risk of infection due to R. slovaca than did men. Dermacentor ticks prefer to bite hairy animals, which may explain why they bite the scalps of humans. Children are more frequently bitten on the scalp, and longer and stronger hair is a predisposing factor for being bitten by Dermacentor ticks (A.L., unpublished data). This fact has been observed with regard to different Dermacentor species in the United States  and in France . This tendency may be the result of the questing height of Dermacentor ticks (i.e., 1–1.5 m, the position on vegetation where the ticks prefer to wait for a potential host) [29, 30].
In addition to their different geographic distributions and clinical presentations, TIBOLA and Mediterranean spotted fever exhibit specific seasonal distributions . In our study, cases of R. slovaca infection occurred mainly during the colder months of the year, whereas cases of R. conorii infection were associated with the summer. A study from Spain reported that Dermacentor ticks more frequently bite humans during the autumn and winter, whereas Rhipicephalus sanguineus ticks bite during the summer . A recent survey of soldiers in central France showed that, in April and May, 25% of the soldiers experienced tick bites; among the 319 ticks removed from patients, 305 were D. marginatus, 11 were D. reticulatus, and 3 were Ixodes ricinus . Therefore, because the identification of a tick as a Dermacentor species after a tick bite should be a reason to suspect R. slovaca infection, we emphasize the importance of preserving ticks for taxonomic identification.
We observed that standard serological testing methods were insensitive. This may be explained by the fact that TIBOLA is a localized disease, unlike R. conorii infection. DNA amplification by PCR was markedly more useful, especially when performed with a skin biopsy specimen or lymph node aspirate. Suicide PCR was very efficient, especially when performed with early-phase serum samples. The 50 patients who presented with similar symptoms, but for whom no clear diagnosis could be made, could have been infected either with R. slovaca that we failed to recognize or with other unidentified pathogens.
Our data suggest that R. slovaca may be a significant cause of tickborne disease in Europe. TIBOLA should be suspected in patients who have been bitten on the scalp by Dermacentor ticks, especially during the colder months from October to May. It is probable that other clinical forms of R. slovaca infection have yet to be described.
We thank B. Toga for identifying the ticks, and we thank R. J. Birtles for reviewing the English text.