During 2010–2014, we tested skin biopsy (7) and cutaneous swab samples from rickettsiosis inpatients and outpatients throughout France. These samples were received frozen or in transport media; when possible, serum samples were also collected and sent at room temperature. For patients with positive Rickettsia results, epidemiologic and clinical data were collected.

We extracted total genomic DNA from samples by using a QIAamp tissue kit (QIAGEN, Hilden, Germany). We screened samples for Rickettsia spp. by using a quantitative PCR assay selective for a 109-bp fragment of a hypothetical protein (8). For positive samples, PCR amplification and sequencing selective for the gltA and ompA genes were performed (8). Samples were cultured in human embryonic lung fibroblasts (9). All serum samples were tested by immunofluorescence assay for SFG rickettsial antigens and typhus group rickettsiae (10). Student t or χ² tests were performed by using Epi Info version 6.0 software (Centers for Disease Control and Prevention, Atlanta, GA, USA). Means were compared by using analysis of variance or the Kruskal-Wallis test, on the basis of results of the Bartlett test for inequality of population variances. Proportions were compared by using the Mantel-Haenszel χ² or Fisher exact tests when the expected value of a cell was <0.05. R. sibirica mongolitimonae seasonality was assessed by using the autocorrelation module of PASW software version 17.02 (http://www.spss.com.hk/statistics/). p <0.05 was considered significant.

We classified patients as definitively having a rickettsiosis if direct evidence of rickettsial infection was found on culture or molecular assays. Of 465 patients examined, 91 (20%) were infected with Rickettsia spp., most commonly R. africae (n = 36, 40%), followed by R. conorii (n = 21, 23%), R. sibirica mongolitimonae (n = 20, 22%), and R. slovaca (n = 14, 15%). Two cases of R. sibirica mongolitimonae infection in France have been reported (11,12).

For patients infected with R. sibirica mongolitimonae, median age ± SD (interquartile range) was 43 ± 21 (2–70) years, and most (12, 60%) were male (Technical Appendix Table 1). The most common Rickettsia species in France was R. sibirica mongolitimonae. Only 1 patient mentioned recent travel to Spain; all others denied recent travel. Five patients mentioned recent outdoor activities, 8 mentioned frequent contact with dogs, and 1 mentioned contact with horses. A tick bite or tick handling was reported by 6 patients. An autocorrelation analysis revealed significant seasonality for R. sibirica mongolitimonae cases (p<0.001). Significantly more cases occurred during spring (April–June) (11 cases, 55%; p = 0.006), followed by summer (July–September) (8 cases, 40%, p = 0.01). One case occurred in October and none in winter.

The symptoms at disease onset included fever for all patients (duration 4–14 days), myalgia (n = 11, 55%), and headache (n = 3, 15%). Generalized maculopapular rash and an inoculation eschar developed in all patients. One patient had 3 eschars (buttocks, right hand, breast). A rope-like lymphangitis from the eschar to the draining lymph node was detected in 7 (35%) patients. One patient was admitted to an intensive care unit. For all 5 patients for whom an initial laboratory examination was available, increased liver enzymes (alanine aminotransferase, aspartate aminotransferase) and thrombocytopenia were found; 2 patients had hypoproteinemia. Oral doxycycline (7–14 days) was given to 19 patients; pristinamycin (7 days) was given to 1 patient. All outcomes were successful.

An eschar swab sample was available for 13 patients (13), and a skin biopsy sample was available for 10; all samples were positive for R. sibirica mongolitimonae. An acute-phase serum sample was also available for 13 patients; results of serologic testing were positive for only 2 (15%). A convalescent-phase serum sample was available from 5 patients; results were positive for 4 (80%). A skin biopsy sample was also positive for R. sibirica mongolitimonae by culture.

Statistical comparison of the 4 rickettsioses (Table) showed that a recent travel history was more common among patients with R. africae infection (p<0.001). R. slovaca infection was associated with absence of fever or rash (p<0.001 for each). Multiple eschars were associated with R. africae infection (p<0.001). An eschar on the neck was a characteristic of infection with R. sibirica mongolitimonae (p = 0.002); on the scalp, R. slovaca (p<0.001); on the trunk, R. conorii (p = 0.05); and on the lower limbs, R. africae (p<0.001). For patients with rope-like lymphangitis, the probability of R. sibirica mongolitimonae infection was 100% (p<0.001). Cervical lymphadenitis was associated with R. slovaca (p<0.001), inguinal lymphadenitis with R. africae (p<0.001), and axillary lymphadenitis with R. sibirica mongolitimonae infection (p = 0.01).

R. sibirica mongolitimonae is considered a rare pathogen; only 30 cases of infection with this organism have been reported in Europe and Africa (Technical Appendix Table 2), of which 11 patients had lymphangitis, 27 inoculation eschars, and 18 a rash. In agreement with previous authors, we found that the most common signs of R. sibirica mongolitimonae infection were fever and rash. The addition of rope-like lymphangitis cases to those in the literature revealed that 17 (35%) of patients with R. sibirica mongolitimonae infection had this manifestation. Ramos et al. proposed that the term lymphangitis-associated rickettsiosis may be unwarranted for R. sibirica mongolitimonae infection because it is not found in all patients infected with this organism and because other rickettsioses produce lymphangitis (14). However, only R. sibirica mongolitimonae infection is associated with rope-like lymphangitis extending from the eschar to the draining lymph node; to our knowledge, only 1 case of mild, local, but not rope-like lymphangitis in a patient with R. africae infection has been described (15). In accordance with previous reports from France and Spain (3,14), we found that R. sibirica mongolitimonae infection was seasonal and that most cases occurred in the spring and summer.

Our strategy for diagnosing Rickettsia spp. infection on the basis of skin biopsy and cutaneous swab samples modified our knowledge of the epidemiology of SFG rickettsioses in France. We provide evidence that R. sibirica mongolitimonae infection is a frequent rickettsiosis, probably more frequent than R. conorii infection, which for decades has been considered the most common Rickettsia species in France.