In agreement with infectious disease practitioners and the World Health Organization (WHO) Collaborating Centre for Rickettsial Diseases, we chose to research murine typhus only in cases of fever of unknown origin lasting >7 days. In accordance with WHO Collaborating Centre procedures, a case was considered confirmed by seroconversion or by a >4-fold increase in antibody response against R. typhi from acute- and convalescent-phase serum samples; by a positive indirect immunofluorescence test for typhus-group rickettsiae confirmed by Western blot; or by positive R. typhi–specific real-time quantitative PCR. We used Western blot to formally distinguish murine typhus from epidemic typhus. Routine analysis (i.e., blood cell count) was performed in Reunion. Murine typhus diagnostic tests were performed by the WHO Collaborating Centre for Rickettsial Diseases (Marseille, France) (7).

We collected and analyzed data for each case by reviewing clinical records and administering an epidemiologic questionnaire by phone to each patient. The following data were collected: symptoms, signs, and complications during illness; results of routine laboratory analyses; treatment and outcome; habitat and environmental characteristics; presence of pets, livestock, or rodents; contact with fleas; daily activities in the 15 days before symptom onset; and travel during the previous 3 months.

During January 2011–January 2013, a total of 8 autochthonous murine typhus cases were confirmed: 5 patients in 2012, 1 in 2013, and 2 who tested positive for retrospective screening of archived serum samples from 2011 (Figures 1, 2; Table). Seven cases occurred during the Southern Hemisphere summer. All the patients lived in private houses in the western and southern parts of the island in periurban areas. Patients’ average age was 46 years (range 21–66 years), and the male:female ratio was 1:1.

For all patients, fever lasted an average of 14.3 days (range 10–21 days). Six patients were hospitalized. The 2 other main symptoms were arthromyalgia (7 patients) and headaches (6 patients). Only 4 patients had a maculopapular rash, but no inoculation eschar was found. In addition, pharyngitis developed in 4 patients. In contrast, gastrointestinal symptoms and ophthalmologic signs, such as uveitis, developed in 2 patients each. One patient had confusion, and 1 had prostration.

Six patients had elevated liver enzymes (reference values aspartate aminotransferase and alanine aminotransferase >33 UI/L; lactate dehydrogenase >480 UI/L). Early lymphopenia (lymphocytes <1,000 cells/mL) occurred in 5 patients, and thrombocytopenia (platelets <150 000/mL) in 4. Renal dysfunction did not develop in any of the 8 patients.

Six patients were treated with doxycycline (2 patients in the acute phase for whom fever decreased during the following 24 hours; 4 others despite previous spontaneous resolution of fever). One of the 2 remaining patients, a pregnant woman, was treated with spiramycin without any complications to the baby during 6 months of follow-up. The other patient did not receive any treatment. All patients recovered completely, although they had weakness lasting at least 1 month.

None of the patients had traveled overseas during the previous 3 months, and no other cases had been identified nearby. Only 1 patient reported having recent insect bites and pruritus. No patient recalled contact with rats in his/her living environment before disease onset.

Five patients completed the investigation questionnaire. Reported risk factors included close contact with domestic pets (4 patients); presence of livestock in the surroundings (4 patients); presence of wild fauna (rats, tenrecs, dogs, cats) in the surroundings (4 patients); rat extermination (3 patients); outdoor activities, such as jogging, picnicking, walking, gardening (3 patients); or house cleaning (2 patients).

Serologic and molecular evidence is sufficient to indicate that autochthonous transmission of murine typhus exists in Reunion. The real impact and the origin (always present but only recently identified vs. recent importation by navigation trade from Asia or from Madagascar) of murine typhus remain unknown. Murine typhus is common in tropical areas but is often not distinguished from other more prevalent febrile bacterial infections, such as scrub typhus, spotted fever rickettsioses, and leptospirosis. However, murine typhus recently has been shown to be a major cause of fever of unknown origin in Indonesia and Laos (8,9). In Reunion, despite a systematic biologic investigation (dengue, chikungunya, and leptospirosis) of dengue-like syndrome, many fevers still remain of unknown origin (10,11). Our results support the hypothesis that murine typhus could be a major cause of fever in Reunion.

Like leptospirosis, murine typhus seems to peak seasonally during the summer (12). As previously described in the tropics, climatic conditions during the hot and wet seasons lead to the proliferation of rats and so increase human exposure to murine typhus (4,13). Because most of the 8 patients had contact with pets, the role of fleas that infest pets (e.g., Ctenocephalides felis) in the transmission of murine typhus needs to be clarified, as does the role of X. cheopis, the rat flea (14).

In conclusion, murine typhus is an emerging disease in Reunion. It is possibly underdiagnosed and has the potential to cause major illness (9). This disease should be considered locally when fevers of unknown origin are investigated, as well as in travelers with fever returning from the island. Specific studies are needed to assess the effect of murine typhus on public health and to describe the epidemiology of the disease. The best way to prevent murine typhus is to minimize exposure to ectoparasite vectors by limiting contact with rodents and fleas. Such measures include keeping a well-maintained yard, minimizing waste, protecting pets from fleas, and using mechanical or chemical protection in case of at-risk activities.