Tropheryma whipplei in Fecal Samples from Children, Senegal Florence Fenollar, Jean-François Trape, Hubert Bassene, Cheikh Sokhna, and Didier Raoult We tested fecal samples from 150 healthy children 2–10 years of age who lived in rural Senegal and found the prevalence of Tropheryma whipplei was 44%. Unique geno- types were associated with this bacterium. Our indings suggest that T. whipplei is emerging as a highly prevalent pathogen in sub-Saharan Africa. T ropheryma whipplei is known mainly as the bacterial pathogen responsible for Whipple disease (1). Until re- cently, it was thought to be a rare bacterium typically caus- ing disease in white men (1). However, recent studies have shown 1%–11% prevalence of the bacterium in fecal sam- ples from the healthy general adult population in Europe (2,3). T. whipplei also was viable in a fecal sample from a patient with Whipple disease (4). In addition, T. whipplei DNA has been detected in sewage and is more prevalent in fecal samples of sewer workers (12%–26%) than in the general population, supporting this environment as a likely ecologic niche (2,3). T. whipplei was highly prevalent in fecal samples of children 2–4 years of age in France who have gastroenteritis but was not detected in a control group of children of the same age who did not have diarrhea (5). For these reasons, we speculated that if T. whipplei is transmitted through the fecal–oral route, it might be more prevalent in countries with poor sanitation, such as devel- oping countries. Because no information is available about T. whipplei and Whipple disease in Africa, we conducted a study to assess the prevalence of T. whipplei in fecal sam- ples from healthy children in Senegal, speciically in the villages of Dielmo and Ndiop (Figure 1). The Study In early April 2008, we sampled fecal specimens from 150 healthy children (79 girls) 2 months–10 years of age (mean 3.5 years ± 2.5 years) living in 2 villages in Senegal (Ndiop, 77 children; Dielmo, 73 children) (6). These vil- lages are included in the Dielmo project, initiated in 1990 for long-term investigations of host–parasite associations in the entire village population, which was enrolled in a longitudinal prospective study (6,7). At the beginning of the current study, parents or legal guardians of all children gave individual informed consent. The national ethics com- mittee of Senegal approved the project (6). Eight wells in the 2 villages (5 from Dielmo, 3 from Ndiop), which are the only sources of drinking water for the communities, also were sampled. After collection, each fecal specimen was mixed with 2.5 mL of absolute ethanol for storage and transportation to our laboratory at room temperature. On arrival, DNA was extracted by using the BioRobot MDx workstation (QIAGEN, Valencia, CA, USA) in accordance with the manufacturer’s recommendations and protocols. T. whip- plei quantitative PCR assays were performed as previously described (8). A case was deined as 2 positive quantita- tive PCR results in assays targeting 2 different T. whipplei DNA sequences. T. whipplei genotyping using 4 variable sequences was performed by using samples from children with high bacterial loads, as previously reported (9). Among the 150 healthy children, the prevalence of T. whipplei was 11% (2/18) in children <8 months of age, 37% (10/27) in children 8–24 months of age, and 44% (46/105) in children 2–10 years of age (Figure 1). None of the 8 wa- ter wells sampled were positive for T. whipplei. DISPATCHES 922 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 15, No. 6, June 2009 Author afiliations: Université de la Méditerranée, Marseille, France (F. Fenollar, J.-F. Trape, H. Bassene, C. Sokhna, D. Raoult); Pôle de Maladies Infectieuses, Marseille (F. Fenollar, D. Raoult); and In- stitut de Recherche pour le Développement, Dakar, Senegal (J.-F. Trape, H. Bassene, C. Sokhna, D. Raoult) DOI: 10.3201/eid1506.090182 Figure 1. Location of Dielmo and Ndiop in Senegal, Africa. Plus- symbol lines deine the Senegal frontiers. The number of fecal samples positive for Tropheryma whipplei and number tested for children in each age group was as follows: Dielmo, 1/13 from children <8 months of age, 5/9 from children 8–24 months of age, and 19/54 from children 2–10 years; Ndiop, 1/5 for children <8 months of age, 5/18 from children 8–24 months of age, and 27/51 from children 2–10 years of age.