Prediction of rickettsial skin eschars in humans using an experimental guinea pig model.

Until now, when a new Rickettsia species was isolated in a tick, it was not possible to predict whether it was a human pathogen or if it would cause a skin eschar at the infection site. Guinea pigs are injected intradermally with 25 different Rickettsia species or subspecies: 16 induced an eschar, 5 induced inflammatory lesions and 4 have no effect. We observed that the occurrence of skin eschars in this model was significantly correlated (P <0.05) with observations of skin eschars in humans (14/16). The most common histological finding was mononuclear cell infiltration. Polymorphonuclear cell infiltration was observed for Rickettsia australis, Rickettsia japonica, and, as in humans, Rickettsia africae. The treatment of guinea pigs with corticosteroids prevents the apparition of eschar following Rickettsia bellii inoculation. Virulent, but not avirulent, Rickettsia prowazekii induced transient inflammatory lesions that were associated with dermal vasculitis, as is Rickettsia typhi. Therefore, the intradermal injection of Rickettsia in guinea pigs appears to be a relevant model for the prediction of the development of escharotic lesions following Rickettsia infection in humans. We speculate that skin eschar is the reflect of a local control avoiding extreme virulence.