Antimicrobial therapy of rickettsial diseases.

Among the rickettsioses, Q fever has specific pathological and clinical features which do not allow it to be reviewed in comparison with other rickettsioses. Arthropod-borne rickettsioses comprise three distinct disease groups: spotted fever rickettsioses, typhus, and scrub typhus. Among the spotted fever rickettsioses, Rocky Mountain spotted fever (RMSF), Mediterranean spotted fever (MSF), and tick typhus of the eastern hemisphere have been widely studied. In this report we review the in vitro and in vivo efficacies of antibiotic treatments for arthropod-borne rickettsioses. Since 1948, when the first cases of RMSF were treated with chloramphenicol (32), these diseases have been curable. Despite this therapeutic evolution, RMSF lethality remains high (4%) (13), as does MSF lethality in several countries (44). In Japan, it was suggested that the apparent increasing incidence of rickettsiosis during the 1970s was related to a change in empiric antibiotic therapy for unexplained febrile illness (56). The ineffective beta-lactams were prescribed, despite the existence of effective drugs, i.e., chloramphenicol and tetracycline. Since spotless fever has been described both in patients with RMSF (33, 58) and in patients with MSF (8), rickettsioses may present early as nonspecific febrile illnesses; these illnesses were previously undiagnosed and were cured by empiric antibiotic therapy, but they developed when beta-lactams began to be used. The delay in administration of an effective antibiotic leads to enhanced mortality (44, 46, 63). The recent therapeutic use of new quinolone compounds and new macrolides that are active against rickettsiosis may be of importance. Another specific problem is the treatment of pregnant women and young children, for whom tetracyclines are toxic, and the use of chloramphenicol exposes these patients to the risk of aplasia. The cost of treatment is of importance in developing countries, and the prescription of new compounds increases this cost. Ultimately, the treatment duration is still questioned, and shortened treatment durations have been proposed. The investigator who advocates changes in accepted treatment regimens for rickettsial diseases should critically evaluate some important issues. First, testing of the in vitro efficacy of a new drug on representative organisms is critical. The availability of a larger number of strains allows the evaluation of heterogeneous susceptibilities. The in vitro data should not be directly extrapolated to treatment regimens for use in humans. The reliability of animal models has not been demonstrated for these diseases. Second, clinical trials should involve only biologically confirmed cases. Preliminary results from noncontrolled trials are a preliminary approach in evaluating the results, and clinical failures are of