Evaluation of Case Reports of Aplastic Anemia in Felbamate Users

Objectives

To determine the scope of the problem of felbamate-induced aplastic anemia, based on an evaluation of the case reports against the background of existing knowledge about the etiology of the disease. The large number of case reports raised sufficient concerns about the safety of this new anticonvulsant that its use has been severely restricted, despite considerable efficacy when other treatments are ineffective.

Methods

All relevant clinical information and, wherever possible, the bone marrow biopsies, were obtained for 31 cases that were reported in the U.S. These were reviewed by a hematologist to determine if they met the criteria for aplastic anemia according to the IAAAS. The confirmed cases were evaluated by epidemiologists who participated in the IAAAS, and based on that experience, the degree of likely confounding for felbamate as an etiologic agent was assessed. Using sales data to estimate the denominator, the incidence of aplastic anemia among felbamate users was calculated under various assumptions (e.g., defining the numerator using all cases, and using only those where there is no apparent confounding for a felbamate association). The estimates were compared with estimates for other anticonvulsants from other sources.

Results

A total of 23 cases were confirmed as aplastic anemia; felbamate was judged to be the only or most plausible cause for 11 of them. The incidence of aplastic anemia among users was estimated to be between 27-209 cases per million per year, with a “most probable” value of 127 per million.