Effects of the clozapine national registry system on incidence of deaths related to agranulocytosis

Publisher Website
Google Scholar
Abstract
Clozapine is the only medication distributed in the U.S. through a national patient registry system that provides the medication only if results of patients' weekly blood tests show no evidence of significant white blood cell suppression, an effect that can be fatal if it progresses to advanced agranulocytosis. This study assessed morbidity and mortality related to agranulocytosis during the first five years of the national registry system. Data from the national registry database maintained by the U.S. manufacturer of clozapine was used to determine the level of treating systems' adherence to the mandated program of weekly white blood cell counts, number of instances in which clozapine treatment was denied because of prior determination of white blood cell suppression, and number of cases of agranulocytosis and deaths related to agranulocytosis among treated patients from February 1990, when clozapine was commercially introduced in the U.S., through December 1994. The actual numbers of cases of agranulocytosis and related deaths were compared with expected outcomes based on clinical research done before the drug became available commercially. Approximately 97 percent of treating systems had a high overall level of adherence to the registry protocol. In 28 instances, the pretreatment authorization requirement resulted in denial of clozapine; after additional data were considered, 15 of the patients were cleared for treatment. The actual incidences of 382 cases of agranulocytosis and 12 related deaths were lower than the expected 995 cases and 149 deaths. The clozapine national registry system fostered early detection of white blood cell suppression, prevented retreatment with clozapine of patients who had previously developed white blood cell suppression, and brought about lower than expected rates of agranulocytosis and associated deaths.

This publication has 4 references indexed in Scilit: