Exploring the potential effect of polypharmacy on the hematologic profiles of clozapine patients

Clozapine, an atypical antipsychotic with documented efficacy in the management of treatment-resistant schizophrenia, is associated with the risk of adverse hematological outcomes. Of particular concern are reductions in white blood cells (WBC) and absolute neutrophil counts (ANC). Individuals who display moderate leukopenia (3000/mm(3) > WBC ≥ 2000/mm) upon initiation of clozapine therapy are at increased risk of developing agranulocytosis, defined as an ANC less than 500/mm. Complications of agranulocytosis can be severe and include increased risk of infection and mortality.

The primary objective of this study was to examine data on clozapine recipients who experienced adverse drug reactions (ADRs) related to decreases in WBC or ANC and ascertain whether other drugs and/or drug interactions had played a role. The analysis included multiple classes of medications.

A retrospective chart review was performed of open and closed medical records of all inpatient recipients of clozapine at a state psychiatric center between January 1, 2004 and June 30, 2011. Laboratory records of patients prescribed clozapine were examined for abnormal WBC counts or ANC. A hematological ADR was considered to have occurred if there was a substantial drop in either WBC or ANC or mild or moderate leukopenia or granulocytopenia. Each episode was analyzed for medications that might have contributed to the ADR. Data were collected for all scheduled and STAT medications started at any point during the clozapine patient's hospitalization. The following seven medication groups, based on the Therapeutic Classification System of the American Hospital Formulary System (AHFS), were chosen for analysis because they were consistently used in the majority of the patient population: antihistamines, anti-infectives, autonomic agents, cardiovascular agents, antipsychotics, vitamins, and gastrointestinal agents. Pearson correlation coefficients were calculated to identify associations between the presence of hematological ADRs and medications administered concomitantly with clozapine.

The following significant correlation coefficients were found between the use of a class of medications and the occurrence of a hematological ADR: antiinfective agents 0.409 (p < 0.01), gastrointestinal agents 0.329 (p < 0.01), and autonomic agents 0.309 (p < 0.01). In the subset of patients who were prescribed a proton-pump inhibitor or ranitidine concomitantly with clozapine, 24/26 (96%) experienced a hematological ADR.

Autonomic agents, anti-infective agents, and proton pump inhibitors and other gastrointestinal agents were all associated with hematological ADRs when co-prescribed with clozapine. Medications from these classes should be initiated cautiously in patients being treated with clozapine to avoid precipitous drops in ANC or WBC that may increase the risk of agranulocytosis.