Tavassoli_2007_Drug.Saf_30_1063

BACKGROUND: Cholinesterase inhibitors (ChEIs) could be involved in several drug-drug interactions (DDIs) because of their complex pharmacodynamic and/or pharmacokinetic properties. AIM: To identify spontaneous reports in the French Pharmacovigilance Database containing DDIs with the three ChEIs marketed in France (donepezil, galantamine or rivastigmine) and to compare the informativity of two national drug references, the French national formulary (Vidal) and the British National Formulary (BNF), for their ability to identify ChEI-related DDIs.
METHODS: Spontaneous reports submitted to the French Pharmacovigilance Database concerning donepezil, galantamine or rivastigmine were reviewed by two clinical pharmacologists from Toulouse Regional Pharmacovigilance Centre. Spontaneous reports containing DDIs were identified according to Vidal, BNF or their own judgement (and with use of the interaction supplement of the French independent drug information bulletin La Revue Prescrire). Then, the potential of DDIs to result in adverse drug reactions (ADRs) was evaluated. Finally, the presentations of the different ChEIs in the two references (Vidal, BNF) were compared for their DDI informativity.
RESULTS: A total of 1058 spontaneous reports were identified that involved ChEIs in the French Pharmacovigilance Database up to 31 March 2006; of these 376 (35.5%) contained at least one DDI according to experts' judgement. In total, 118 DDIs (31.4%) were the cause of ADRs. Most of the DDIs were due to pharmacodynamic interactions (247 cases, 65.7%). The most frequently encountered drugs involved in DDIs were bradycardic (205 cases, 54.5%) and anticholinergic (118 cases, 31.4%) drugs. DDIs were found in 309 spontaneous reports (29.2%) according to Vidal and in 127 (12.0%) according to BNF. In total, 88 'serious' ADRs were related to DDIs (including seven deaths, mainly due to cardiovascular ADRs). The most frequently observed ADRs due to DDIs were cardiovascular (67 cases, mainly bradycardia, atrioventricular block and arterial hypotension) and neurological (33 cases, mainly mental confusion). Comparison of the different presentations of summary of product characteristics (SPC) showed that Vidal was more informative than BNF for all the ChEIs, and that galantamine had the most complete data in the two references. CONCLUSION: DDIs were present in more than one-third of spontaneous reports including ChEIs registered in the French Pharmacovigilance Database. Approximately, one-third of these DDIs were the cause of ADRs. The informativity of European drug dictionaries differs substantially and Vidal was found to be more informative than BNF for all the ChEIs.