The present pharmacoepidemiologic study was performed to characterize the profile of adverse drug reactions (ADRs) reported with selegiline, a monoamine oxidase B (MAO-B) inhibitor used in the treatment of Parkinson's disease and previously reported to induce an excess of mortality. The analysis was performed with use of the French Pharmacovigilance Database between 1989 and 1997. This database includes all ADRs reported by French practitioners (and especially "serious" and "unexpected" ADRs). Three different analyses were performed: identification of ADRs reported with selegiline, comparison with the ADR profile observed with other antiparkinsonian drugs, and a case/non-case study investigating the occurrence of cardiovascular ADRs with selegiline in comparison with other drugs in general and other antiparkinsonian drugs (e.g., levodopa [L-Dopa], dopamine agonists) in particular. The most often reported ADRs with selegiline were psychiatric (delirium, hallucinations, agitations), cardiovascular (orthostatic hypotension, arterial hypertension, etc.) and neurologic (sedation, abnormal movements, etc.). Psychiatric and cardiovascular ADRs were more frequently reported with selegiline than with L-Dopa or dopamine agonists. The case/ non-case study found an increased risk of cardiovascular ADRs (OR = 1.72; 95% Cl = 1.16-2.55)when selegiline was associated with L-Dopa. These data show that the profile of selegiline-induced ADRs differs from that of other antiparkinsonian drugs (L-Dopa, dopamine agonists) with more psychiatric and cardiovascular ADRs. We suggest that the higher frequency of cardiovascular ADRs could explain, at least partially, the previously reported increase in mortality rate.