There was increasing evidence suggesting that angiotensin I-converting enzyme may play an important role in the pathogenesis of PD. Our former study has shown that angiotensin I-converting enzyme gene (ACE) may confer a susceptibility for the risk of Parkinson's disease (PD). Meanwhile, recent studies have emphasized that genetic factors may involve in the occurrence of the adverse effects of chronic L-dopa therapy in PD patients. This study was designed to assess whether genetic polymorphism of the ACE could be a predictor of L-dopa-induced adverse effects in PD. There were 251 patients included in this study and their mean age at onset of disease was 63.3+/-11.4 years. The duration of disease and the treatment with L-dopa was 6.3+/-5.1 and 5.0+/-4.3 years, respectively. The frequency of the homozygote ACE-II genotype of the ACE in PD patients with L-dopa-induced psychosis was significantly higher than that in PD patients without the adverse effect (63.3% vs 43.0%; chi(2)=6.347, OR=1.435, 95%CI=1.105-1.864, p=0.012). However, the ACE polymorphism was not associated with the risk to develop dyskinesia or motor fluctuation induced by L-dopa. Furthermore, a logistic regression analysis confirmed that the ACE-II genotype was an independent risk factor for L-dopa-induced psychosis in PD patients (OR=2.542, p=0.012). In conclusion, results of the study showed that ACE-II genotype might confer a primary predictor for the occurrence of psychosis in L-dopa-treated PD.