RESUMO
BACKGROUND AND PURPOSE: Dopamine agonists (DAs) are generally considered to be deprived of the highly dyskinetic effect of levodopa in Parkinson's disease (PD) patients. However, the risk for dyskinesia induced by DA monotherapy and the contribution of clinically significant factors in the development of this disorder have never been systematically assessed. METHODS: A systematic literature search was conducted for randomized, levodopa-controlled trials of DAs in early PD. A meta-analysis was performed to calculate the combined odds ratio (OR) for dyskinesia. Meta-regressions were subsequently performed on dyskinesia OR including individually as covariates the effects of mean disease duration, treatment duration and DA dose. In an additional analysis the effect of adjunct levodopa on the odds for dyskinesia was investigated. RESULTS: DA monotherapy resulted in an 87% lower risk for dyskinesia compared with treatment with levodopa (OR = 0.13, 95% confidence interval 0.09-0.19, P < 0.001). The risk for dyskinesia was independent of the dose of DA, disease duration and treatment duration. A dose-related pattern was revealed between adjunct levodopa in the DA group and dyskinesia. Nevertheless, the odds for dyskinesia in the DA group were constantly lower than in the levodopa group. CONCLUSION: Initial DA treatment encompasses a lower risk for dyskinesia even after the unavoidable introduction of levodopa that increases the risk for dyskinesia in a dose-related manner. As the dose and treatment duration with DAs are factors independent of the risk of dyskinesia, monotherapy with DAs in early PD is suggested at doses that ensure efficacy and delay the need for levodopa, always following an adequate evaluation of the risks DAs can pose in individual patients.