Background and Objectives: Studies examining the risk factors for dyskinesia in Parkinson's disease (PD) have been inconsistent, and racial differences exist. Since there have been no systematic studies of the characteristics of dyskinesia in the Mainland Chinese population, we sought to elucidate the risk factors for dyskinesia. Methods: A total of 1974 PD patients from Mainland China were systematically investigated by univariable and multivariable analyses. PD patients with and without dyskinesia were stratified into 4 groups according to levodopa equivalent daily dose (LEDD) and analyzed by a Cox proportional hazards model. A longitudinal study of 87 patients with dyskinesia was classified into 3 groups according to the duration from onset of PD to the initiation of levodopa, and comparisons among groups were analyzed by the Mann-Whitney test. Results: Early age of onset, long disease duration, being female, high LEDD, low UPDRS III scores (ON-state) and high Hoehn-Yahr stage (ON-state) were predictors of dyskinesia. Dyskinesia was levodopa dosage-dependent, and the incidence increased remarkably when LEDD exceeded 300 mg/d (p < 0.05). The emergence of dyskinesia had no association with the initiation time of levodopa, and if the latter was more than 4 years, the duration of time on chronic levodopa free of motor complications was significantly shortened. Conclusions: We found risk factors for the prediction of dyskinesia. Our data shows that physicians should be cautious if the LEDD exceeds 300 mg/d. The development of dyskinesia was not correlated with the time of levodopa initiation.
Keywords: Parkinson's disease; disease duration; dyskinesia; levodopa; risk factors.