Differential progression of motor impairment in levodopa-treated Parkinson's disease

OBJECTIVE: To monitor comparative progression of clinical impairment over 4 years in patients with Parkinson's disease (PD) who present on levodopa at two different levels of Hoehn and Yahr (HY) stages, II and III. BACKGROUND: The rate of clinical impairment progression in patients with PD being treated with levodopa has not been studied in detail using current, standardized assessment tools. Sample size estimates for all levodopa adjunctive treatment studies and proper definition of study groups require a solid estimate of longitudinal motor impairment progression. DESIGN/METHODS: From our computer database, we identified two groups of patients with PD being treated with levodopa based on their initial HY stage at presentation to our center (II or III). Fifty randomly selected subjects in each stage were monitored in the ON state with annual Unified Parkinson's Disease Rating Scale (UPDRS) motor scores, dyskinesia ratings, and antiparkinsonian medication doses using a repeated measures analysis of variance. RESULTS: The stage II and stage III subjects had similar disease duration. In stage II subjects, parkinsonian impairment was maintained without progression over 4 years, but in association with significantly higher dyskinesia scores and dopaminergic medication doses. In stage III subjects, UPDRS motor scores deteriorated despite more medication and increased dyskinesias. Of the established six factors comprising the UPDRS motor scale, bradykinesia accounted for the increased impairment. Initial UPDRS motor score and disease duration did not influence progression of motor impairment. CONCLUSION: In subjects with similar disease duration, progression of PD motor impairment differs significantly between stage II and stage III subjects over 4 years. Whereas in stage II subjects, parkinsonian impairment can be stabilized at the expense of increased dyskinesia and dopaminergic drugs, once subjects reach stage III, motor impairment progresses. Power estimates and sample size calculations for these groups of patients should be calculated separately.