The motor phenotype of Parkinson's disease in relation to age at onset

Background: Parkinson's disease (PD) is heterogeneous and age at onset may define variation in clinical phenotype. Most previous studies have used various age cut-offs and have been based on clinical case series. Methods: We have studied the association between clinical features and age of onset in 358 community-based and regional patients with PD. Results: Tremor at presentation is twice as common in those with onset over 64 years as compared to those with onset under 45 (early onset PD - EOPD) and becomes more common with increasing age at onset (p values for trend ≤ 0.004). Dystonia affects 60% of those with EOPD, shows a curvilinear relationship with age at onset (cubic versus linear p=0.01) with highest risk in patients whose disease began before 48 years. In this study age at onset was a strong predictor of the development of dyskinesias, with younger age associated with a higher risk of dyskinesias. Following multivariable analysis, allowing for possibly confounding factors (disease duration, L-DOPA dosage, L-DOPA treatment duration) younger age at onset, (less than 55 years) predicted the development of L-DOPA induced dyskinesia (odds ratio <45 years 2.1, 95% CI 1.0, 4.8; odds ratio < 55 years 3.8, 95% CI 1.8, 8.0). Only 2/70 (2.9%) EOPD patients carried pathogenic parkin or PINK1 mutations and the clinical differences between early and late onset disease were not explained by the presence of mutations in these genes. Discussion: This study highlights the clinical differences between early and late onset PD, which have important implications for diagnosis and management.