Abstract
BACKGROUND: The aim of this work was to identify early clinical predictors of important outcomes in Parkinson's disease (PD). In PD, treatment-resistant (e.g., dementia, falling) and other important functional outcomes (e.g., declines in quality of life [QOL] and activities of daily living [ADL]) emerge and become increasingly disabling. METHODS: We analyzed longitudinal data from 491 early, untreated PD subjects who enrolled in the PreCEPT trial, had baseline SPECT dopamine transporter deficit, and have continued in the PostCEPT observational cohort. After PreCEPT, antiparkinsonian medications were added if needed. Baseline clinical precursors were examined as potential predictors of selected outcomes. Separate and multivariate logistic regressions, adjusted for certain baseline factors, were performed for dichotomized outcomes evaluated at the last PostCEPT visit. RESULTS: On enrollment, subjects had average disease duration of 0.8 years and were followed for an average of 5.5 years. Some baseline precursors were found to be predictive: disease stage, cognitive, and ADL scores for dementia; disease stage, ADL, and motor and freezing scores for hallucinations; disease stage, depression, ADL, and freezing and walking scores for falling; and ADL, depression, and motor and walking scores and disease stage for QOL decline. No baseline clinical feature predicted decline in ADL. Being on levodopa was not a significant predictor of any outcome, but subjects on a dopamine agonist were significantly less impaired with respect to falling, abnormal Mini-Mental State Examination, and QOL. CONCLUSIONS: Although there are limitations, results support the value of longitudinal follow-up of clinical trial populations to identify early clinical precursors of important outcomes and thereby identify high-risk patients early on.