A 12-year population-based study of freezing of gait in Parkinson's disease.

INTRODUCTION Freezing of gait (FOG) is a potentially disabling motor problem in Parkinson's disease (PD) with uncertain etiology. Longitudinal studies of FOG in PD are scarce. We determined the prevalence, incidence, and associated clinical risk factors and concomitants of FOG during prospective long-term follow-up of a population-based PD cohort. METHODS A community-based prevalent cohort of 232 PD patients was followed prospectively over 12 years. Reassessments were conducted at 4 and 8 years, and then annually. FOG, as well as severity of parkinsonism, motor complications, and psychotic symptoms were assessed by the Unified PD Rating Scale, and cognitive impairment by the Mini-Mental State Examination. Generalized estimating equations were applied to investigate baseline risk factors and concomitants of FOG over time. RESULTS The point prevalence of FOG at baseline was 27% (95% confidence interval (95%-CI) 22-33%). By study end, 63% (95%-CI 56-69%) of patients had developed FOG. The incidence rate of FOG was 124.2 (95%-CI 101.5-152.1) per 1000 person-years. Motor fluctuations (odds ratio (OR) 3.45; p = 0.036) and higher levodopa dose (OR 1.30/100 mg, p = 0.009) at baseline were independent risk factors of incident FOG. Prevalent FOG over time was additionally associated with features thought to reflect extrastrial, non-dopaminergic pathologies, including PIGD (postural instability/gait difficulty, OR 6.30/10 points, p < 0.001) and psychosis (OR 1.85; p = 0.016). CONCLUSION These findings demonstrate that FOG affects the majority of patients in the general PD population and provide support to the hypothesis that alterations in both basal ganglia and extrastriatal brain areas are involved in the pathogenesis of FOG in PD.