Long-term cognitive follow-up of Parkinson's disease patients with impulse control disorders.

This study investigated cognitive functions in Parkinson's disease (PD) patients with impulse control disorders (ICDs) and aimed to identify possible predictors of behavioral outcome. In this longitudinal cohort study, 40 PD outpatients with ICDs and 40 without, were matched for sex, age at PD onset, age and disease duration at cognitive assessment. All patients had two neuropsychological assessments at least 2 years apart (mean, 3.5 years). Multivariate logistic regression analysis was performed to identify predictors of ICDs remission at follow-up. The PD patients with and without ICDs had overall comparable cognitive performance at baseline. When evaluating changes between baseline and follow-up, we found significant group × time interactions in several frontal lobe-related tests, with the ICDs group showing a less pronounced worsening over time. ICDs remission was associated with better performance at baseline in working memory-related tasks, such as digit span (odds ratio [OR] = 2.69 [95% confidence interval (CI), 1.09-6.66]) and attentive matrices (OR=1.19 [95%CI, 1.03-1.37]). ICDs remitters and non-remitters had no remarkable differences in baseline PD-related features and therapy management strategies (including the extent of dopamine agonist dose reduction). In conclusion, ICDs in PD patients are not related to greater cognitive impairment or executive dysfunction, but rather show relatively lower cognitive decline over time. The impaired top-down inhibitory control characterizing ICDs is likely attributable to a drug-induced overstimulation of relatively preserved prefrontal cognitive functions. Full behavioral remission in the long term was predicted by better working memory abilities.