Abstract
BACKGROUND: Freezing of gait (FOG) is an important milestone in the individual disease trajectory of people with Parkinson's disease (PD). Based on the cognitive model of FOG etiology, the mechanism behind FOG implies higher executive dysfunction in PD(FOG+). To test this model, we investigated the FOG-related phenotype and cognitive subdomains in idiopathic PD (iPD) patients without genetic variants linked to PD from the Luxembourg Parkinson's study. METHODS: A cross-sectional analysis comparing iPD(FOG+) (n = 118) and iPD(FOG-) (n = 378) individuals was performed, followed by the application of logistic regression models. Consequently, regression models were fitted for a subset of iPD(FOG+) (n = 35) vs. iPD(FOG-) (n = 126), utilizing a detailed neuropsychological battery to assess the association between FOG and cognitive subdomains. Both regression models were adjusted for sociodemographic confounders and disease severity. RESULTS: iPD(FOG+) individuals presented with more motor complications (MDS-UPDRS IV) compared to iPD(FOG-) individuals. Moreover, iPD(FOG+) individuals exhibited a higher non-motor burden, including a higher frequency of hallucinations, higher MDS-UPDRS I scores, and more pronounced autonomic dysfunction as measured by the SCOPA-AUT. In addition, iPD(FOG+) individuals showed lower sleep quality along with lower quality of life (measured by PDSS and PDQ-39, respectively). The cognitive subdomain analysis in iPD(FOG+) vs. iPD(FOG-) indicated lower scores in Benton's Judgment of Line Orientation test and CERAD word recognition, reflecting higher impairment in visuospatial, executive function, and memory encoding. CONCLUSION: We determined a significant association between FOG and a clinical endophenotype of PD with higher non-motor burden. While our results supported the cognitive model of FOG, our findings point to a more widespread cortical impairment across cognitive subdomains beyond the executive domain in PD(FOG+) with additional higher impairment in visuospatial function and memory encoding.