The effect of morphometric brain changes on gait-cognitive impairment of patients with Parkinson’s disease

With the progression of Parkinson`s Disease (PD), patients suffer from impairments of gait and cognition [1,2]. More severe cognitive impairment is often accompanied by worse gait performance [3]. It is widely believed that ongoing neurodegeneration underlies both disorders. While several studies published their findings regarding brain morphology changes correlating either with cognitive or gait parameters [4], it stays open if there are volumetric differences in brain-regions related to gait between patients with and without cognitive impairment. Is there a relationship between dual-task gait results and brain atrophy in PD patients and is it depending on the existence of cognitive impairment? We included 79 drug-naive PD patients (59.5±12 years) and 54 healthy controls (HC) (60.5 ±9.0 years). All subjects underwent the MoCA test and a single-task (ST) and dual-task (DT) gait assessment. In DT the subjects had to count down from 100 by sevens. The DT cost was calculated for velocity, stride length and cadence, as follows: [(DT − ST)/ST × 100]. Additionally, brain MRI including a T1-weighted images with 1mm3 isotropic resolution was done on a 3 T scanner Based on gait and MoCA results, the subjects in both groups were further divided into cognitively unimpaired subjects, with MoCA ≥24 and first PCA component of DT cost >-0.178 (PD 44, 58.63±12.2 years; HC 43, 61.05±9.3 years), and cognitively impaired subjects with MoCA <24 and first PCA component of DT cost of ≤ -0.178 (PD 35, 60.65±12.3 years; HC 11, 58.55±7.5 years). A voxel-based-morphometry analysis using a multiple regression model with the covariates DT-cost, total intracranial volume, age and sex was performed using the CAT12 software. For cognitively unimpaired PD patients, a cluster in the primary motor cortex positively correlated with the stride-length DT cost (pFWEcorr=0.027). This correlation was not detectable for either the cognitively impaired PD group or both HC subgroups. No significant correlations were found for velocity DT-cost and cadence DT-cost. Fig. 1: Results of VBM, positive correlation of DT-cost for stride length for cognitive unimpaired PD patients (significant peak at -21,-12,78)Download : Download high-res image (120KB)Download : Download full-size image By filtering the group of PD patients into a cognitively impaired and cognitively unimpaired group, we were able to detect a correlation between the precentral gyrus volume and stride-length DT cost in patients without cognitive impairment. This indicates that the poorer DT performance in these patients is mainly driven by degeneration of motor brain regions. This correlation is not detectable in the whole PD dataset, pointing to the fact that more extensive brain atrophy involving different brain regions, might be responsible for worse results in DT in cognitively impaired subjects. In conclusion, the impact of cognitive impairment on gait analysis in PD patients should be considered an important influencing factor. Supported by the Czech Ministry of Health (Grant No. NU20-04-00327)