White matter lesions volume and motor performances in the elderly

TLDR

The study investigates how white matter lesion volume and its localization relate to motor performance in older adults. White matter lesion volumes were quantified automatically, and motor function was measured at baseline and after eight years in 1,702 and 1,086 participants, respectively. High white matter lesion volumes, especially periventricular lesions, predict slower walking speed and greater decline over eight years, indicating a vascular contribution to motor decline.

Abstract

To investigate the cross-sectional and longitudinal associations between performance-based measures of motor function and volume of white matter lesions (WMLs), and to examine the influence of the localization of these lesions.At baseline, motor performances (maximum walking speed, Tinetti gait and balance subscales) were assessed in 1,702 subjects aged 80 years or younger from the Dijon (France), France center of the Three-City study. Volumes of WMLs lesions (total, periventricular, deep) were measured using an automated method of tissue segmentation and quantification of lesion size. At 8-year follow-up, walking speed was evaluated in 1,086 subjects.At baseline, mean and 95% confidence interval (CI) walking speed was lower in subjects with total volumes of WMLs >or=90th percentile (1.50 [1.45-1.55] m/s) than in subjects with lower volumes (1.56 [1.55-1.58] m/s; p = 0.004). Baseline total volumes of WMLs above the 90th percentile predicted walking speed decline during follow-up (odds ratio [95% CI] for having the greatest walking speed decline = 2.3 [1.3-4.1], p = 0.006). Moreover, high volumes of periventricular but not deep WMLs were associated with slower walking speed at baseline (p = 0.005) and over time (p = 0.001), and with lower Tinetti gait subscore (p = 0.02).Our study shows a cross-sectional and longitudinal association between high total volumes of WMLs, in particular volumes above the 90th percentile, and impaired mobility. These associations were independent of several confounders, including cognition, and were mainly accounted for by volumes of periventricular WMLs. These findings support the hypothesis of a vascular contribution to motor decline in the elderly.