Recovery of balance and gait after stroke is deteriorated by confluent white matter hyperintensities: Cohort study.

BACKGROUND White matter hyperintensities (WMHs) are well known to affect post-stroke disability, mainly by cognitive impairment. Their impact on post-stroke balance and gait disorders is unclear. OBJECTIVES We aimed to test the hypothesis that WMHs would independently deteriorate post-stroke balance and gait recovery. METHODS This study was performed in 210 individuals of the cohort Determinants of Balance Recovery After Stroke (DOBRAS), consecutively enrolled after a first-ever hemisphere stroke. Clinical data were systematically collected on day 30±3 (D30) post-stroke and at discharge from the rehabilitation ward. WMHs were searched on MRI, graded with the Fazekas scale, and dichotomized as no/mild (absence/sparse) or moderate/severe (confluent). The primary endpoint was the recovery of the single limb stance, assessed with the Postural Assessment Scale for Stroke (PASS). The secondary endpoint was the recovery of independent gait, assessed with the modified Fugl-Meyer Gait Assessment (mFMA). The adjusted hazard ratios (aHRs) of achievements of these endpoints by level of WMHs were estimated by using Cox models, accounting for other relevant clinical and imaging factors. RESULTS Individuals with moderate/severe WMHs (n=86, 41%) had greater balance and gait disorders and were more often fallers than others (n=124, 59%). Overall, they had worse and slower recovery of single limb stance and independent gait (p<0.001). Moderate/severe WMHs was the most detrimental factor for recovery of balance (aHR 0.46, 95% confidence interval [CI] 0.32-0.68, p<0.001) and gait (0.51, 0.35-0.74, p<0.001), along with age, stroke severity, lesion volume and disrupted corticospinal tract. With cerebral infarct, endovascular treatments had an independent positive effect, both on the recovery of balance (aHR 1.65, 95% CI 1.13-2.4, p=0.009) and gait (1.78, 1.24-2.55, p=0.002). CONCLUSIONS WMHs magnify balance and gait disorders after stroke and worsen their recovery. They should be better accounted for in post-stroke rehabilitation, especially to help establish a prognosis of mobility. ClinicalTrials.gov registration: NCT03203109.