Exercise and white matter hyperintensities: a randomized clinical trial

Subcortical vascular cognitive impairment (SVCI) is a risk factor for dementia. White matter hyperintensity (WMH) lesions are considered a hallmark of SVCI. We investigated the impact of progressive resistance exercise training (RT) on WMH in individuals with SVCI. This 12-month, single-blinded, proof-of-concept randomized clinical trial involved 89 community-dwelling individuals aged 55 years or older living with SVCI. Participants were randomized (1:1 ratio) to twice-weekly (60 min/day) RT (n = 44) or a balance and tone (BAT, n = 45) control intervention. Whole-brain and region-specific WMH volumes were measured using T₂- and proton density-weighted magnetic resonance imaging. Group differences in the 12-month change in WMH from baseline were determined via analysis of covariance, adjusting for baseline WMH, intracranial volume, age, sex and training format (remote versus in-person). Exploratory analyses tested associations between changes in WMH and changes in growth factors. Participants were mostly females (n = 58, 65.17%) with an average of 74.50 (SD 5.71) years. There were no differences between groups in whole-brain WMH volume after 12 months [estimated mean difference (95% CI): -0.016 (-0.052 to 0.019) log-transformed cm³, P = 0.364]. Region-specific analysis showed main effects of training on WMH volume in the temporal lobe [-0.115 (-0.207 to -0.023), P = 0.015] and splenium of the corpus callosum [-0.120 (-0.215 to -0.025), P = 0.014], favouring the RT group. There was a significant group-by-sex interaction, indicating that intervention effects varied by sex [estimated mean difference-of-differences (95% CI) = -0.093 (-0.170 to -0.016), P = 0.019]. Post hoc contrasts showed that RT (versus BAT) significantly reduced whole-brain WMH volume in female [-0.049 (-0.093 to -0.005), P = 0.029] but not male [0.044 (-0.017 to 0.105), P = 0.152] participants. Exploratory analysis showed that increasing levels of vascular endothelial growth factor [standardized β (95% CI) = -0.503 (-0.838 to -0.167), P = 0.004] were associated with a reduction in whole-brain WMH volume. No associations between insulin-like growth factor 1 and WMH volumes were found (all P > 0.05). Based on these findings, we conclude that progressive RT may reduce WMH progression in community-dwelling individuals living with SVCI. This trial is registered with ClinicalTrials.gov, number NCT02669394; submitted for registration on 1 February 2016; the first patient enrolment was on 26 May 2016.