BACKGROUND: To what extent cerebral small vessel disease (SVD) contributes to functional disability remains unclear. We investigated the longitudinal effect of SVD burden on functional decline. METHODS: In this investigator-initiated, prospective, multicenter, and observational study, patients receiving oral antithrombotic therapy for cerebrovascular or cardiovascular diseases were enrolled from 52 hospitals across Japan (2016-2019) and followed for 24 months. SVD score (range, 0-4) was calculated on baseline magnetic resonance imaging obtained under prespecified conditions. The outcomes were an increase in modified Rankin Scale (mRS) score from baseline to end of follow-up (ΔmRS score ≥1). Logistic regression evaluated the association between ΔmRS score ≥1 and SVD score, adjusting for age, sex, premorbid mRS, vascular risk factors, antithrombotic therapy, magnetic resonance imaging field strength, and bleeding or ischemic events during follow-up. Bleeding and ischemic events were treated as time-dependent covariates. Causal mediation analyses using a 4-way decomposition separated the effects of SVD score ≥3 on ΔmRS score ≥1 into direct and indirect effects (mediated by major bleeding or ischemic events). RESULTS: Of 5378 patients enrolled, 613 were excluded mainly due to the lack of clinical or imaging data. Finally, 4765 patients were analyzed (1573 women; median age, 73 years); ΔmRS score ≥1 was observed in 13.2% of patients with SVD score 0, 16.1% with 1, 20.7% with 2, 28.1% with 3, and 28.2% with 4. SVD score of 3 (adjusted odds ratio, 1.65 95% CI, 1.27-2.15) and SVD score of 4 (1.62 95% CI, 1.19-2.19) were associated with ΔmRS score ≥1 compared with SVD score of 0. Causal mediation analysis indicated that 73.05% of the total effect of SVD score ≥3 on ΔmRS score ≥1 was attributed to the controlled direct effect (coefficient, 0.497 95% CI, 0.26-0.73). CONCLUSIONS: SVD burden was associated with functional decline beyond significant bleeding or ischemic events.
Arakaki et al. (Mon,) studied this question.