Supplementary Material for: White Matter Hyperintensities and Functional Outcomes at 2 Weeks and 1 Year after Stroke

Background: Many stroke survivors remain at a functionally disabled state. Early prediction of functional outcome is an important step in the treatment and care of stroke patients. Brain imaging measures have received particular attention as one of the predictors of stroke outcomes. However, the associations between white matter hyperintensities (WMHs) and post-stroke recovery have been controversial. This study aimed to investigate whether deep and periventricular WMHs (DWMHs and PVWMHs, respectively) were associated with functional outcomes at 2 weeks and 1 year after stroke separately and interactively. Methods: A total of 408 patients were evaluated at the time of admission and 2 weeks after stroke, and 284 (78%) were followed 1 year later in order to evaluate consequences of stroke at both acute and chronic stages. Stroke outcomes were dichotomized into good and poor by applying a modified Rankin Scale (mRS) score cutoff of 1 (no significant disability)/2 (slight disability). Measures of stroke severity (National Institutes of Health Stroke Scale: NIHSS), physical disability (Barthel Index: BI), and cognitive function (Mini-Mental State Examination: MMSE) were administered. Brain magnetic resonance imaging (MRI) was performed at admission. DWMHs and PVWMHs were rated visually on axial FLAIR images using the 4-point modified Fazeka’s scale and categorized into mild (grades 0 and 1) or severe (grades 2 and 3). The demographic and clinical covariates of age, gender, years of education, previous history of stroke, depression, stroke location, and vascular risk factors were obtained. Associations of WMHs with stroke outcomes (mRS) and other measures (NIHSS, BI, and MMSE) at 2 weeks and at 1 year after stroke were investigated using logistic regression and repeated measures ANOVA after adjustment for the potential covariates. Results: Severe PVWMHs, but not severe DWMHs, were significantly associated with worse outcomes both at 2 weeks and at 1 year after stroke. Furthermore, significant interactive effects of PVWMHs and DWMHs on poor outcomes were found. Significant PVWMH group by time interactions on NIHSS and BI scores, and significant group effects of both DWMHs and PVWMHs on MMSE score were found. Conclusions: PVWMHs predicted poorer functional outcomes after stroke both in the acute and chronic phases, independently and interactively with DWMHs. In addition, PVWMHs were independently associated with worsening of stroke severity and physical disability. More careful evaluation and management of stroke patients with a high risk of functional dependence are indicated. Overall, PVWMH might be a prognostic marker of short- and long-term stroke outcomes.