Abstract TP250: Migraine and Deep White Matter Hyperintensities are Independently Associated With Post-Stroke Recrudescence

Introduction: Transient reemergence of previous stroke-related deficits (or post-stroke recrudescence, PSR) may be triggered by infections, hypotension and hyponatremia. Patients with PSR (PSR+) are more often women, non-White, have more diabetes, dyslipidemia, history of smoking, infarcts from small-vessel disease and more severe deficits at stroke onset than patients without PSR (PSR-). Whether increased white matter hyperintensities (WMH) burden is also more frequent in PSR+ than in PSR- remained to be determined. We compared WMH burden in patients with stroke who experienced PSR and in controls. Methods: We assessed WMH burden in MRIs of 75 patients with stroke and PSR (PSR+) and 150 age- and gender-matched controls with DWI-confirmed ischemic stroke (PSR-). We evaluated the extension of periventricular WMH (PWMH) and deep WMH burden (DWMH). A conditional logistic regression model was used to identify independent predictors of PSR. Results: Multivariate analysis showed that patients with diabetes (Odds ratio, OR, 5.02), hyperlipidemia (OR 2.97), higher NIHSS (OR 1.2), DWMH (OR 3.94) and migraine (OR 8.06) were significantly more likely, and that White patients were less likely (OR 0.22) to have PSR. Conclusions: DWMH and migraine were more frequent in PSR+ than in PSR-. PSR is not a migraine attack because transient worsening of residual focal deficits, or transient recurrence of previous stroke-related deficits are required for diagnosis. It has been hypothesized that recurring migraine attacks are associated to ischemia, and that DWMH may be caused by ischemia. Migraine, diabetes and hyperlipidemia may increase the risk of DWMH. DWMH may disrupt brain connectivity and increase susceptibility to PSR. Alternatively, spreading depression or other abnormal neuronal activity may lead to migraine attacks and to PSR. DWMH, migraine and PSR may share a common, yet unveiled underlying mechanism.