Computational fluid dynamics-derived flow helicity and recurrent stroke risk in symptomatic intracranial stenosis.

BACKGROUND: Patients with symptomatic intracranial atherosclerotic stenosis (ICAS) remain at risk of recurrent ischemic events despite treatment. Stenosis severity is routinely used for risk assessment, but does not describe the lesion-level flow field. We used computational fluid dynamics (CFD) to assess whether flow helicity within the culprit stenosis was associated with 1-year target-vessel stroke. METHODS: We studied 204 adults with symptomatic ICAS who underwent CT angiography-based vascular reconstruction and CFD analysis. The prespecified exposure was maximum absolute flow helicity at the narrowest cross-sectional plane of the culprit stenosis, transformed using log(1+x) and standardized per SD. The primary outcome was 1-year target-vessel stroke, and the secondary endpoint was target-territory stroke or vascular death. Multivariable Cox models adjusted for major clinical, anatomical, and hemodynamic covariates were supplemented by Kaplan-Meier, cutpoint, tertile, subgroup, and Winsorization analyses. RESULTS: Multivariable analyses included 203 patients. During follow-up, 31 target-vessel strokes and 33 composite events occurred. Each 1 SD increase in log-transformed maximum absolute helicity was associated with target-vessel stroke (HR 1.461, 95% CI 1.049 to 2.035, P=0.0249). The association with the composite endpoint was directionally similar but not significant (HR 1.351, 95% CI 0.973 to 1.875, P=0.0720). Median-split Kaplan-Meier curves separated for both outcomes, and tertile analysis showed higher target-vessel stroke risk in the highest versus lowest tertile (HR 2.951, 95% CI 1.067 to 8.160, P=0.0371). CONCLUSIONS: In symptomatic ICAS, higher lesion-level maximum absolute helicity was associated with 1-year target-vessel stroke, supporting further evaluation of helicity as a CFD-derived marker of flow disturbance.