Nonlinear associations of dietary selenium intake with all-cause and cardiovascular mortality after stroke in US adults: findings from NHANES 1999-2018

Abstract

Abstract Background: The association between dietary selenium intake and mortality in stroke patients remains unclear. We aimed to explore these associations through NHANES 1999-2018. Methods: Stroke participants ≥20 years of age were enrolled and were followed until December 31, 2019. Dietary selenium intake and stroke diagnosis were obtained by self-report on the questionnaire. Mortality information was obtained through prospective matching with the National Death Index database. Results: 1334 stroke survivors were included. Dietary selenium intake was significantly and inversely associated with all-cause mortality in stroke patients after adjusting for all confounders (hazard ratio [HR]=0.997, p=0.024). Compared to the reference population, stroke patients with dietary selenium intake in quartiles Q2, Q3, and Q4 were associated with reduced all-cause mortality (p for trend=0.02). Dietary selenium intake at Q2 was associated with reduced CVD mortality after stroke and not significantly associated with cancer mortality. Restricted cubic spline analysis indicated that dietary selenium intake was nonlinearly associated with all-cause (p for nonlinear=0.0361) and CVD mortality (p for nonlinear=0.0189) in stroke patients. Segmented regression showed that dietary selenium intake was inversely associated with all-cause and CVD mortality at <81 mcg/d and 87 mcg/d, respectively, with no additional protective effects thereafter. Stratified analyses suggested that these effects were only present in specific ages and genders. Conclusions: In a national longitudinal cohort, dietary selenium intake was nonlinearly associated with all-cause and CVD mortality but not cancer mortality after stroke in U.S. adults. Dietary selenium intake at 81 mcg/d and 87 mcg/d had the optimal protective effect on post-stroke all-cause and CVD mortality, respectively.