Dose-response association between circulating selenium and blood pressure: a systematic review and meta-analysis

Abstract

Aim and scopeHypertension is considered a global health issue due to its growing prevalence. Although several studies investigated the relationship between circulating selenium and hypertension, their results are inconsistent. Therefore, we aimed to examine this association through a systematic review and dose-response analysis. MethodThe research protocol has been registered with PROSPERO (CRD42024555841). A systematic search was performed across Medline/PubMed, Scopus, Clarivate Web of Science, and Google Scholar until December 2023. The analysis calculated summary relative risks (RR) and 95 confidence intervals (95 CI) for the risk of hypertension by comparing the highest versus lowest levels of circulating selenium and using random-effects models. Additionally, dose-response relationships were assessed. ResultsOverall, 28,312 participants from 17 observational studies were included in the study. The highest versus lowest level of circulating selenium was not significantly related to risk of hypertension (RR:1.15; 95 CI: 0.99-1.33). Subgroup analysis indicated that the highest vs. lowest circulating selenium was associated with a 36 higher risk of hypertension in women (95 CI: 1.03-1.81). Nevertheless, there was no significant association in the subgroups of men (RR: 1.23; 95 CI: 0.79-1.90) or both genders together (RR: 1.02; 95 CI: 0.86-1.21). Additionally, non-linear dose-response analysis indicated a U-shaped association between circulating selenium and risk of hypertension (P non-linearity < 0.001). However, based on linear dose-response analysis, there was no significant association per 50 <mu>g/L increment in circulating selenium and risk of hypertension. When analyses were confined to representative studies, similar results were found. ConclusionThis systematic review and dose-response meta-analysis suggests a U-shaped association between circulating selenium and the risk of hypertension in representative and general adults. However, more prospective studies should be conducted to define the causality and mechanisms of this relationship.