Associations of selenium status with all-cause and cause-specific mortality: a systematic review and meta-analysis of cohort studies

Abstract

Objective: To provide a systematic review and meta-analysis of population-based cohort studies on the association of selenium status with all-cause and cause-specific mortality.

Methods: Relevant studies were identified through systematic searches of MEDLINE and ISI Web of Knowledge. Risk ratios (RRs) reported across categories of selenium biomarkers were recalculated as continuous RR estimations per standard deviation (SD) using generalized least squares for linear trend estimation and pooled in random effects meta-analyses.

Results: The literature search identified 20 studies, including 17 studies on all-cause mortality, 9 studies on cardiovascular mortality and 7 on cancer mortality. An increase of selenium biomarker concentration by one SD was associated with 13 % lower all-cause mortality (RR [95 %-confidence interval], 0.87 [0.83-0.90]), 11 % lower cardiovascular mortality (0.89 [0.84-0.94]) and 15 % lower cancer mortality (0.85 [0.78-0.94]). Although moderate heterogeneity was observed, the inverse association with all-cause mortality was robust across countries with low or adequate selenium supply, selenium measurement methods, recruitment years, study quality scores, follow-up lengths and sample sizes. The trim and fill method showed no indications of relevant publication bias.

Conclusion: Selenium biomarkers are inversely associated with all-cause, cardiovascular and cancer mortality in the general population and clinical trials among selenium deficient populations are still needed.

Keywords: Cancer; Cardiovascular diseases; Cohort studies; Meta-analysis; Mortality; Selenium; Systematic review.

Fig. 1

Flow chart of literature search.

Fig. 2

Meta-analysis of risk ratios for all-cause mortality per one standard deviation increase of plasma/serum selenium or selenoprotein P concentrations Abbreviations: 95 %-CI, 95 % confidence interval; RR, risk ratio.

Fig. 3

Subgroup meta-analyses by the background Se levels for the association of plasma/serum selenium or selenoprotein P concentrations (risk ratios per 1 standard deviation) with all-cause mortality Abbreviations: 95 %-CI, 95 % confidence interval; RR, risk ratio; Se, selenium.

Fig. 4

Subgroup meta-analyses by selenium measurement method for the association of selenium biomarkers (risk ratios per 1 standard deviation) with all-cause mortality Abbreviations: 95 %-CI, 95 % confidence interval; AAS, atomic absorption spectrometry; ELISA, enzyme-linked immunosorbent assay; RR, risk ratio; ICP-MS, inductively coupled plasma mass spectrometry; SELENOP, selenoprotein P.

Fig. 5

Meta-analysis of risk ratios for cardiovascular mortality per one standard deviation increase of plasma/serum selenium or selenoprotein concentrations Abbreviations: 95 %-CI, 95 % confidence interval; RR, risk ratio.

Fig. 6

Meta-analysis of risk ratios for cancer mortality per one standard deviation increase of plasma/serum selenium or selenoprotein concentrations Abbreviations: 95 %-CI, 95 % confidence interval; RR, risk ratio.

Fig. 7

Functions, metabolic pathways and health effects of selenoproteins ^a Abbreviations: GPXs, glutathione peroxidases; H₂O₂, hydrogen peroxide, LOOH, lipid hydroperoxides; GSH, reduced glutathione; GSSG, oxidized glutathione; NADPH, reduced form of nicotinamide adenine dinucleotide phosphate; NADP⁺, oxidized form of nicotinamide adenine dinucleotide phosphate; TXNRDs, thioredoxin reductases; TXN–S–S, oxidized thioredoxin; TXN(SH)₂, reduced form of thioredoxin; MSRB1, methionine sulfoxide reductase B1; Met-R-O, methionine-R-sulfoxide; Met, methionine; DIOs, iodothyronine deiodinases; T₄, thyroxine; T₃, triiodothyronine; rT₃, reverse triiodothyronine; T₂, diiodothyronine; ER, endoplasmic reticulum; SELENOP, selenoprotein P; ROS, reactive oxygen species. ^a Created in BioRender. Cui, Z. (2025) https://BioRender.com/psa1qx3.