Selenium supplementation for the primary prevention of cardiovascular disease.

Randomized controlled trials evaluating selenium-only supplementation for the primary prevention of CVD. 12 trials with 19,715 participants; populations included general adults and individuals at high risk of CVD; many male participants; some cancer-related subgroups; study designs largely double-blind with placebo or no-intervention controls; follow-up spanned from weeks to up to around 12 years; baseline selenium status varied by country.

Selenium supplementation (single ingredient) given orally; regimens varied across trials: dosages ranged from 36 to 800 µg/day; forms included selenomethionine, selenium yeast, and selenium-enriched foods; durations ranged from about 2 weeks to up to 12 years; administration included tablets and selenium-enriched foods (bread, biscuits, milk, chicken meat).

No significant benefit of selenium supplementation on major cardiovascular outcomes or mortality. Pooled results: all-cause mortality RR 0.97 (95% CI 0.88–1.08); CVD mortality RR 0.97 (95% CI 0.79–1.20); non-fatal CVD events RR 0.96 (95% CI 0.89–1.04); all CVD events RR 1.03 (95% CI 0.95–1.11). Small, non-significant increased risk of type 2 diabetes (RR 1.06, 95% CI 0.97–1.15); SELECT reported alopecia (RR 1.28) and dermatitis (RR 1.17). Lipids: total cholesterol −0.11 mmol/L (95% CI −0.30 to 0.07); non-HDL cholesterol reduced in UK PRECISE (−0.20 mmol/L, 95% CI −0.41 to 0.00); HDL unchanged; LDL reduction in one trial but not significant. Blood pressure effects not evident. Conclusion: The limited trial evidence does not support selenium supplements for the primary prevention of CVD, especially in selenium-replete populations; potential benefits may exist only in subgroups with low selenium status; more research is needed, particularly in diverse or deficient populations.

Results dominated by two large US trials (NPC and SELECT); predominantly male and selenium-replete populations; wide variation in dose, form, and duration; many outcomes were secondary endpoints; limited representation of women and of populations with low selenium intake; generalizability uncertain.