The Role of Nutrient Supplements in Female Infertility: An Umbrella Review and Hierarchical Evidence Synthesis.
Level 1 - systematic review of randomized trials
Systematic review and umbrella review of meta-analyses of randomized controlled trials
PubMed 39796491 · doi:10.3390/nu17010057
What was done
Authors conducted an umbrella review searching five electronic databases (August 2017 to January 2024) for systematic reviews, meta-analyses, and umbrella reviews of randomized controlled trials examining nutrient supplements in women with infertility attempting spontaneous conception or medically assisted reproduction (MAR). Primary outcomes were live birth, clinical pregnancy, and biochemical pregnancy rates; secondary outcomes included miscarriage, ectopic pregnancy, and multiple pregnancy. Evidence quality was assessed via AMSTAR 2 and GRADE.
What was found
Four meta-analyses were included. Multiple micronutrients (OR 2.59) and antioxidants (OR 1.81) were associated with increased live birth rates versus control, both with very low certainty evidence. Clinical pregnancy rates in women with PCOS and/or undergoing MAR were increased with L-carnitine (OR 11.14), CoQ10 (OR 2.49), melatonin (OR 1.66), myo-inositol (RR 1.52), N-acetylcysteine (OR 2.15), and vitamin D (OR 1.49), all with very low certainty. Vitamin D did not increase biochemical pregnancy rates (very low certainty). NAC, vitamin D, and pooled antioxidants showed no effect on miscarriage or multiple pregnancy rates (low to very low certainty), and pooled antioxidants showed no effect on ectopic pregnancy rates (low certainty).
Why it matters
Despite widespread clinical use and popular interest, the current randomized trial evidence is insufficient to recommend nutrient supplementation for female infertility, though no evidence of significant harm was identified.
Limits
Only four meta-analyses were eligible for inclusion. All efficacy outcomes were graded as very low certainty by GRADE. Specific supplement dosages, formulations, durations, and participant sample sizes are not reported in the abstract.