Association between circulating essential metals and pregnancy loss: A systematic review and meta-analysis.
Abstract
Objective
To systematically compare circulating concentrations of iron (Fe), magnesium (Mg), manganese (Mn), and selenium (Se) between women with spontaneous pregnancy loss (SPL) or recurrent pregnancy loss (RPL) and pregnant controls.
Methods
We searched PubMed, Embase, the Cochrane Library, and Web of Science from inception to 25th May 2025 for observational studies comparing circulating metal concentrations between women with SPL/RPL and pregnant controls. Standardized mean differences (SMDs) and 95% confidence intervals (CIs) were pooled using random-effects models. Subgroup, sensitivity, and influence analyses were performed. Meta-regression and assessments of small-study effects were considered exploratory. The certainty of evidence was assessed using GRADE.
Results
Eighteen studies were included (10 on Fe, 7 on Mg, 9 on Mn, and 8 on Se), encompassing 879 women with pregnancy loss (647 with SPL and 232 with RPL) and 2355 pregnant controls from 11 countries. No statistically significant overall differences in circulating metal concentrations were identified between women with pregnancy loss and pregnant controls. The pooled SMD was -0.05 for Fe (95% CI: -0.44-0.34; I² = 91.3%), -0.38 for Mg (95% CI: -1.02-0.27; I² = 93.5%), 0.17 for Mn (95% CI: -0.38-0.72; I² = 93.6%), and 0.29 for Se (95% CI: -0.73-1.32; I² = 97.2%). Findings were also not statistically significant within the SPL and RPL subgroups. Substantial heterogeneity persisted despite subgroup and exploratory meta-regression analyses. Influence and sensitivity analyses indicated limited robustness for Mg, while the Se estimate remained highly imprecise and heterogeneous. Small-study effect assessment was inconclusive because of the limited number of studies. GRADE assessment rated the certainty of evidence as very low for all metal-outcome comparisons.
Conclusion
Available observational evidence does not demonstrate consistent differences in circulating Fe, Mg, Mn, or Se concentrations between women with SPL or RPL and pregnant controls. However, the evidence remains inconclusive because of substantial heterogeneity, possible reverse causality, residual confounding, variation in outcome and control-group definitions, and very low certainty of evidence. Well-designed prospective studies using standardized outcome definitions and pre-event metal measurements are needed. These findings should not be interpreted as evidence for or against routine essential-metal supplementation to prevent pregnancy loss.