The association between serum selenium level and gestational diabetes mellitus: A systematic review and meta-analysis.

Abstract

The relationship between serum selenium level and gestational diabetes mellitus (GDM) is controversial. The purpose is to update and summarize previous studies to understand the relationship in more detail. PubMed, The Cochrane Library, EMBASE, Web of science, CNKI, WANFANG DATA and Cqvip were searched for studies published up to 3 September 2021. The random-effects model was used to measure the combined estimation. The overall effect was reported in a standard mean difference (SMD) and 95% confidence interval (95% CI). All data were analysed by Review Manager 5.4. Twenty-seven studies involving 1588 patients with GDM and 2450 healthy pregnant women contributed to this meta-analysis. Selenium level was significantly lower in women with GDM than those without GDM (SMD=-1.29; 95% CI: -1.60 to -0.97, p<0.00001). Subgroup analyses showed that such trend was consistent within the non-European population (Asia: SMD=-1.44; 95% CI: -1.79 to -1.08, p<0.00001; Africa: SMD=-2.62; 95% CI: -4.50 to -0.74, p=0.006) and in the second and third trimesters (the second trimester: SMD=-1.41; 95% CI: -1.82 to -0.99, p<0.00001; the third trimester: SMD=-1.54; 95% CI: -2.09 to -0.98, p<0.00001), but not within the European population (SMD=-0.47; 95% CI: -1.09 to 0.16, p=0.14) or in the first trimester (SMD=-0.52; 95% CI: -1.13 to 0.10, p=0.10). This meta-analysis showed that the serum selenium level of patients with GDM was lower than that in healthy pregnant women, especially within the non-European population and in the second and third trimesters.