Aims: To examine the associations of serum concentrations of adiponectin and leptin and leptin/adiponectin ratio (LAR) in early pregnancy with risk of gestational diabetes mellitus (GDM) in Chinese women. The predictive ability of those biomarkers for GDM was also assessed.
Methods: Within the Tongji-Shuangliu Birth Cohort, a nested case-control study was established with 332 GDM cases and 664 matched controls at 1:2 ratio on age (±3 years) and gestational age (±4 weeks). Serum adiponectin and leptin levels were measured in early pregnancy (median gestational week, 11; range, 6-15). Conditional logistic regression models with adjustment for potential covariates were used to evaluate the associations.
Results: Multivariable-adjusted odds ratios (ORs) comparing extreme quartiles of adiponectin, leptin and LAR were 0.55 (95 % CI, 0.35, 0.85), 1.96 (95 % CI, 1.19, 3.24), and 2.72 (95 % CI, 1.63, 4.54) for GDM, respectively (All P-trend < 0.02). Adding adiponectin and leptin to a conventional prediction model (including traditional risk factors and fasting glucose) increased the C-statistics from 0.708 (95 % CI, 0.674, 0.741) to 0.728 (95 % CI, 0.695, 0.760), and achieved a net reclassification improvement of 0.292.
Conclusions: Our findings indicate that adiponectin is inversely associated with GDM, while leptin and LAR are positively associated with GDM in Chinese pregnant women.
Keywords: Adiponectin; Association; Gestational diabetes mellitus; Leptin; Nested case-control study.
Copyright © 2022 Elsevier B.V. All rights reserved.