Serum level and polymorphisms of retinol-binding protein-4 and risk for gestational diabetes mellitus: a meta-analysis.

BACKGROUND Retinol-binding protein-4 (RBP4) has been reported to be potentially involved in the pathogenesis of gestational diabetes mellitus (GDM); however, the findings are inconsistent. Our aims were to review the studies that investigated the association of serum levels and polymorphisms of RBP4 with GDM risk, and to provide recommendations for future research. METHODS The databases PubMed, EBSCO, ScienceDirect, and Web of Knowledge were searched up to October 2015 to find out studies evaluating the relationship between serum RBP4 level/ RBP4 polymorphisms and GDM risk. In the meta-analysis of serum RBP4 levels the key inclusion was that studies were designed as BMI-matched studies or had observed non-significant differences in BMI between cases and controls. RESULTS Fourteen case-control studies (647 cases and 620 controls) reporting the association between serum RBP4 level and GDM risk, and three studies (1012 cases and 1605 controls) investigating the association between RBP4 polymorphisms and GDM risk were involved. Our results showed that high serum RBP4 levels represent a risk factor for GDM (pooled standardized mean difference =0.758, 95% confidence interval [0.387, 1.128]). The results of subgroup analyses based on "gestational age at blood sampling" or "diagnostic criteria" are consistent with the overall results. However, the postpartum subgroup and "before 24 weeks" subgroup both only include one article and indicate no association between serum RBP4 level and GDM risk. The meta-analysis on the association between rs3758539 polymorphism and GDM risk shows that RBP4 rs3758539 polymorphism is not associated with the development of GDM. CONCLUSIONS The results of this meta-analysis support the hypothesis that RBP4 is a modest independent risk factor for GDM (i.e., nonobese patients with GDM might express RBP4 at abnormal levels). The serum RBP4 level is associated with the risk of GDM. However, the association in the first-trimester and postpartum period should be validated by further research. The association between RBP4 rs3758539 polymorphism and GDM risk was not confirmed.