Diabetes and anti-diabetic interventions and the risk of gynaecological and obstetric morbidity: an umbrella review of the literature

Background: Diabetes has reached epidemic proportions in recent years with serious health ramifications. The aim of this study was to evaluate the strength and validity of associations between diabetes and anti-diabetic interventions and the risk of any type of gynaecological or obstetric conditions. Methods: Design: Umbrella review of systematic reviews and meta-analyses. Data sources: PubMed, Medline, Embase, Cochrane Database of Systematic Reviews, manual screening of references. Eligibility criteria: Systematic reviews and meta-analyses of observational and interventional studies investigating the relationship between diabetes and anti-diabetic interventions with gynaecological or obstetric outcomes. Meta-analyses that did not include complete data from individual studies, such as relative risk, 95% confidence intervals, number of cases/controls, or total population were excluded. Data analysis: The evidence from meta-analyses of observational studies was graded as strong, highly suggestive, suggestive or weak according to criteria comprising the random effects estimate of meta-analyses and their largest study, the number of cases, 95% prediction intervals, I2 heterogeneity index between studies, excess significance bias, small study effect and sensitivity analysis using credibility ceilings. Interventional meta-analyses of randomised controlled trials were assessed separately based on the statistical significance of reported associations, the risk of bias and quality of evidence (GRADE) of included meta-analyses. Results: A total of 117 meta-analyses of observational cohort studies and 200 meta-analyses of randomised clinical trials that evaluated 317 outcomes were included. Strong or highly suggestive evidence only supported a positive association between gestational diabetes and caesarean section, large for gestational age babies, major congenital malformations and heart defects and an inverse relationship between metformin use and ovarian cancer incidence. Only a fifth of the randomised controlled trials ...