Prematurity and size at birth and the risk of gestational diabetes mellitus
The median GDM rate in the control groups of the included cohort studies was 2.9% (range: 0.5% to 22%). Figure 2 presents a forest plot summarising the studies that assessed the association between preterm birth or size at birth with GDM. Premature birth, LBW, and SGA status were associated with a higher GDM risk (pooled OR, 1.84; 95% CI: 1.54 to 2.20; I2 = 78.3%; τ2 = 0.07). Supplementary Table S5 summarises the absolute risk difference in pregnant women born with LBW, SGA status, or born preterm in the low- (2.0% risk of GDM in the control group), medium- (10%), and high- (20%) GDM prevalence groups. The absolute risk increases were 1.6% (95% CI: 1.0 to 2.1%), 7.0% (95% CI: 4.6 to 9.6%), and 11.5% (95% CI: 7.8 to 15.5%) in low-, moderate-, and high-prevalence settings, respectively. The certainty of evidence was low due to serious concerns of risk of bias and publication bias.
Figure 3 presents study estimates in a funnel plot. The plot appeared asymmetrical, and Egger’s test for funnel plot asymmetry was statistically significant (p -value = 0.030). Supplementary Figure S2 shows the contour-enhanced funnel plot, which suggests the existence of some missing studies on the left-hand side of the plot; these studies would have yielded statistically non-significant findings.
Data on the birth weight of mothers with or without GDM, obtained from three studies excluded from the meta-analysis, are presented in Supplementary Table S5. These studies consistently reported that mothers with GDM were born with lower birth weights than those without GDM.