Late-onset diagnosis of gestational diabetes after normal mid-pregnancy screening in women with large for gestational age or polyhydramnios: a systematic review and meta-analysis.

Abstract

To determine the detection rate of late-onset gestational diabetes mellitus and associated perinatal outcomes in women with normal mid-pregnancy glucose screening who undergo repeat oral glucose tolerance test due to suspected large-for-gestational-age fetuses or polyhydramnios. We systematically searched PubMed/MEDLINE, Embase, Web of Science, and the Cochrane Library for studies published from January 2010 to November 2024. We included cohort studies reporting late gestational diabetes mellitus detection rates in women with documented normal mid-pregnancy glucose testing (negative glucose challenge test or normal 75 g oral glucose tolerance test at 24-28 weeks) who underwent repeat oral glucose tolerance test after 28 weeks due to sonographic findings of suspected large for gestational age or polyhydramnios. We excluded studies of women with preexisting diabetes, those lacking documented normal mid-pregnancy screening, and those examining routine late testing without specific clinical indication. Study quality was assessed using the Newcastle-Ottawa Scale. Pooled detection rates with 95% confidence intervals were calculated using random-effects meta-analysis. Subgroup analyses were performed by clinical indication, timing of testing, initial screening method (2-step vs 1-step), and maternal body mass index. Perinatal outcomes were compared using pooled odds ratios calculated with the Mantel-Haenszel method. Six cohort studies including 2166 women met inclusion criteria. The pooled detection rate was 15.0% (95% confidence interval, 9.9%-21.0%; I²=91%). Detection rates varied significantly by indication: suspected large for gestational age 20.8% (95% confidence interval, 17.4%-24.6%) vs isolated polyhydramnios 4.8% (95% confidence interval, 2.0%-10.8%). Women with late-onset gestational diabetes mellitus had significantly higher rates of neonatal hypoglycemia (odds ratio, 1.82; 95% confidence interval, 1.18-2.81), overall cesarean delivery (odds ratio, 2.00; 95% confidence interval, 1.47-2.72), elective cesarean for large for gestational age/macrosomia (odds ratio, 3.37; 95% confidence interval, 2.01-5.64), emergent cesarean (odds ratio, 1.64; 95% confidence interval, 1.05-2.57), and induction of labor (odds ratio, 2.27; 95% confidence interval, 1.32-3.89). Macrosomia by birthweight and large for gestational age at delivery were not significantly elevated. Late oral glucose tolerance test detects gestational diabetes mellitus in approximately one in 6 women with normal mid-pregnancy screening who develop large for gestational age or polyhydramnios. Late-onset gestational diabetes mellitus is associated with significantly increased neonatal hypoglycemia and cesarean delivery, with the largest effect for elective cesarean for suspected large for gestational age/macrosomia. These findings may inform clinical decision-making regarding repeat glucose testing in the third trimester.