This multicenter retrospective cohort study evaluated whether associations between gestational diabetes mellitus (GDM) and neonatal and long-term offspring outcomes differ by plurality. The investigators noted that diagnostic thresholds and risk stratification for GDM are largely extrapolated from singleton pregnancies and that evidence about GDM effects in multifetal pregnancies is limited. The study aimed to compare short-term neonatal complications and longer-term healthcare utilization among offspring from multifetal versus singleton pregnancies complicated by GDM, and to assess whether plurality or gestational age at delivery modified those associations.
The study used electronic medical records from six university-affiliated medical centers in Israel over the period 1 January 2010 to 30 June 2024. Inclusion was all live births at 22 weeks’ gestation or later or with birthweight greater than 500 g. The analytic sample included 134,517 pregnancies resulting in 136,579 live-born offspring to 104,023 mothers.
Four exposure groups were defined by plurality and GDM status using ICD-9 coding (648.8x): singleton without GDM, singleton with GDM, multifetal without GDM, and multifetal with GDM. Short-term neonatal complications assessed included neonatal hypoglycemia, neonatal intensive care unit (NICU) admission, respiratory distress, intraventricular hemorrhage, and growth abnormalities. Long-term offspring outcomes were evaluated indirectly by measuring hospital-based health-care utilization across specialties such as neurology, ophthalmology, and cardiology.
Multivariable Firth logistic regression and ordinary least squares regression were used. Models adjusted for maternal age, body mass index (BMI), parity, and fertility treatment. Additional models included adjustment for gestational age. The analysis incorporated false discovery rate correction for multiple comparisons and specifically tested GDM-by-plurality interaction terms to evaluate effect modification by plurality.
Overall, GDM complicated 10,605 singleton pregnancies (7.9% of singletons) and 256 multifetal pregnancies (0.2% of multifetal pregnancies) in the cohort. Key outcome differences by plurality included:
In multifetal pregnancies, GDM was not associated with increased short-term neonatal complications. Specifically, GDM in multifetal gestations was associated with lower adjusted odds of neonatal hypoglycemia (adjusted odds ratio [aOR] 0.30; 95% CI 0.18–0.49).
Additionally, multifetal pregnancies with GDM showed lower adjusted odds of later specialty healthcare utilization in neurology (aOR 0.37; 95% CI 0.24–0.59) and ophthalmology (aOR 0.49; 95% CI 0.31–0.77).
By contrast, GDM in singleton pregnancies was associated with increased neonatal morbidity (details of specific effect sizes for singletons beyond the cited contrasts were not reported in the abstract).
When gestational age was added to models, associations for NICU admission were attenuated, indicating part of the relationship between GDM and NICU admission was mediated or confounded by gestational age at delivery. However, adjustment for gestational age did not fully account for the observed interactions between GDM and plurality for neonatal hypoglycemia or selected long-term outcomes. The authors applied GDM-by-plurality interaction terms to formally test effect modification and used false discovery rate correction to address multiple testing.
The findings indicate that the association between GDM and both neonatal and longer-term offspring outcomes differs by plurality. In this cohort, GDM in multifetal pregnancies was not linked to increased short-term neonatal complications and was associated with lower odds of neonatal hypoglycemia and reduced later neurology and ophthalmology healthcare utilization, whereas GDM in singleton pregnancies was associated with higher neonatal morbidity. These results suggest that risk stratification, diagnostic thresholds, and counseling practices derived from singleton-pregnancy data may not fully apply to multifetal pregnancies.
Clinicians and guideline developers should be cautious when extrapolating singleton-based GDM risk estimates to multifetal gestations. The study underscores the need to consider plurality and gestational age when assessing GDM-associated perinatal risk and when designing follow-up strategies for offspring.
The abstract reports the study design, sample size, exposures, outcomes, and main adjusted associations, but does not provide detailed information on several points in the abstract: methods for GDM ascertainment beyond ICD-9 coding, diagnostic criteria used across centers, glycemic control or treatment strategies for GDM, neonatal management protocols, specific absolute risks for outcomes, and full effect estimates for singleton pregnancies. The study is retrospective and based on hospital electronic records, which may limit available covariates and introduce misclassification. Further details on these limitations and on subgroup analyses would be found in the full text but were not reported in the abstract.
In this large multicenter cohort of over 134,000 pregnancies, associations between gestational diabetes mellitus and neonatal and offspring healthcare outcomes differed by plurality. GDM in multifetal pregnancies was not associated with increased short-term neonatal complications and was linked to lower adjusted odds of neonatal hypoglycemia and select specialty healthcare utilization, whereas GDM in singleton pregnancies was associated with increased neonatal morbidity. The authors conclude that risk stratification strategies derived from singleton pregnancies may not fully capture the risk profile in multifetal pregnancies, implying the need for plurality-specific consideration in diagnosis, management, and counseling for GDM.