In the 1970s, studies suggested an association between maternal diabetes, including gestational diabetes mellitus (GDM), and neonatal respiratory distress syndrome (RDS). Robert et al reported that the risk of RDS in neonates of mothers with diabetes (cohort included women with both GDM and preexisting diabetes) was 23.7 times greater than in neonates born to mothers without diabetes (P<.001).1 This increased risk was postulated to be the result of delayed fetal lung maturation. Women with GDM or class A diabetes, as well as class B and C diabetes, achieved lecithin to sphingomyelin ratios greater than 2.0 in their amniotic fluid, a marker of fetal lung maturation, 1–5 weeks later in gestation than their euglycemic counterparts.2 Since those initial studies, some studies have suggested that GDM is associated with an increased risk of RDS whereas other studies have found no association with poor neonatal respiratory outcomes. In 1989, a prospective cohort study of 2,204 Californian women, 97 of whom had GDM, found no difference in the rates of RDS in neonates born to women with GDM compared with neonates born to women without GDM.3 More recently, a prospective Israeli study demonstrated that, among neonates born prematurely, there was no increased neonatal respiratory morbidity in the offspring of women whose pregnancies had been complicated by GDM.4 These studies stand in contrast to two French retrospective data studies of hundreds of thousands of women that suggest that GDM is associated with increased odds of RDS (odds ratio 1.1, 95% CI 1.0–1.3 and odds ratio 1.2, 95% CI 1.1–1.3).5,6
Because the available data are inconsistent, we sought to evaluate neonatal respiratory morbidity in pregnancies with and without GDM in a contemporary cohort of US pregnancies in which the respiratory neonatal outcomes were collected prospectively. We chose to focus on pregnancies at high risk for late preterm delivery because late preterm birth is associated with significantly more neonatal respiratory morbidity than term birth.
This is a secondary analysis of the Antenatal Late Preterm Delivery Steroids study, a multi-centered, randomized, controlled trial that tested the effectiveness of betamethasone compared with placebo to reduce neonatal morbidity in women at high risk for late preterm delivery.7 The study was performed in 17 centers across the United States in which 2,831 singleton pregnancies at high risk for delivery between 34 0/7 and 36 5/7 weeks of gestation were enrolled. The parent protocol was approved by the Institutional Review Board at all participating sites and written informed consent was obtained from all participants. Women were considered to have had GDM if they were given the diagnosis at any point during their pregnancy before delivery. Women with preexisting diabetes mellitus were excluded from the parent trial. Information about GDM diagnostic criteria or treatment was not collected.
We compared pregnancy outcomes between those diagnosed with GDM (exposed pregnancies) and those without a GDM diagnosis (unexposed). The primary outcome was a composite of neonatal respiratory morbidity identical to that of the primary study. The composite included significant respiratory morbidity (continuous positive airway pressure or high-flow nasal cannula for at least 2 hours, supplemental oxygen with a fraction of inspired oxygen of at least 0.30 for at least 4 hours, extracorporeal membrane oxygenation, or mechanical ventilation), stillbirth, or neonatal death within the first 72 hours of life. Secondary outcomes included severe neonatal respiratory complications, neonatal intensive care unit (NICU) admission lasting longer than 3 days, and neonatal hyperbilirubinemia. Severe neonatal respiratory complications were defined as any of the following: continuous positive airway pressure or high-flow nasal cannula for at least 12 continuous hours, supplemental oxygen with a fraction of inspired oxygen of at least 0.30 for at least 24 continuous hours, extracorporeal membrane oxygenation, mechanical ventilation, stillbirth or neonatal death within the first 72 hours of life. Hyperbilirubinemia was considered present if the peak total neonatal bilirubin was at least 15 mg or if the neonate was treated with phototherapy.
Significant additional maternal and neonatal data were collected on each participant. These data were collected by local research staff members at each site but were reviewed centrally for consistency. Data were also available regarding study group assignment (betamethasone or placebo). All study staff were blinded to the original study intervention, betamethasone or placebo. Study staff, including the neonatal care team, were aware of a woman's GDM status.
The relative risk of the primary outcome was assessed using modified Poisson regression.8 Relative risks were adjusted for covariates that differed significantly in univariable analysis (P<.05) between pregnancies complicated by GDM and those that were not. Models incorporating a product interaction term between GDM status and treatment arm (betamethasone or placebo) were also evaluated. Analyses were conducted using the SAS System version 9.3.
A power calculation, after the primary study but before performing this secondary analysis, was performed and revealed 80% power to detect a 38% difference between neonatal respiratory morbidity in pregnancies exposed to GDM and those unexposed. The Antenatal Late Preterm Delivery Steroids trial had a cumulative respiratory morbidity incidence of 13% and so we assumed a baseline respiratory morbidity incidence of 13% in most of those pregnancies that were not complicated by GDM. We anticipated being able to detect a difference if the rate of respiratory morbidity in pregnancies complicated by GDM was 18% or greater.
Gestational diabetes mellitus data were collected on all 2,831 participants in the parent trial who delivered between October 2010 and February 2015.7 In total, 306 of the 2,831 (10.8%) pregnancies were complicated by GDM. Among pregnancies with GDM, an equal number of women were randomized to betamethasone and placebo. Compared with women whose pregnancies were not complicated by GDM, women with GDM were older, more likely to be parous, and more often developed hypertensive disorders of pregnancy (Table 1). At the time of study enrollment, women with GDM were less likely to have preterm labor or premature rupture of membranes compared with women without GDM. There was no difference between women with GDM and those without regarding race, gestational age at randomization, or study group assignment. The mean gestational age at delivery was similar for women with and without GDM (36.1 weeks).
Composite respiratory morbidities were not significantly different among neonates born to women with GDM compared with those born to women without GDM (12.1% vs 13.1%; adjusted RR 0.84, 95% CI 0.61–1.17) nor were severe respiratory complications or prolonged NICU admission (Table 2). Neonates born to women with GDM were more likely to have hyperbilirubinemia (15.0% vs 10.3%, P=.04). This association persisted after adjustment for age, parity, and hypertensive disorders of pregnancy (relative risk 1.39 [95% CI 1.03–1.88]). The occurrence of cesarean delivery among women with GDM and those without was 37% compared with 31%, P=.02. When we included mode of delivery in the model, GDM still was not associated with an increased risk of RDS (0.81; 95% CI 0.56–1.18). We also did not include indication for trial entry in the principle adjusted model because these diagnoses occurred temporally after the diagnosis of GDM and likely represent modifiers rather than confounders. However, as a precaution, we performed additional models in which they were included and no significant difference in results was found. Finally, there were no significant differences in the association between GDM and the composite and secondary outcomes by initial study intervention arm (betamethasone or placebo) (P>.05 for interaction for all comparisons).
The original Antenatal Late Preterm Delivery Steroids trial demonstrated that for pregnancies at high risk of late preterm delivery, antenatal steroids reduced neonatal respiratory complications. In this secondary analysis, we found that pregnancies complicated by GDM at high risk for late preterm delivery had similar neonatal respiratory outcomes as pregnancies without GDM. Gestational diabetes mellitus was not associated with an increased risk of the composite neonatal respiratory morbidity, nor was it associated with prolonged NICU admission or severe neonatal respiratory complications. The equivalent neonatal respiratory morbidity frequencies suggest that women with GDM delivering in the late preterm interval should be managed under the same guidelines as those without GDM. The original study suggested that 25 women at risk for late preterm labor must be treated with antenatal steroids to prevent one case of neonatal respiratory morbidity. Our findings suggest that the number needed to treat in women with GDM is similar to that required in women without GDM.
The strengths of this study lie primarily with the large sample size as well as the detailed definition of the outcome data before initiation of the parent study. Trained research nurses collected the data, which was regularly and iteratively audited for completeness and accuracy. However, the parent study did not include women with preexisting diabetes, only those with GDM. Another limitation was the study design not examining GDM as an independent variable or collecting data about each participant's glucose control, diabetes treatment, or timing of diagnosis. Therefore, we cannot perform any subanalyses assessing GDM severity and risk of respiratory morbidities. It is quite possible that the degree of dysglycemia affects neonatal respiratory morbidity and this study was unable to detect those differences. Based on the data available, it is likely that the women in this study represent a typical cross-section of women with GDM; many needed no medication to achieve euglycemia, some used insulin and some used oral hypoglycemic. Women with GDM in this study population had an increased frequency of common risk factors for GDM (age, parity, and hypertensive disorders) compared with the non-GDM group.3,9 Neonates born to women with GDM in this cohort also had an increased risk of neonatal hyperbilirubinemia compared with neonates born to women without GDM, a well-established and consistent neonatal complication of maternal hyperglycemia.10 This study was also limited by a predetermined sample size as it was a secondary analysis and therefore we may not have been powered to detect differences in all confounders (eg, racial–ethnic differences). The final limitation of this study is that neonatal care providers were not blinded to the women's GDM status, which could result in bias, but would likely be in favor of interventions, such as supplemental O2, in the GDM-affected neonates.
Currently in the United States, 7.6% of pregnancies are complicated by GDM.11 As the burden of GDM grows, it is vital that clinical management be guided by high quality evidence to optimize maternal and neonatal outcomes. In this contemporary cohort, GDM was not associated with a clinically significant difference in neonatal respiratory outcomes. This suggests that evidence-based interventions known to reduce morbidity (eg, induction at 37 weeks for preeclampsia without severe features and induction at 39 weeks for low-risk women), should be considered in all women regardless of their GDM status.
1. Robert MF, Neff RK, Hubbell JP, Taeusch HW, Avery ME. Association between maternal diabetes and the respiratory-distress syndrome in the newborn. N Eng J Med 1976;294:357–60.
2. Gluck L, Kulovich MV. Lecithin/sphingomyelin ratios in amniotic fluid in normal and abnormal pregnancies. Am J Obstet Gynecol 1973;115:64–70.
3. Jacobson JD, Cousing L. A population-based study of maternal and perinatal outcome in patients with gestational diabetes. Am J Obstet Gynecol 1989;161:981–6.
4. Bental Y, Reichman B, Shiff Y, Weisbrod M, Boyko V, Lerner-Geva L, et al. Impact of maternal diabetes mellitus on mortality and morbidity of preterm infants. Pediatrics 2011;128:848–55.
5. Billionnet C, Mitanchez D, Weill A, Nizard J, Alla F, Hartemann A, et al. Gestational diabetes and adverse perinatal outcomes from 716,152 births in France in 2012. Diabetologia 2017;60:636–44.
6. Mitanchez D, Yzydorczyk C, Simeoni U. What neonatal complications should the pediatrician be aware of in case of maternal gestational diabetes? World J Diabetes 2015;6:734–43.
7. Gyamfi-Bannerman C, Thom EA, Blackwell A, Tita AT, Reddy UM, Saade GR, et al. Antenatal betamethasone for women at risk for late preterm delivery. N Engl J Med 2016;374:1311–20.
8. Zou G. A modified Poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004;159:702–6.
9. Landon MB, Spong CY, Thom E, Carpenter MW, Ramin SM, Casey B, et al. A multi-center, randomized trial of treatment for mild gestational diabetes. N Engl J Med 2009;361:1339–48.
10. The HAPO Study Cooperative Research Group. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med 2008;358:1991–2002.
11. Stark Casagrande S, Linder B, Cowie CC. Prevalence of gestational diabetes and subsequent type 2 diabetes among U.S. women. Diabetes Res Clin Pract 2018;141:200–8.