Gestational Diabetes and Neonatal Hypoglycemia Risk Post-Betamethasone in the Late Preterm Population: A Secondary Analysis of the ALPS Trial

Background: Neonatal hypoglycemia is a significant clinical concern, particularly in infants born to mothers with gestational diabetes mellitus (GDM). Betamethasone, administered antenatally to enhance fetal lung maturity in late preterm pregnancies, induces transient maternal hyperglycemia, potentially exacerbating the neonatal insulin response and subsequent risk of hypoglycemia. Corticosteroids increase maternal blood glucose levels by increasing hepatic gluconeogenesis, inhibiting glucose uptake in adipose tissue, and antagonizing insulin synthesis (Schäcke et al., Pharmacology & Therapeutics 2002). Elevated maternal blood glucose stimulates fetal pancreatic insulin secretion, predisposing neonates to hypoglycemia after birth (Arimitsu et al., Endocrine Journal 2023). The original Antenatal Late Preterm Steroids (ALPS) Trial reported reduced neonatal respiratory morbidity but noted increased neonatal hypoglycemia incidence with antenatal betamethasone administration. However, whether gestational diabetes specifically modifies this hypoglycemia risk after betamethasone exposure remains uncertain. Objective: To determine whether gestational diabetes is associated with increased incidence of neonatal hypoglycemia following antenatal betamethasone administration during the late preterm period. Methods: This is a secondary analysis of the ALPS Trial, a randomized, multicenter study examining neonatal outcomes following antenatal betamethasone administration between 34 weeks 0 days and 36 weeks 5 days’ gestation. Inclusion criteria were limited to participants from the active treatment arm randomized to receive two betamethasone doses, 24 hours apart (initial N = 1,428). Participants lacking essential data, including prepregnancy weight (n = 36), neonatal hypoglycemia status (n = 2), maternal ethnicity (n = 3), and second dose administration status (n = 1), were excluded, leaving a final analyzed cohort of N = 1,386. The primary outcome was neonatal hypoglycemia, defined as blood glucose levels below 40 mg/dL at any time after birth. Information about GDM diagnostic criteria or treatment was not available. Notably, patients with GDM were originally excluded from the ALPS Trial due to concerns about the potential for unblinding with corticosteroids. Eligibility criteria were later amended to enroll patients with gestational diabetes after the trial was underway. Pre-gestational diabetes was an original exclusion criterion of the trial and remained so for the duration. Baseline clinical and maternal characteristics were compared using chi-square tests for categorical variables. Logistic regression was performed to account for confounding factors on univariate analysis: mode of delivery, race or ethnicity, primary source of medical payment for prenatal care, and indication for trial entry. The interval between randomization and delivery was analyzed to determine if subgroup analyses for betamethasone doses were warranted. Analyses for both the ALPS Trial and this secondary analysis followed intention-to-treat principles. Subgroup analysis of participants who received 1 or 2 doses of betamethasone was performed as-treated. Relative risks (RRs) with 95% confidence intervals (CIs) were calculated for unadjusted group comparisons. Adjusted odds ratios (aORs), accounting for statistically significant baseline differences (p < 0.05), are reported for outcomes of interest. Results: Hypoglycemia occurred in 28.7% of neonates born to mothers with GDM and 23.4% of those born to mothers without GDM (p = 0.1422). Compared to euglycemic patients, those with GDM were more likely to have prepregnancy obesity, be Hispanic or Asian, be aged 35 or older, and have late preterm delivery due to rupture of membranes, preeclampsia or gestational hypertension, or other indications. Notably, in the parent trial, only 60.5% (n = 839) of participants received the prespecified two doses of betamethasone. Unadjusted subgroup analysis revealed nearly equal odds of hypoglycemia among neonates born to mothers with GDM who received two doses. However, women with GDM receiving only one dose of betamethasone had 82% increased odds of having a hypoglycemic neonate (OR 1.82; 95% CI: 1.06–3.14; p = 0.0303). This significant result may relate to timing, with 43% delivering the same day as randomization and 96% delivering within one day, corresponding closely with peak maternal hyperglycemia post-betamethasone (Itoh et al. Endocrine Journal 2016). Multiple logistic regression analyses showed the adjusted odds ratio (aOR) for neonatal hypoglycemia among infants born to mothers with GDM randomized to two doses was 1.132 (95% CI: 0.75–1.68). Among infants whose mothers received a single dose, the aOR was 1.540 but was not statistically significant (95% CI: 0.83–2.79). A larger trial may be needed to detect a true effect. Conversely, mothers who received both doses had similar odds for neonatal hypoglycemia (aOR 0.8658; 95% CI: 0.48–1.50). Conclusion: Among women receiving antenatal corticosteroids in the late preterm period, maternal GDM was not associated with an increased risk for neonatal hypoglycemia compared with euglycemic patients. These results support the practice of administering antenatal steroids in late preterm pregnancies complicated by GDM. A large randomized clinical trial is warranted to determine whether GDM patients receiving antenatal late preterm steroids are at increased risk for neonatal hypoglycemia.