Design

This study was a secondary analysis of data collected in a study by the Eunice Kennedy Shriver National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network, where 495 women at risk for preterm delivery were randomly assigned to a single course vs repeat weekly courses of antenatal corticosteroids.

Inclusion criteria for the primary study included having an entry gestational age between 23 ^0/7 and 31 ^6/7 weeks, intact amniotic membranes, being at risk for preterm delivery, and having received a single full course of antenatal corticosteroids between 6–10 days before enrollment (ie, 2 intramuscular injections of 12 mg of betamethasone 24 hours apart or 4 intramuscular injections of dexamethasone 12 hours apart). The primary study exclusion criteria included documented fetal lung maturity, chorioamnionitis, nonreassuring fetal monitoring, active preterm labor (cervical dilation ≥5 cm or ≥6 contractions per hour), insulin-requiring diabetes at study enrollment, and concurrent use of systemic corticosteroids. In addition to these criteria, this analysis focused on women with a singleton pregnancy who received only 1 course of antenatal corticosteroids (ie, the placebo arm of the study).

Given that this study was using an existing deidentified database, it was considered exempt from full review by the institutional review board at Oregon Health & Science University.

Data

Maternal data regarding age (<35 years or ≥35 years), race (white or other), prepregnancy weight, and height were extracted from the original dataset. Maternal prepregnancy BMI was calculated using the formula BMI = prepregnancy weight in kilograms/(height in meters). Patients were initially categorized into the BMI categories of underweight (BMI < 18.5), normal weight (18.5 ≤ BMI <25), overweight (25 ≤ BMI <30), obese (30 ≤ BMI <40), and morbidly obese (BMI ≥40), but secondary to limited numbers of subjects in certain BMI categories (14, 82, 48, 33, and 6, respectively), the decision was made to categorize the groups into BMI <25 and BMI ≥25.

Data regarding gestational age at delivery and birthweight were collected in addition to neonatal morbidity and mortality data, which were used as outcomes of interest. The primary outcome was a composite outcome, including neonatal respiratory distress syndrome, bronchopulmonary dysplasia, chronic lung disease, IVH, periventricular leukomalacia, necrotizing enterocolitis, retinopathy of prematurity, sepsis, and stillbirth or neonatal death. These individual neonatal outcomes, which were defined previously in reports of the study, were considered as secondary outcomes.

Statistical analysis

Analyses were completed using SAS (SAS Institute, Cary, NC). Statistical analysis for continuous variables used descriptive statistics as well as the Wilcoxon rank-sum test. Categorical variables were analyzed with Fisher exact test; exact confidence intervals were also constructed for odds ratios. Multivariable logistic regression was then used to estimate the effect of prepregnancy maternal BMI on the primary outcome, while adjusting for maternal age, maternal race, gestational age at delivery, birthweight, and gestational diabetes. The analysis was performed with prepregnancy maternal BMI as a categorical and continuous variable.

Results were presented as odds ratios (ORs), 95% confidence intervals (CIs), and P values. A result was considered to have statistical significance if the 95% CI did not include 1.0 or with a P value less than .05. No adjustments for multiple comparisons were made.