Materials and Methods
This is a secondary analysis of the Beneficial Effects of Antenatal Magnesium randomized, controlled trial conducted through the Eunice Kennedy Shriver National Institute of Child Health and Development’s Maternal-Fetal Medicine Units (MFMU) Network performed at 20 US academic centers from December 1997 through May 2004. The objective of the parent study was to assess whether administration of antenatal magnesium decreased the rate of cerebral palsy or death in children likely to be born preterm. Data were collected for women randomized between 22-31 weeks of gestation at risk of preterm birth due to of rupture of membranes, spontaneous labor, or anticipated indicated preterm delivery within 2-24 hours. The details of the study have previously been reported.
The current study was approved by the institutional review board at Columbia University Medical Center. We included nonanomalous singleton gestations delivered between 23 4/7 and 25 6/7 weeks, as the earliest survivor in our cohort was delivered at 23 4/7 weeks We excluded women with missing exposure (mode of delivery) or outcome data (Bayley II scores). Our primary exposure of interest was mode of delivery. The primary outcomes were mental and psychomotor Bayley II scores <70 at 2 years of age.
The Bayley Scales of Infant and Toddler Development are an individually administered instrument to assess neurodevelopmental functioning of infants and young children between 1-42 months of age. It is used both clinically and in the research setting. Clinically, its primary goal is to identify children with developmental delay and allow for intervention planning. As a research tool, it is used to determine the developmental outcomes associated with a history of prematurity and drug exposures. Prior to 2006 and the introduction of the Bayley III, the Bayley II was considered the standard neurodevelopmental tool in young children. The Bayley II evaluates sensory-perception, knowledge, memory, problem solving, and early language.
In the parent study, all children were evaluated by Bayley II as a prespecified secondary outcome. The test was administered by either a trained psychologist or psychometrist at 2 years of age, corrected for prematurity.
The Strengthening the Reporting of Observational Studies in Epidemiology guidelines were followed. Continuous variables were compared with the Student t test or Wilcoxon rank sum and categorical variables with χ 2 or Fisher exact test as appropriate. Bivariate analyses were used to identify independent variables that were associated with mode of delivery and neurodevelopmental outcomes defined by the Bayley II scores. The association between mode of delivery and Bayley scores was estimated using log linear regression adjusting for potential confounding variables, including gestational age at delivery, presentation at time of delivery, chorioamnionitis, years of maternal education, maternal body mass index (BMI), and original study treatment group. Our sample size was fixed by the parent study and our necessary exclusions, with 67 patients undergoing a cesarean delivery and 91 delivered vaginally. Prior data suggest that approximately 35% of early preterm infants would have an abnormal Bayley II score. Thus, with a power of 80% and a type I error rate of 0.05, we would be able to detect true relative risks of ≤0.408 or ≥1.674 in exposed subjects relative to unexposed subjects. We used a continuity-corrected χ 2 statistic or Fisher exact test to evaluate this null hypothesis. All tests were 2-tailed. We calculated adjusted odds ratios with 95% confidence intervals (CIs) for variables of interest. We then fit a log linear regression model to adjust for possible confounders. Software (SAS, version 8; SAS Institute, Cary, NC) was used for the analysis.
Results
From the original population of 2444 neonates in the parent trial, we sequentially excluded 406 twins, 1775 infants delivered at ≥26 weeks, 103 infants without Bayley II scores, and 2 infants with major congenital anomalies yielding a total of 158 meeting our inclusion and exclusion criteria. Our analysis group contained 91 neonates delivered vaginally and 67 neonates delivered by cesarean. The earliest gestational age of delivery of eligible patients was 23 4/7 weeks in the vaginal delivery group and 24 2/7 weeks in those delivered by cesarean.
Demographic and obstetric characteristics of the study population are presented in Table 1 . Exposure to magnesium sulfate, maternal education, chorioamnionitis, years of maternal education, and maternal BMI were similar in both groups. There is a significant difference between groups in presentation at time of delivery with vertex presentation by far the most common in the vaginal delivery group and breech presentation being the most common in the cesarean delivery group ( Table 1 ).
| Characteristic | Vaginal delivery | Cesarean delivery | P value |
|---|---|---|---|
| Age, y | 26.4 ± 5.2 | 27.9 ± 6.0 | .12 |
| Race | .63 | ||
| African American | 44 (48.4%) | 29 (43.3%) | |
| Caucasian | 27 (29.7%) | 21 (31.3%) | |
| Hispanic | 16 (17.6%) | 16 (23.9%) | |
| Asian | 2 (2.2%) | 0 | |
| Native American/other | 2 (2.2%) | 1 (1.5%) | |
| BMI [IOM criteria] | .18 | ||
| Underweight | 8 (8.8%) | 13 (19.4%) | |
| Normal weight | 39 (42.9%) | 23 (34.3%) | |
| Overweight | 19 (20.9%) | 10 (14.9%) | |
| Obese | 25 (27.5%) | 21 (31.3%) | |
| Presentation at time of delivery | < .0001 | ||
| Vertex | 80 (87.9%) | 22 (32.8%) | |
| Breech | 11 (12.1%) | 40 (59.7%) | |
| Transverse | 0 | 5 (7.5%) | |
| Years of education | 11.7 ± 2.4 | 12.1 ± 2.6 | .25 |
| Chorioamnionitis | 24 (26.4%) | 10 (14.9%) | .08 |
| GA at delivery, wk | 25 1/7 ± 0.54 | 25 1/7 ± 0.51 | .82 |
| PPROM | 75 (98.7%) | 57 (95%) | .32 |
| 5 min Apgar <7 | 42 (46.2%) | 27 (40.9%) | .51 |
| Sepsis a | 47 (51.7%) | 30 (44.8%) | .39 |
a Confirmed by culture or in absence of culture where there is evidence of cardiovascular collapse and x-ray findings confirm infection in setting of neonate with clinical sepsis.
Neurodevelopmental outcomes by Bayley II scores are presented in Table 2 . On unadjusted analyses, children of women who delivered vaginally or by cesarean delivery did not show a difference in Bayley Mental Developmental Index (MDI) <70 or Psychomotor Developmental Index (PDI) <70 at 2 years of age. Additionally, there was no difference in MDI or PDI <85 between vaginal and cesarean deliveries. The mean Bayley MDI and PDI scores also did not significantly differ between groups.
| Variable | Vaginal delivery, n = 91 | Cesarean delivery, n = 67 | P value | Adjusted RR (95% CI) a |
|---|---|---|---|---|
| Mental Bayley | ||||
| Mean MDI Bayley (SD) | 75.8 (20.0) | 75.5 (17.3) | .91 | |
| MDI Bayley <70, n (%) | 34 (37.4) | 26 (38.8) | .85 | 0.78 (0.47–1.31) |
| MDI Bayley <85, n (%) | 59 (64.8) | 44 (65.7) | .91 | 0.92 (0.70–1.21) |
| Motor Bayley | ||||
| Mean PDI Bayley (SD) | 80.4 (20.0) | 80.4 (19.8) | .99 | |
| PDI Bayley <70, n (%) | 28 (30.8) | 23 (34.3) | .64 | 0.98 (0.54–1.77) |
| PDI Bayley <85, n (%) | 50 (55.0) | 37 (55.2) | .97 | 1.02 (0.72–1.45) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
