Background
Maternal obesity is associated with increased systemic inflammation and an increased risk of preterm premature rupture of membranes. There is an established association between an inflammatory intrauterine environment and adverse neonatal outcomes that is independent of gestational age and mediated by the fetal inflammatory response. It is unknown whether the maternal systemic inflammation that is present in obese women influences the intrauterine environment and predisposes the fetus to adverse neonatal outcomes after preterm premature rupture of membranes.
Objective
The purpose of this study was to determine whether maternal obesity is associated with adverse neonatal outcomes in pregnancies that are complicated by preterm premature rupture of membranes.
Study Design
This was a secondary analysis of the Maternal-Fetal Medicine Units Network Randomized Clinical Trial on the Beneficial Effects of Antenatal Magnesium Sulfate. Women with singleton pregnancies that were affected by preterm premature rupture of membranes who delivered live-born infants between 24 + 0 and 33 + 6 weeks of gestation were included. An adverse neonatal outcome was defined as a composite outcome of neonatal death, severe necrotizing enterocolitis, respiratory distress syndrome, sepsis, or severe intraventricular hemorrhage. The rates of the composite outcome were compared between obese (body mass index, ≥30 kg/m 2 ) and nonobese women. Multivariable logistic regression was used to evaluate the independent effect of obesity on neonatal outcomes. Magnesium sulfate administration, steroid administration, maternal diabetes mellitus, gestational age at delivery, indomethacin exposure, birthweight, and chorioamnionitis were all considered as possible covariates in the multivariable regression models.
Results
Three hundred twenty-five of the 1288 women (25.2%) who were included were obese, and 202 of these women (62.2%) had neonates with adverse outcomes. In univariable analysis, maternal prepregnancy obesity was associated with increased odds of an adverse neonatal outcome (odds ratio, 0.30, 95% confidence interval, 1.00–1.68). However, in our multivariable logistic regression model, gestational age at delivery (odds ratio, 0.93, 95% confidence interval, 0.92–0.94 per day), but not maternal obesity (odds ratio, 1.02, 95% confidence interval, 0.75–1.38), was associated with adverse neonatal outcomes. Obese African American women experienced preterm premature rupture of membranes (189 vs 196 days; p < .001) and delivery (199 vs 205 days; p < .001) earlier than nonobese African American women. This difference was not seen in non–African American women.
Conclusions
Maternal obesity was not associated independently with adverse neonatal outcomes in pregnancies that were affected by preterm premature rupture of membranes after adjustment for gestational age at birth. However, obese African American women rupture and deliver earlier than other women, which causes increased neonatal morbidity.
Maternal obesity is associated with an increased risk of preterm premature rupture of membranes (PPROM). Maternal obesity is also associated with increased systemic inflammation. Both systemic inflammation and local infection have been proposed as explanations for the increased risk of PPROM seen in obese women.
There is an established association between an inflammatory intrauterine environment and adverse neonatal outcomes. This association is independent of gestational age and is mediated by the fetal inflammatory response syndrome. This syndrome results in severe neonatal morbidity that involves the respiratory, nervous, and gastrointestinal systems and in increased mortality rates.
Although the mechanism responsible for the increased risk of PPROM in obese women remains unknown, we hypothesized that the increase in systemic inflammation that is observed in obesity predisposes these neonates to adverse outcomes. Therefore, the purpose of this study was to evaluate whether prepregnancy obesity is associated with adverse neonatal outcomes in pregnancies that are complicated by PPROM, as compared with nonobese women.
Materials and Methods
Our study is a secondary analysis of the Maternal-Fetal Medicine Units Network (MFMU) Beneficial Effects of Antenatal Magnesium Sulfate (BEAM) randomized clinical trial of magnesium sulfate for the prevention of cerebral palsy in neonates who are at risk for preterm delivery. Women with a diagnosis of preeclampsia at the time of enrollment were excluded. Details of the methods of the original trial were published previously. The BEAM study enrolled women at participating MFMU clinical centers from 24-32 weeks of gestation with advanced preterm labor or PPROM. Trained research nurses recorded neonatal outcomes prospectively. Informed consent was obtained from all women who participated in the original MFMU BEAM trial. The Colorado Multiple Institutional Review Board approved our secondary analysis of these data.
For this analysis, we included all women with a singleton gestation who were affected by PPROM and who were delivered of a live-born infant between 24 + 0 weeks and 33 + 6 weeks gestational age. Despite enrollment in the BEAM trial from 24-32 weeks of gestation, many women went on to deliver at ≥32 weeks of gestation. Women with pregnancies that were affected by multiple gestations, stillbirths, major congenital anomalies, and chromosomal abnormalities were excluded.
Women were categorized as obese or nonobese. Prepregnancy obesity was defined as maternal body mass index (BMI) ≥30 kg/m 2 that is consistent with the Institute of Medicine guidelines. Prepregnancy height and weight was self-reported at the time of study enrollment. Demographic and clinical characteristics of the 2 groups were compared in univariable analysis with the use of t -test and chi-square, as appropriate.
Magnesium sulfate administration, steroid administration, maternal diabetes mellitus, gestational age at delivery, indomethacin exposure, birthweight, and chorioamnionitis were all considered as possible covariates in the multivariable regression models. Chorioamnionitis was defined as ≥1 temperatures at ≥37.8°C in addition to a clinical diagnosis made by the managing provider. These variables were preselected by the authors to be clinically significant covariates that have been associated previously with adverse neonatal outcomes.
The primary outcome was a composite outcome of ≥1 of the following occurrences: neonatal death before discharge, necrotizing enterocolitis stage 2-3, respiratory distress syndrome, culture-proven sepsis, or severe intraventricular hemorrhage grade 3-4. Secondary maternal and neonatal outcomes included a 5-minute Apgar score of <7, postpartum endometritis, and duration of latency from PPROM to delivery. In the original study and in this analysis, endometritis was defined as a clinical diagnosis made by the treating physician.
Demographic data were compared between obese and nonobese women. Given the known association between African American race and preterm labor, the women were stratified by racial groups and by various markers of socioeconomic status. Differences between racial groups were noted during the planned analysis. We therefore also performed an unplanned subanalysis of only African American women to further assess the observed associations between neonatal morbidity and maternal race.
Univariable analyses were used to compare the groups of women (obese and nonobese) for the composite neonatal outcome and for each of the components of the composite outcome. The odds of the composite neonatal outcome were compared between women with and without prepregnancy obesity with the use of simple and multiple logistic regression models. Latency, a nonbinary outcome, was modeled via Poisson regression. Data analysis was performed in R software (version 3.1.1), and statistical significance was set at the .05 level.
A power calculation was performed for the known sample size after our inclusion criteria were applied to the database. We determined that the number of women who were available in the dataset for analysis would allow us to detect an odds ratio of 1.78 with 80% power and a 0.05 false-positive rate.
Results
One thousand two hundred eighty-eight women met our inclusion criteria. Three hundred twenty-five of these women (25.2%) were obese (BMI, ≥30 kg/m 2 ) and 963 women (74.8%) were not obese.
The obese women in our cohort were more likely to be older and have diabetes mellitus than the women who were not obese ( Table 1 ).
| Pregnancy characteristic | Body mass index | P value | |
|---|---|---|---|
| <30 kg/m 2 (n = 963) | ≥30 kg/m 2 (n = 325) | ||
| Maternal age, n (%) | |||
| 0-17 Y | 46 (4.8) | 8 (2.5) | < .001 |
| 18-25 Y | 431 (44.8) | 112 (34.5) | |
| 26-34 Y | 355 (36.9) | 160 (49.2) | |
| 35-40 Y | 115 (12.0) | 42 (12.9) | |
| >40 Y | 14 (1.5) | 3 (0.9) | |
| Race, n (%) | |||
| African American | 428 (44.5) | 169 (52.0) | .148 |
| White | 387 (40.3) | 017 (32.9) | |
| Hispanic | 122 (12.7) | 43 (13.2) | |
| Asian | 13 (1.4) | 3 (0.9) | |
| Native American | 0 | 0 | |
| Other | 11 (1.1) | 3 (0.9) | |
| Marital status, n (%) | |||
| Married/live-in partner | 442 (46.0) | 171 (52.6) | .119 |
| Single | 519 (54.0) | 154 (47.4) | |
| Mean education, y a | 12 ± 2.4 | 12.2 ± 2.0 | .187 |
| Parity, n (%) | |||
| 0 | 326 (33.9) | 125 (38.5) | .157 |
| ≥1 | 635 (66.1) | 200 (61.5) | |
| Diabetes mellitus, n (%) | 27 (2.8) | 37 (11.4) | < .001 |
| Previous preterm birth, n (%) | 278 (28.9) | 83 (25.5) | .270 |
| Mode of delivery, n (%) | |||
| Vaginal | 609 (63.4) | 192 (59.1) | .189 |
| Cesarean | 352 (36.6) | 133 (40.9) | |
| Mean gestational age at preterm premature rupture of membranes, d a | 195 ± 18.7 | 191 ± 18.5 | < .001 |
| Mean gestational age at birth, d a | 207 ± 17.2 | 202 ± 18.1 | < .001 |
| Neonatal sex, n (%) | |||
| Male | 522 (54.3) | 165 (50.8) | .296 |
| Female | 439 (45.7) | 160 (49.2) | |
| Mean birthweight, kg a | 1.37 ± 0.45 | 1.29 ± 0.48 | .007 |
The average gestational age of neonates at PPROM and delivery was lower for obese women than women who were not obese (191 vs 195 days [ p < .001]; 202 vs 207 days [ p < .001], respectively). Accordingly, neonates who were born to obese women weighed less than neonates who were born to women who were not obese ( Table 1 ). We examined the impact of maternal BMI on birthweight using univariable and multivariable logistic regression analysis. There was no significant association between maternal obesity (≥30 kg/m 2 ) and neonatal birthweight in this cohort of premature neonates (data not shown).
The primary composite outcome of neonatal death and morbidity was present in 740 of the neonates (57.4%) overall. Two hundred two of 325 neonates (62.2%) of obese women and 538 of 963 neonates (55.9%) of women who were not obese were affected by our composite neonatal outcome. Obese women had greater odds of having neonates who were affected by the composite outcome than were women who were not obese (odds ratio [OR], 1.30; 95% confidence interval [CI], 1.00–1.68) in univariable analysis. However, in multivariable analysis, this finding was no longer significant (OR, 1.02; 95% CI, 0.75–1.38; Table 2 ). Gestational age at delivery of the affected neonates (215 vs 186 days) was the only covariate that remained associated with the composite outcome (OR, 0.93; 95% CI, 0.92–0.94) in multivariable analysis.
| Outcome | Body mass index | Univariable analysis | Multivariable regression | |||
|---|---|---|---|---|---|---|
| <30 kg/m 2 , n (%) | ≥30 kg/m 2 , n (%) | Odds ratio | 95% Confidence interval | Odds ratio | 95% Confidence interval | |
| Composite outcome | 538 (55.9%) | 202 (62.2%) | 1.30 | 1.00–1.68 | 1.02 | 0.75–1.38 |
| Individual outcome | ||||||
| Neonatal death | 47 (4.9) | 26 (8.0) | 1.69 | 1.03–2.79 | 1.29 | 0.76–2.19 |
| Necrotizing enterocolitis (stage 2-3) | 37 (3.8) | 19 (5.8) | 1.55 | 0.88–2.74 | 1.38 | 0.76–2.48 |
| Respiratory distress syndrome | 493 (51.2) | 182 (56.0) | 1.21 | 0.94–1.56 | 0.94 | 0.70–1.26 |
| Neonatal sepsis | 158 (16.4) | 62 (19.1) | 1.20 | 0.87–1.66 | 0.89 | 0.64–1.29 |
| Severe Intraventricular hemorrhage (grade III/IV) | 18 (1.9) | 8 (2.5) | 1.32 | 0.57–3.08 | 1.06 | 0.44–2.55 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
