Magnesium Sulfate Exposure and Neonatal Intensive Care Unit Admission at Term
Total Page:16
File Type:pdf, Size:1020Kb
Journal of Perinatology (2015) 35, 181–185 © 2015 Nature America, Inc. All rights reserved 0743-8346/15 www.nature.com/jp ORIGINAL ARTICLE Magnesium sulfate exposure and neonatal intensive care unit admission at term AI Girsen1, MB Greenberg2, YY El-Sayed1, H Lee3, B Carvalho4 and DJ Lyell1 OBJECTIVE: The aim of this study was to investigate the effect of maternal magnesium sulfate (MgSO4) exposure for eclampsia prophylaxis on neonatal intensive care unit (NICU) admission rates for term newborns. STUDY DESIGN: A secondary analysis of the Maternal–Fetal Medicine Unit Network Cesarean Registry, including primary and repeat cesarean deliveries, and failed and successful trials of labor after cesarean was conducted. Singleton pregnancies among women with preeclampsia and 437 weeks of gestation were included. Pregnancies with uterine rupture, chorioamnionitis and congenital malformations were excluded. Logistic regression analysis was used to determine associations between MgSO4 exposure and NICU admission. Po0.05 was considered statistically significant. RESULT: Two thousand one hundred and sixty-six term pregnancies of women with preeclampsia were included, of whom 1747 (81%) received MgSO4 for eclampsia prophylaxis and 419 (19%) did not. NICU admission rates were higher among newborns exposed to MgSO4 vs unexposed (22% vs 12%, Po0.001). After controlling for neonatal birth weight, gestational age and maternal demographic and obstetric factors, NICU admission remained significantly associated with antenatal MgSO4 exposure (adjusted odds ratio 1.9, 95% confidence interval 1.3 to 2.6, Po0.001). Newborns exposed to MgSO4 were more likely to have Apgar scores o7 at 1 and 5 min (15% vs 11% unexposed, P = 0.01 and 3% vs 0.7% unexposed, P = 0.008). There were no significant differences in NICU length of stay (median 5 (range 2 to 91) vs 6 (3 to 15), P = 0.5). CONCLUSION: Antenatal maternal MgSO4 treatment was associated with increased NICU admission rates among exposed term newborns of mothers with preeclampsia. This study highlights the need for studies of maternal MgSO4 administration protocols that optimize maternal and fetal benefits and minimize risks. Journal of Perinatology (2015) 35, 181–185; doi:10.1038/jp.2014.184; published online 16 October 2014 INTRODUCTION The primary aim of the study was to determine the effect of An estimated 5% of all pregnant women in the United States MgSO4 exposure for eclampsia prophylaxis on NICU admission receive eclampsia prophylaxis each year.1 The most common rates. Because of the reports of increased NICU admissions among eclampsia prophylaxis medication is magnesium sulfate (MgSO ), neonates exposed to MgSO4 for preterm labor and eclampsia 4 fl which has been shown to be the best choice of currently available prophylaxis, and trends toward more extensive uid, nutritional 2,3 and respiratory support needed for MgSO4-exposed term neonates, anticonvulsants. MgSO4 causes smooth muscle relaxation by antagonizing the effect of calcium and inhibiting the release of we hypothesized that a focused investigation limited to term neonates born to mothers with preeclampsia would reveal that acetylcholine in the neuromuscular junction. MgSO is the most 4 antenatal maternal MgSO exposure immediately prior to delivery commonly used tocolytic agent in North America4,5 and has been 4 increases odds of NICU admission. shown to be effective as a neuroprotective agent for preterm infants.6–8 The neonatal effect of maternal MgSO4 exposure is controver- MATERIALS AND METHODS sial. MgSO crosses the placenta, and studies 440 years ago 4 This is a secondary analysis of the Maternal–Fetal Medicine Unit Network identified the occurrence of neonatal hypermagnesemia after 9–11 (MFMU) Cesarean Registry, a previously reported prospective cohort antenatal maternal MgSO4 exposure. However, comparative observational study that was conducted by the Eunice Kennedy Shriver studies of the optimal eclampsia prophylactic did not find increased National Institute of Child Health and Development Maternal–Fetal 2,12 16 morbidity among MgSO4-exposed preterm and term neonates. Medicine Units Network from 1999 to 2002 at 19 participating centers. More recently, retrospective studies found increased neonatal The MFMU Cesarean Registry study was carried out primarily to determine the risk of uterine rupture among women undergoing a trial of labor based intensive care unit (NICU) admission rates after maternal MgSO4 fi exposure for both tocolysis for preterm labor and eclampsia on multiple vs one previous cesarean delivery. For the rst 2 years of the 13–15 study, all women delivering by a vaginal birth after a cesarean delivery or prophylaxis. There were also trends toward more frequent by cesarean delivery (primary or repeat) were enrolled. For the final 2 years neonatal respiratory compromise, need for nutritional support and of the study (2001 to 2002), only women who underwent repeat cesarean 13–15 longer stay in the NICU among MgSO4-exposed neonates. delivery or vaginal birth after cesarean who delivered infants 420 weeks 1Department of Obstetrics and Gynecology, Lucile Packard Children’s Hospital at Stanford University, Stanford, CA, USA; 2Department of Obstetrics and Gynecology, Kaiser Permanente Northern California, Oakland Medical Center, Oakland, CA, USA; 3Department of Pediatrics, Lucile Packard Children’s Hospital at Stanford University, Stanford, CA, USA and 4Department of Anesthesiology, Stanford University Medical Center, Stanford, CA, USA. Correspondence: Dr AI Girsen, Department of Obstetrics and Gynecology, Lucile Packard Children’s Hospital at Stanford University, 300 Pasteur Dr HH333 MC5317, Stanford, CA 94305, USA. E-mail: [email protected] This study was presented as a poster at the Society for Maternal–Fetal Medicine 33rd Annual Meeting, in San Francisco, CA, 11–16 February 2013. Received 13 May 2014; revised 21 August 2014; accepted 2 September 2014; published online 16 October 2014 Magnesium sulfate exposure and NICU admission AI Girsen et al 182 4 16,17 or 500 g were enrolled. Data were collected through a detailed Table 1. chart review at delivery, and information regarding neonatal morbidity Maternal and neonatal demographics was collected from discharge summaries. Data regarding patient and Magnesium Non-exposed P-valuea hospital were de-identified by the MFMU.16,17 The initial study was exposed N = 419 approved by the Human Subjects committee at each participating center, N = 1747 and the study subjects gave written informed consent before participating. fi Because of the de-identi ed data from the MFMU, this secondary analysis Maternal age (years) received a waiver of exemption by the Stanford University Institutional o18 124 (7%) 14 (3%) 0.02 Review Board. 18–35 1400 (80%) 354 (85%) All singleton pregnancies from the Cesarean Registry that were coded 435 223 (13%) 51 (12%) for maternal preeclampsia and 437 weeks of gestation at delivery were fi identi ed for the analysis. All neonates in the current study were born to Maternal race mothers with preeclampsia, with comparisons made between those whose Caucasian 389 (22%) 144 (34%) o0.001 mothers received MgSO4 for eclampsia prophylaxis before delivery vs African American 710 (41%) 135 (32%) those whose mothers did not receive MgSO4 before delivery. Preeclampsia fi Hispanic 584 (33%) 124 (30%) was de ned as a patient exhibiting hypertension with proteinuria. The Asian 13 (1%) 4 (1%) criteria for both MgSO4 administration and NICU admission were based on Native American 2 (o1%) 1 (o1%) the judgment of the medical providers. The analysis included both primary Other 49 (3%) 11 (3%) and repeat cesareans and failed and successful vaginal births after cesarean delivery. Cases with uterine rupture, chorioamnionitis and major Receipt of public insurance 896 (51%) 181 (43%) o0.001 congenital malformations were excluded from the analysis. Only cases with Preexisting maternal 487 (28%) 127 (30%) 0.35 available data on each variable were analyzed. comorbidity Neonatal respiratory distress syndrome was coded if the neonate Chronic hypertension 124 (7%) 44 (11%) 0.03 received a clinical diagnosis of respiratory distress syndrome Type I and ⩾ ⩾ oxygen therapy (FiO2 0.40) for 24 h or died before 24 h of age and Cesarean delivery 1460 (84%) 371 (89%) 0.02 received a clinical diagnosis of respiratory distress syndrome Type I with Primary 1050 (60%) 202 (48%) ⩾ oxygen therapy (FiO2 0.40). Neonatal seizure was recorded when the Repeat 410 (24%) 169 (40%) attending physician or nurse believed that a seizure definitely occurred, with or without a positive electroencephalogram to confirm the seizure VBAC 165 (9%) 25 (6%) 0.03 diagnosis. Prolonged neonatal hypotonicity was defined as a finding of Failed VBAC 122 (7%) 23 (5%) 0.26 decreased tone. Induction of labor 744 (43%) 155 (36%) 0.02 The primary outcome was NICU admission at any time before the hospital discharge. The secondary outcomes were NICU length of stay Indications for inductionb: (LOS) and neonatal death. Prolonged NICU LOS was defined using the 75th Postterm 30 (2%) 15 (4%) 0.03 percentile value for this cohort or ⩾ 8 days. PROM 14 (1%) 2 (1%) 0.71 Statistical analysis was performed using R (version 2.15.0, R Develop- IUGR 5 (o1%) 0 (0%) 0.60 ment Core Team, 2011, Vienna, Austria). Comparisons of continuous Macrosomia 9 (1%) 3 (1%) 0.90 variables were performed using t-tests or Wilcoxon Rank-Sum tests, when Oligohydramnios 16 (1%) 7 (2%) 0.28 applicable. Categorical variables were compared with the Chi-square test. Chronic hypertension 8 (o1%) 4 (1%) 0.09 After bivariable analysis was conducted, unadjusted and multivariable Non-reassurring fetal testing 9 (1%) 6 (1%) 0.39 logistic regression models were created. Covariates were included into the Preeclampsia/PIH 602 (35%) 100 (24%) o0.001 model based on their significance in bivariable analysis, defined by Po0.1. Chronic hypertension 39 (2%) 9 (2%) 0.93 The final model (including covariates: public insurance, maternal age, race/ with preeclampsia/PIH ethnicity, type of delivery, neonatal birth weight, and gestational age) Diabetes 12 (1%) 4 (1%) 0.80 was built using stepwise backwards exclusion.