- Email: [email protected]
Maternal obesity and neuroprotective magnesium sulfate
Research
ajog.org
OBSTETRICS
Maternal obesity and neuroprotective magnesium sulfate Jessica McPherson, MD; Sarah Smiley, MPH; David Stamilio, MD OBJECTIVE: Given the association between risk of cerebral palsy and
children born to obese women, the study aim was to estimate whether maternal obesity is associated with reduced effectiveness of conventional antenatal magnesium sulfate dosing for the prevention of cerebral palsy and death. STUDY DESIGN: This is a secondary cohort analysis of a multicenter
randomized clinical trial completed by the Maternal-Fetal Medicine Units Network. Women were included in the original trial if deemed high risk for preterm delivery in the subsequent 24 hours. The present study included singleton, nonanomalous fetuses that were randomized to and received magnesium sulfate with complete data available. Outcomes between obese (body mass index 30 kg/m2) and nonobese women were compared. A secondary analysis of outcomes between morbidly obese (body mass index 40 kg/m2) and nonmorbidly obese women was performed. The primary outcome was a composite of cerebral palsy or perinatal death before 15 months corrected age. Secondary outcomes included moderate to severe cerebral palsy or death, any cerebral palsy, moderate to severe cerebral palsy, and death. A logistic regression analysis was used to estimate the odds ratio of each outcome. Based on sample size,
exposure rate (obesity) and an outcome rate of 10%, assuming an 80% power and a 0.05 alpha error, this study had sufficient power to detect a 2-fold increase in the primary outcome. RESULTS: Of 1188 women randomized to receive magnesium sulfate,
806 were included in this analysis. After adjusting for gestational age at delivery, maternal obesity was not associated with an increased risk of cerebral palsy or death in children born to women who received magnesium sulfate. Women with morbid obesity had higher rates of the primary outcome and cerebral palsy in an unadjusted analysis but did not have increased risks after adjusting for gestational age at delivery. In analyses stratified on gestational age, morbidly obese women who delivered after 28 weeks had increased risks of children with cerebral palsy or death and cerebral palsy, although case numbers were small. CONCLUSION: Among women receiving antenatal neuroprotective magnesium sulfate, maternal obesity is not associated with an increased risk of having a child with cerebral palsy or death.
Key words: magnesium sulfate, neuroprotection, obesity
Cite this article as: McPherson J, Smiley S, Stamilio D. Maternal obesity and neuroprotective magnesium sulfate. Am J Obstet Gynecol 2015;213:xx-xx.
I
n the United States, overweight and obese women constitute greater than half of all pregnant women.1,2 Obesity is associated with adverse pregnancy outcomes including miscarriage, preterm delivery, the need for cesarean delivery, preeclampsia, and gestational diabetes.1,3 There is emerging evidence that the inflammatory environment that is present in obese women is responsible
for some of the adverse outcomes.2,3 The inflammatory state of obesity may be an independent risk factor for the development of cerebral palsy in children of obese mothers because the inflammatory environment may increase susceptibility to hypoxic-ischemic injury.2,4 Antenatal magnesium sulfate (MgSO4) administered to mothers at risk of preterm delivery reduces the risk of cerebral
From the Department of Obstetrics and Gynecology (Drs McPherson and Stamilio), The University of North Carolina School of Medicine (Ms Smiley), Chapel Hill, NC. Received March 2, 2015; revised April 27, 2015; accepted June 17, 2015. The views expressed herein are those of the authors and not necessarily those of the Eunice Kennedy Shriver National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network or the National Institutes of Health. The authors report no conflict of interest. Presented in poster format at the 35th annual meeting of the Society for Maternal-Fetal Medicine, San Diego, CA, Feb. 2-7, 2015. Corresponding author: Jessica A. McPherson, MD. [email protected] 0002-9378/$36.00 ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajog.2015.06.046
palsy.5,6 Magnesium sulfate crosses the placenta rapidly and levels in the neonate are similar to maternal levels at birth.7 The exact mechanism of neuroprotection is unknown but is likely multifactorial including stabilization of blood vessels and antiinflammatory effects.5,7,8 It is biologically plausible that there is a minimum amount of magnesium that must reach the fetus for the medication to be effective. Given that many pharmacological therapies, such as antibiotics, need escalated dosing for obese women because of increased vascular volume and that obesity may be an independent risk factor for cerebral palsy given the proinflammatory state, it is essential to determine whether the standard dose of neuroprotective magnesium sulfate is as effective in obese women. The aim of this study therefore was to determine whether obesity is associated with a
MONTH 2015 American Journal of Obstetrics & Gynecology
1.e1
Research
Obstetrics
reduced effectiveness of conventional antenatal magnesium sulfate dosing for neuroprotection, resulting in an increased risk of cerebral palsy and death in children born to mothers who received magnesium sulfate.
M ATERIALS
AND
M ETHODS
This study is a secondary cohort analysis of a previously reported multicenter randomized clinical trial. The initial study was completed by the MaternalFetal Medicine Units Network and subjects were recruited at 20 centers between 1997 and 2004.5 All centers and the data-coordinating center received institutional review board approval. This study, based on publicly available deidentified data, was exempt by the Institutional Review Board at the University of North Carolina. The data collection procedures for the original study have been previously described.5 Infant follow-up data were obtained at 6, 12, and 24 months of age. All data were edited and validated on a regular basis. The inclusion criteria and randomization protocol for the initial prospective trial are previously detailed.5 Briefly, women were included in the initial trial if considered to be at high risk for preterm delivery at 24 through 31 weeks. Women were considered high risk of preterm delivery if they presented with rupture of membranes (22-31 weeks), spontaneous labor with cervical dilation of 4-8 cm, or providers anticipated an indicated preterm delivery within 24 hours. The primary analysis of the current study included only singleton, nonanomalous fetuses (diagnosed before or after birth) randomized to MgSO4 infusion that received study drug. However, we performed a supplementary analysis stratified on obesity of all patients in the trial (allocated to study drug vs placebo) to help assess whether maternal obesity modified the effect of magnesium sulfate on the risk of cerebral palsy or death. Magnesium sulfate was administered via a loading dose (6 g over 20-30 minutes), followed by a maintenance infusion of 2 g/h. After 12 hours, if delivery had not occurred and was no longer
ajog.org determined imminent, the MgSO4 was discontinued. If delivery again threatened, the MgSO4 was restarted; if more than 6 hours had passed since the drug was given, the MgSO4 loading dose was repeated upon reinitiating therapy. Retreatment did not occur if open-label magnesium sulfate became indicated for preeclampsia, if a delay in delivery for retreatment would increase risk to mother or fetus, or the gestational age reached 34 weeks. Study groups in the current study were defined by prepregnancy body mass index (BMI) determined at the time of enrollment. For the primary analyses, BMI of 30 kg/m2 was considered obese and was defined as the exposed group. BMI <30 kg/m2 was considered nonobese. The primary outcome was a composite of cerebral palsy (mild, moderate, or severe) or death (stillbirth or death before 15 months corrected age). Secondary outcomes included moderate to severe cerebral palsy or death, any cerebral palsy, moderate to severe cerebral palsy, and death. Cerebral palsy was assessed and diagnosed at or beyond 2 years of life by an annually certified pediatrician or pediatric neurologist according to strict criteria.5 Additional secondary analyses compared outcomes in morbidly obese women (defined as BMI of 40 kg/m2) with women with a BMI less than 40 kg/m2. To reduce the chance of masking an association by including overweight women in the nonobese reference group, the analysis for obese women (BMI 30 kg/m2) was repeated excluding women with BMI of 25-30 kg/m2 from the nonobese group. Similarly, the analysis for morbidly obese patients was repeated excluding overweight and obese women from the nonmorbidly obese group. To assess for interaction between gestational age and obesity or morbid obesity, we also performed analyses stratified on prematurity severity using delivery gestational age categories of less than 28 weeks, 28-32 weeks, and more than 32 weeks. Lastly, we performed a supplementary stratified analysis of the entire clinical
1.e2 American Journal of Obstetrics & Gynecology MONTH 2015
trial population comparing the MgSO4 group vs the placebo group for infant outcomes stratified on maternal obesity severity to determine whether obesity was a modifier of the study drug effect on infant neurological and mortality outcomes. The assessment of obesity as an effect modifier was performed in 2 ways: first, as an intent-to-treat analysis as randomly allocated and then as an as-treated analysis comparing all patients who received study MgSO4 or open-label MgSO4 to patients who received no MgSO4. We tested for evidence of interaction using the MantelHaenszel c2 test for heterogeneity and logistic regression analysis. Baseline characteristics between obese and nonobese women were described. Covariates associated with the outcome of cerebral palsy or death were determined using the Student t test or Mann-Whitney U test for continuous variables and the c2 or the Fisher exact test for categorical variables as appropriate. Differences in incidences of the primary and secondary outcomes between study groups were estimated and tested using a c2 or a Fisher exact test and the unadjusted relative risks (RRs) with 95% confidence intervals (CIs) were estimated. Effect modifiers and potential confounding factors were identified via literature review and in bivariate analyses. Logistic regression models were developed to estimate the independent risk of obesity for each outcome, adjusting for confounding factors. All variables considered as potential confounders were included in the initial logistic regression model and were removed one by one while assessing the magnitude of change in the effect size with the remaining covariates. Variables that changed the odds ratio by greater than 10% or were known historically to be confounders were included in the final regression models and adjusted odds ratios with 95% CIs were estimated (P < .05 was considered significant). Goodness of fit was assessed with the Hosmer-Lemeshow test for each regression model. Statistical analyses were performed using STATA version
Obstetrics
ajog.org 13.0, special edition (StataCorp, College Station, TX). Because the sample size was fixed, we did not perform a sample size estimate. However, prior to performing the study, we calculated the detectable difference in outcome rates between exposure groups. Based on the total sample size, the exposure rate (obesity), and the outcome rate of 10%, assuming a power of 80% and an alpha of 0.05, this study had sufficient statistical power to detect a 2-fold increased risk for the primary outcome (cerebral palsy or death).
Research
FIGURE
Flow diagram of study participants
R ESULTS In the initial randomized trial, 1188 fetuses (1096 women) were randomized to receive antenatal magnesium sulfate. There were 841 women available for this analysis after exclusions. There were 204 women considered obese with a BMI of 30 kg/m2, 31 of whom had a BMI of 40 kg/m2 and were considered morbidly obese and 637 women with a BMI <30 kg/m2 (Figure). In summary statistics, obese women were more likely to be African American or Hispanic and less likely to be white. Obese women were less likely to smoke, use illicit drugs, or have a history of a preterm delivery. Obese women were more likely to have diabetes and be nulliparous. Importantly, the gestational age at randomization, gestational age at delivery, rates of chorioamnionitis, rates of steroid administration, hours of magnesium sulfate received, frequency of magnesium sulfate running at delivery, and eligibility criteria between obese and nonobese women were similar. Obese women, on average, had more time elapse since receiving magnesium sulfate at the time of delivery (Table 1). When obese women were compared with nonobese women, there was no difference in the risk of the composite primary outcome of any cerebral palsy or death when adjusted for gestational age at delivery (adjusted odds ratio, 1.3; 95% CI, 0.8e2.1) (Table 2). There was no significant difference in secondary outcomes including the composite of moderate to severe cerebral palsy or
BMI, body mass index. McPherson. Neuroprotective magnesium sulfate and obesity. Am J Obstet Gynecol 2015.
death, any cerebral palsy, moderate to severe cerebral palsy, or death after adjusting each model for gestational age at delivery; however, there was a nonsignificant trend toward worse outcomes in obese women. In analyses comparing morbidly obese women with nonmorbidly obese women, there was no difference in any of the primary or secondary outcomes (Table 3). However, we observed a possible interaction between obesity status and delivery gestational age; when stratified by delivery gestational age category (<28 weeks, 28-32 weeks, and >32 weeks), morbidly obese women who delivered at 28-32 weeks appeared to have a 5-fold increased risk of cerebral palsy or death (RR, 5.0; 95% CI, 1.8e14.4) and a 10-fold increased risk of cerebral palsy (RR, 10.1; 95% CI, 2.4e43.1) when compared with nonmorbidly obese women, although case
numbers were small because of stratification. When we performed a sensitivity analysis comparing the obese group and the morbidly obese group each to a reference group containing only normalweight women (BMI <25 kg/m2), the results were similar to the initial analysis, with no difference in the primary or secondary outcomes (data not shown). Lastly, we performed an analysis of the entire clinical trial population stratified on obesity category (nonobese, obese, or morbidly obese) to estimate whether obesity modified the effect of magnesium sulfate on the risk of infant cerebral palsy or death. Obesity status did not significantly modify the effect of magnesium sulfate on any of the outcomes including any cerebral palsy, moderate to severe cerebral palsy, death and any cerebral palsy or death (data not shown). These negative results for interaction
MONTH 2015 American Journal of Obstetrics & Gynecology
1.e3
Research
Obstetrics
ajog.org
TABLE 1
Baseline characteristics of obese vs nonobese women receiving magnesium sulfate for neuroprotection (n [ 841) Obese (n [ 204)
Characteristic
Nonobese (n [ 637)
Maternal age, y
27.1 5.9
25.8 6.4
Advanced maternal age (35 y old)
27 (13.2%)
75 (11.8%)
96 (47.1%)
278 (43.6%)
White
66 (32.4%)
258 (40.5%)
Hispanic
38 (18.6%)
86 (13.5%)
4 (1.9%)
15 (2.4%)
Educational level, y
12.0 2.5
11.9 2.3
Alcohol use
14 (6.9%)
64 (10.1%)
Tobacco use
38 (18.6%)
194 (30.5%)
Illicit drug use
8 (3.9%)
75 (11.8%)
Race African American
Other
Diabetes
22 (10.8%)
16 (2.5%)
Nulliparous
82 (40.2%)
212 (33.3%)
Previous preterm delivery
46 (22.6%)
193 (30.3%)
7 (3.4%)
46 (7.2%)
Gestational age at randomization, wks
27.7 2.6
28.5 2.4
Gestational age at delivery, wks
29.3 3.3
30.2 3.0
Chorioamnionitis
28 (13.7%)
74 (11.6%)
199 (97.6%)
621 (97.5%)
14.4 7.8
15.6 8.9
No prenatal care
Steroid administration Total duration of MgSO4, h Study MgSO4 on at delivery (yes)
110 (53.9%)
357 (56.1%)
Any MgSO4 (open label or study), running at delivery
113 (55.4%)
358 (56.2%)
Hours since last MgSO4 infusion (all patients)
173.9 411.6
152.8 360.3
MgSO4 received prior to randomization Open-label MgSO4 received after randomization Open-label MgSO4 received anytime
15 (7.4%)
71 (11.2%)
3 (1.5%)
16 (2.5%)
33 (16.2%)
135 (21.2%)
181 (88.7%)
559 (87.8%)
18 (8.8%)
61 (9.6%)
5 (2.5%)
17 (2.7%)
Eligibility criteria Premature rupture of membranes Advanced preterm labor Indicated preterm delivery
Mean SD or absolute number (percentage) unless otherwise specified. Obese indicated a BMI of 30 kg/m2; nonobese indicated a BMI <30 kg/m2. BMI, body mass index; MgSO4, magnesium sulfate. McPherson. Neuroprotective magnesium sulfate and obesity. Am J Obstet Gynecol 2015.
were stable across intention-to-treat and as-treated analyses.
C OMMENT To our knowledge, based on a thorough literature search, this study is the first to
investigate the effect of maternal obesity on the effectiveness of antenatal magnesium sulfate for neuroprotection for pregnancies at high risk for preterm delivery. We observed that among atrisk women who received magnesium
1.e4 American Journal of Obstetrics & Gynecology MONTH 2015
sulfate, obese women did not have higher rates of infant cerebral palsy or death compared with nonobese women. Furthermore, obesity did not appear to modify the effect of magnesium sulfate on infant cerebral palsy or death in an analysis of the entire clinical trial population that was randomly allocated to either study drug or placebo. Because maternal obesity does not appear to alter the risk reduction of magnesium sulfate neuroprotective therapy, this suggests that escalated dosing is not required for obese patients to realize benefit. In 2008, Rouse et al5 evaluated the effect of magnesium sulfate for cerebral palsy prevention and found that moderate or severe cerebral palsy was lower in women in the magnesium sulfate group compared with placebo (1.9% vs 3.5%; RR, 0.6; 95% CI, 0.3e1.0). Although antenatal magnesium sulfate for neuroprotection in high-risk patients has become the standard of care, optimal dosing has not been defined in general or selected populations. In 2014, McPherson et al9 found that the duration of magnesium sulfate given for neuroprotection was not associated with the rates of cerebral palsy, but specific dosing recommendations could not be made, especially for a specialized population such as obese patients. There is evidence that obesity leads to increased rates of cerebral palsy secondary to increased susceptibility of fetal brain injury related to the inflammatory environment. In 2013, Crisham Janik et al2 performed a retrospective cohort study linking cases of childhood cerebral palsy to maternal obesity and found that obese women had 27% increase in the odds of having a child with cerebral palsy (95% CI, 1.1e1.5), and morbidly obese women had a >2.5-fold increase in the odds of having a child with cerebral palsy (95% CI, 1.8e3.7). In 2014, Pan et al4 found that morbidly obese women were at 3 times increased odds of having a child with cerebral palsy (95% CI, 1.1e8.4). Our study has several strengths. The prospective cohort is derived from a wellcharacterized, comprehensive multicenter clinical trial study population
Obstetrics
ajog.org
Research
TABLE 2
Outcomes of obese vs nonobese women receiving magnesium sulfate (n [ 806) Outcome
Obese (n [ 199)
Nonobese (n [ 607)
RR (95% CI)
aORa (95% CI)
Any cerebral palsy or death
28 (14.1%)
54 (8.9%)
1.60 (1.03e2.43)
1.27 (0.76e2.14)
Moderate to severe cerebral palsy or death
24 (12.1%)
44 (7.2%)
1.66 (1.04e2.66)
1.31 (0.75e2.28)
Any cerebral palsy
7 (3.5%)
17 (2.8%)
1.26 (0.52e2.98)
1.04 (0.42e2.59)
Moderate to severe cerebral palsy
3 (1.5%)
6 (0.9%)
1.53 (0.38e6.04)
1.27 (0.31e5.25)
21 (10.6%)
40 (6.6%)
1.60 (0.97e2.66)
1.22 (0.68e2.21)
b
Death
Obese indicates a BMI of 30 kg/m ; nonobese indicates a BMI <30 kg/m . 2
2
aOR, adjusted odds ratio; CI, confidence interval; RR, relative risk. a
Adjusted for gestational age at delivery in weeks; b Any cerebral palsy includes cerebral palsy that precedes death.
McPherson. Neuroprotective magnesium sulfate and obesity. Am J Obstet Gynecol 2015.
that has been previously validated. Information bias is minimized, given the prospective nature of the data collection. The retrospective design of this analysis, however, is a potential limitation because we are restricted to variables that exist in the established database and there are missing data, however, rarely greater than 5% within a single variable. One notable variable that was not available is gestational weight gain. Therefore, we were unable to assess the independent or additive effect of increased gestational weight gain on magnesium sulfate neuroprotective efficacy or neonatal risk. Notably, the study groups defined by maternal BMI differed in several baseline characteristics. Multivariable regression analysis, however, was used to adjust for these measured differences, minimizing
confounding and leaving only unmeasured potential residual bias. Although our main analysis indicates that maternal obesity does not affect the neuroprotective effect of antenatal magnesium sulfate, a subanalysis of morbidly obese women suggests that the risk for infant cerebral palsy may be increased in those who deliver after 28 gestational weeks. We hypothesize that the excess cerebral palsy risk associated with morbid obesity was observed only after 28 gestational weeks because prior to 28 weeks, the majority of risk is likely attributed to extreme prematurity rather than maternal characteristics. Owing to the small case numbers from stratification and the potential for type 1 error with multiple comparisons, we consider this observation as hypothesis generating and recommend that this
specific population be studied more rigorously before concluding that magnesium sulfate dosing must be increased in the morbidly obese patient at risk for preterm delivery. Based on the known effects of morbid obesity on drug levels in general and on the inflammatory milieu, it is biologically plausible that neuroprotective efficacy could be reduced. And given the increasing prevalence of morbid obesity compounded with our observations, this is an important subgroup to study further. In conclusion, we did not observe an excess of infant cerebral palsy among obese women who received antenatal magnesium sulfate. Therefore, we cannot recommend an escalated magnesium sulfate dosing for obese patients who meet criteria for antenatal neuroprotective therapy. -
TABLE 3
Outcomes of morbidly obese vs nonmorbidly obese women receiving magnesium sulfate (n [ 806) Outcome
Morbidly obese (n [ 31)
Nonmorbidly obese (n [ 775)
RR (95% CI)
aORa (95% CI)
Any cerebral palsy or death
7 (22.6%)
75 (9.7%)
2.33 (1.17e4.64)
2.24 (0.86e5.82)
Moderate to severe cerebral palsy or death
5 (16.1%)
63 (8.1%)
1.98 (0.85e4.58)
1.68 (0.68e4.87)
Any cerebral palsyb
3 (9.6%)
21 (2.7%)
3.5 (1.12e11.34)
3.28 (0.90e11.96)
Moderate to severe cerebral palsy
1 (3.2%)
8 (1.0%)
3.13 (0.40e24.22)
2.67 (0.32e22.61)
Death
4 (12.9%)
57 (7.34%)
1.76 (0.68e4.54)
1.39 (0.44e4.42)
Morbidly obese indicates a BMI of 40 kg/m2; nonmorbidly obese indicates a BMI <40 kg/m2. aOR, adjusted odds ratio; CI, confidence interval; RR, relative risk. a
Adjusted for gestational age at delivery; b Any cerebral palsy includes cerebral palsy that precedes death.
McPherson. Neuroprotective magnesium sulfate and obesity. Am J Obstet Gynecol 2015.
MONTH 2015 American Journal of Obstetrics & Gynecology
1.e5
Research
Obstetrics
ACKNOWLEDGMENT We acknowledge the Eunice Kennedy Shriver National Institute of Child Health and Human Development, the Maternal-Fetal Medicine Units Network, and the Protocol Subcommittee for making this database publicly available.
REFERENCES 1. American College of Obstetricians and Gynecologists. Obesity in pregnancy. ACOG Committee opinion no. 549. Obstet Gynecol 2013;121:213-7. 2. Crisham Janik MD, Newman TB, Cheng YW, Xing G, Gilbert WM, Wu YW. Maternal diagnosis
ajog.org of obesity and risk of cerebral palsy in the child. J Pediatr 2013;163:1307-12. 3. Bergehella V. Obstetric evidence based guidelines, 2nd ed. New York, NY: Informa Healthcare; 2012. 4. Pan C, Deroche CB, Mann JR, McDermott S, Hardin JW. Is pregnancy obesity associated with risk of cerebral palsy and epilepsy in children? J Child Neorol 2014;29: NP196-201. 5. Rouse DJ, Hirtz DG, Thom E, et al. A randomized, controlled trial of magnesium sulfate for the prevention of cerebral palsy. N Engl J Med 2008;359:895-905. 6. American College of Obstetricians and Gynecologists. Magnesium sulfate before anticipated preterm birth for neuroprotection. ACOG
1.e6 American Journal of Obstetrics & Gynecology MONTH 2015
Committee opinion no. 455. Obstet Gynecol 2010;116:669-71. 7. Sugimoto J, Romani A, Valenti-Torres A, et al. Magnesium decreases inflammatory cytokine production: a novel innate immunomodulatory mechanism. J Immunol 2012;188:6338-46. 8. Suzuki-Kaksaka H, Sugimoto J, Tetarbe M, Romani A, Ramirez Kitchen C, Berstein H. Magnesium sulfate increases intracellular magnesium reducing inflammatory cytokine release in neonates. Am J Reprod Immunol 2013;70: 213-20. 9. McPherson J, Rouse D, Grobman W, Palatnik A, Stamilio D. Association of duration of neuroprotective magnesium sulfate with neonatal and maternal outcomes. Obstet Gynecol 2014;124:749-55.
ajog.org
OBSTETRICS
Maternal obesity and neuroprotective magnesium sulfate Jessica McPherson, MD; Sarah Smiley, MPH; David Stamilio, MD OBJECTIVE: Given the association between risk of cerebral palsy and
children born to obese women, the study aim was to estimate whether maternal obesity is associated with reduced effectiveness of conventional antenatal magnesium sulfate dosing for the prevention of cerebral palsy and death. STUDY DESIGN: This is a secondary cohort analysis of a multicenter
randomized clinical trial completed by the Maternal-Fetal Medicine Units Network. Women were included in the original trial if deemed high risk for preterm delivery in the subsequent 24 hours. The present study included singleton, nonanomalous fetuses that were randomized to and received magnesium sulfate with complete data available. Outcomes between obese (body mass index 30 kg/m2) and nonobese women were compared. A secondary analysis of outcomes between morbidly obese (body mass index 40 kg/m2) and nonmorbidly obese women was performed. The primary outcome was a composite of cerebral palsy or perinatal death before 15 months corrected age. Secondary outcomes included moderate to severe cerebral palsy or death, any cerebral palsy, moderate to severe cerebral palsy, and death. A logistic regression analysis was used to estimate the odds ratio of each outcome. Based on sample size,
exposure rate (obesity) and an outcome rate of 10%, assuming an 80% power and a 0.05 alpha error, this study had sufficient power to detect a 2-fold increase in the primary outcome. RESULTS: Of 1188 women randomized to receive magnesium sulfate,
806 were included in this analysis. After adjusting for gestational age at delivery, maternal obesity was not associated with an increased risk of cerebral palsy or death in children born to women who received magnesium sulfate. Women with morbid obesity had higher rates of the primary outcome and cerebral palsy in an unadjusted analysis but did not have increased risks after adjusting for gestational age at delivery. In analyses stratified on gestational age, morbidly obese women who delivered after 28 weeks had increased risks of children with cerebral palsy or death and cerebral palsy, although case numbers were small. CONCLUSION: Among women receiving antenatal neuroprotective magnesium sulfate, maternal obesity is not associated with an increased risk of having a child with cerebral palsy or death.
Key words: magnesium sulfate, neuroprotection, obesity
Cite this article as: McPherson J, Smiley S, Stamilio D. Maternal obesity and neuroprotective magnesium sulfate. Am J Obstet Gynecol 2015;213:xx-xx.
I
n the United States, overweight and obese women constitute greater than half of all pregnant women.1,2 Obesity is associated with adverse pregnancy outcomes including miscarriage, preterm delivery, the need for cesarean delivery, preeclampsia, and gestational diabetes.1,3 There is emerging evidence that the inflammatory environment that is present in obese women is responsible
for some of the adverse outcomes.2,3 The inflammatory state of obesity may be an independent risk factor for the development of cerebral palsy in children of obese mothers because the inflammatory environment may increase susceptibility to hypoxic-ischemic injury.2,4 Antenatal magnesium sulfate (MgSO4) administered to mothers at risk of preterm delivery reduces the risk of cerebral
From the Department of Obstetrics and Gynecology (Drs McPherson and Stamilio), The University of North Carolina School of Medicine (Ms Smiley), Chapel Hill, NC. Received March 2, 2015; revised April 27, 2015; accepted June 17, 2015. The views expressed herein are those of the authors and not necessarily those of the Eunice Kennedy Shriver National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network or the National Institutes of Health. The authors report no conflict of interest. Presented in poster format at the 35th annual meeting of the Society for Maternal-Fetal Medicine, San Diego, CA, Feb. 2-7, 2015. Corresponding author: Jessica A. McPherson, MD. [email protected] 0002-9378/$36.00 ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajog.2015.06.046
palsy.5,6 Magnesium sulfate crosses the placenta rapidly and levels in the neonate are similar to maternal levels at birth.7 The exact mechanism of neuroprotection is unknown but is likely multifactorial including stabilization of blood vessels and antiinflammatory effects.5,7,8 It is biologically plausible that there is a minimum amount of magnesium that must reach the fetus for the medication to be effective. Given that many pharmacological therapies, such as antibiotics, need escalated dosing for obese women because of increased vascular volume and that obesity may be an independent risk factor for cerebral palsy given the proinflammatory state, it is essential to determine whether the standard dose of neuroprotective magnesium sulfate is as effective in obese women. The aim of this study therefore was to determine whether obesity is associated with a
MONTH 2015 American Journal of Obstetrics & Gynecology
1.e1
Research
Obstetrics
reduced effectiveness of conventional antenatal magnesium sulfate dosing for neuroprotection, resulting in an increased risk of cerebral palsy and death in children born to mothers who received magnesium sulfate.
M ATERIALS
AND
M ETHODS
This study is a secondary cohort analysis of a previously reported multicenter randomized clinical trial. The initial study was completed by the MaternalFetal Medicine Units Network and subjects were recruited at 20 centers between 1997 and 2004.5 All centers and the data-coordinating center received institutional review board approval. This study, based on publicly available deidentified data, was exempt by the Institutional Review Board at the University of North Carolina. The data collection procedures for the original study have been previously described.5 Infant follow-up data were obtained at 6, 12, and 24 months of age. All data were edited and validated on a regular basis. The inclusion criteria and randomization protocol for the initial prospective trial are previously detailed.5 Briefly, women were included in the initial trial if considered to be at high risk for preterm delivery at 24 through 31 weeks. Women were considered high risk of preterm delivery if they presented with rupture of membranes (22-31 weeks), spontaneous labor with cervical dilation of 4-8 cm, or providers anticipated an indicated preterm delivery within 24 hours. The primary analysis of the current study included only singleton, nonanomalous fetuses (diagnosed before or after birth) randomized to MgSO4 infusion that received study drug. However, we performed a supplementary analysis stratified on obesity of all patients in the trial (allocated to study drug vs placebo) to help assess whether maternal obesity modified the effect of magnesium sulfate on the risk of cerebral palsy or death. Magnesium sulfate was administered via a loading dose (6 g over 20-30 minutes), followed by a maintenance infusion of 2 g/h. After 12 hours, if delivery had not occurred and was no longer
ajog.org determined imminent, the MgSO4 was discontinued. If delivery again threatened, the MgSO4 was restarted; if more than 6 hours had passed since the drug was given, the MgSO4 loading dose was repeated upon reinitiating therapy. Retreatment did not occur if open-label magnesium sulfate became indicated for preeclampsia, if a delay in delivery for retreatment would increase risk to mother or fetus, or the gestational age reached 34 weeks. Study groups in the current study were defined by prepregnancy body mass index (BMI) determined at the time of enrollment. For the primary analyses, BMI of 30 kg/m2 was considered obese and was defined as the exposed group. BMI <30 kg/m2 was considered nonobese. The primary outcome was a composite of cerebral palsy (mild, moderate, or severe) or death (stillbirth or death before 15 months corrected age). Secondary outcomes included moderate to severe cerebral palsy or death, any cerebral palsy, moderate to severe cerebral palsy, and death. Cerebral palsy was assessed and diagnosed at or beyond 2 years of life by an annually certified pediatrician or pediatric neurologist according to strict criteria.5 Additional secondary analyses compared outcomes in morbidly obese women (defined as BMI of 40 kg/m2) with women with a BMI less than 40 kg/m2. To reduce the chance of masking an association by including overweight women in the nonobese reference group, the analysis for obese women (BMI 30 kg/m2) was repeated excluding women with BMI of 25-30 kg/m2 from the nonobese group. Similarly, the analysis for morbidly obese patients was repeated excluding overweight and obese women from the nonmorbidly obese group. To assess for interaction between gestational age and obesity or morbid obesity, we also performed analyses stratified on prematurity severity using delivery gestational age categories of less than 28 weeks, 28-32 weeks, and more than 32 weeks. Lastly, we performed a supplementary stratified analysis of the entire clinical
1.e2 American Journal of Obstetrics & Gynecology MONTH 2015
trial population comparing the MgSO4 group vs the placebo group for infant outcomes stratified on maternal obesity severity to determine whether obesity was a modifier of the study drug effect on infant neurological and mortality outcomes. The assessment of obesity as an effect modifier was performed in 2 ways: first, as an intent-to-treat analysis as randomly allocated and then as an as-treated analysis comparing all patients who received study MgSO4 or open-label MgSO4 to patients who received no MgSO4. We tested for evidence of interaction using the MantelHaenszel c2 test for heterogeneity and logistic regression analysis. Baseline characteristics between obese and nonobese women were described. Covariates associated with the outcome of cerebral palsy or death were determined using the Student t test or Mann-Whitney U test for continuous variables and the c2 or the Fisher exact test for categorical variables as appropriate. Differences in incidences of the primary and secondary outcomes between study groups were estimated and tested using a c2 or a Fisher exact test and the unadjusted relative risks (RRs) with 95% confidence intervals (CIs) were estimated. Effect modifiers and potential confounding factors were identified via literature review and in bivariate analyses. Logistic regression models were developed to estimate the independent risk of obesity for each outcome, adjusting for confounding factors. All variables considered as potential confounders were included in the initial logistic regression model and were removed one by one while assessing the magnitude of change in the effect size with the remaining covariates. Variables that changed the odds ratio by greater than 10% or were known historically to be confounders were included in the final regression models and adjusted odds ratios with 95% CIs were estimated (P < .05 was considered significant). Goodness of fit was assessed with the Hosmer-Lemeshow test for each regression model. Statistical analyses were performed using STATA version
Obstetrics
ajog.org 13.0, special edition (StataCorp, College Station, TX). Because the sample size was fixed, we did not perform a sample size estimate. However, prior to performing the study, we calculated the detectable difference in outcome rates between exposure groups. Based on the total sample size, the exposure rate (obesity), and the outcome rate of 10%, assuming a power of 80% and an alpha of 0.05, this study had sufficient statistical power to detect a 2-fold increased risk for the primary outcome (cerebral palsy or death).
Research
FIGURE
Flow diagram of study participants
R ESULTS In the initial randomized trial, 1188 fetuses (1096 women) were randomized to receive antenatal magnesium sulfate. There were 841 women available for this analysis after exclusions. There were 204 women considered obese with a BMI of 30 kg/m2, 31 of whom had a BMI of 40 kg/m2 and were considered morbidly obese and 637 women with a BMI <30 kg/m2 (Figure). In summary statistics, obese women were more likely to be African American or Hispanic and less likely to be white. Obese women were less likely to smoke, use illicit drugs, or have a history of a preterm delivery. Obese women were more likely to have diabetes and be nulliparous. Importantly, the gestational age at randomization, gestational age at delivery, rates of chorioamnionitis, rates of steroid administration, hours of magnesium sulfate received, frequency of magnesium sulfate running at delivery, and eligibility criteria between obese and nonobese women were similar. Obese women, on average, had more time elapse since receiving magnesium sulfate at the time of delivery (Table 1). When obese women were compared with nonobese women, there was no difference in the risk of the composite primary outcome of any cerebral palsy or death when adjusted for gestational age at delivery (adjusted odds ratio, 1.3; 95% CI, 0.8e2.1) (Table 2). There was no significant difference in secondary outcomes including the composite of moderate to severe cerebral palsy or
BMI, body mass index. McPherson. Neuroprotective magnesium sulfate and obesity. Am J Obstet Gynecol 2015.
death, any cerebral palsy, moderate to severe cerebral palsy, or death after adjusting each model for gestational age at delivery; however, there was a nonsignificant trend toward worse outcomes in obese women. In analyses comparing morbidly obese women with nonmorbidly obese women, there was no difference in any of the primary or secondary outcomes (Table 3). However, we observed a possible interaction between obesity status and delivery gestational age; when stratified by delivery gestational age category (<28 weeks, 28-32 weeks, and >32 weeks), morbidly obese women who delivered at 28-32 weeks appeared to have a 5-fold increased risk of cerebral palsy or death (RR, 5.0; 95% CI, 1.8e14.4) and a 10-fold increased risk of cerebral palsy (RR, 10.1; 95% CI, 2.4e43.1) when compared with nonmorbidly obese women, although case
numbers were small because of stratification. When we performed a sensitivity analysis comparing the obese group and the morbidly obese group each to a reference group containing only normalweight women (BMI <25 kg/m2), the results were similar to the initial analysis, with no difference in the primary or secondary outcomes (data not shown). Lastly, we performed an analysis of the entire clinical trial population stratified on obesity category (nonobese, obese, or morbidly obese) to estimate whether obesity modified the effect of magnesium sulfate on the risk of infant cerebral palsy or death. Obesity status did not significantly modify the effect of magnesium sulfate on any of the outcomes including any cerebral palsy, moderate to severe cerebral palsy, death and any cerebral palsy or death (data not shown). These negative results for interaction
MONTH 2015 American Journal of Obstetrics & Gynecology
1.e3
Research
Obstetrics
ajog.org
TABLE 1
Baseline characteristics of obese vs nonobese women receiving magnesium sulfate for neuroprotection (n [ 841) Obese (n [ 204)
Characteristic
Nonobese (n [ 637)
Maternal age, y
27.1 5.9
25.8 6.4
Advanced maternal age (35 y old)
27 (13.2%)
75 (11.8%)
96 (47.1%)
278 (43.6%)
White
66 (32.4%)
258 (40.5%)
Hispanic
38 (18.6%)
86 (13.5%)
4 (1.9%)
15 (2.4%)
Educational level, y
12.0 2.5
11.9 2.3
Alcohol use
14 (6.9%)
64 (10.1%)
Tobacco use
38 (18.6%)
194 (30.5%)
Illicit drug use
8 (3.9%)
75 (11.8%)
Race African American
Other
Diabetes
22 (10.8%)
16 (2.5%)
Nulliparous
82 (40.2%)
212 (33.3%)
Previous preterm delivery
46 (22.6%)
193 (30.3%)
7 (3.4%)
46 (7.2%)
Gestational age at randomization, wks
27.7 2.6
28.5 2.4
Gestational age at delivery, wks
29.3 3.3
30.2 3.0
Chorioamnionitis
28 (13.7%)
74 (11.6%)
199 (97.6%)
621 (97.5%)
14.4 7.8
15.6 8.9
No prenatal care
Steroid administration Total duration of MgSO4, h Study MgSO4 on at delivery (yes)
110 (53.9%)
357 (56.1%)
Any MgSO4 (open label or study), running at delivery
113 (55.4%)
358 (56.2%)
Hours since last MgSO4 infusion (all patients)
173.9 411.6
152.8 360.3
MgSO4 received prior to randomization Open-label MgSO4 received after randomization Open-label MgSO4 received anytime
15 (7.4%)
71 (11.2%)
3 (1.5%)
16 (2.5%)
33 (16.2%)
135 (21.2%)
181 (88.7%)
559 (87.8%)
18 (8.8%)
61 (9.6%)
5 (2.5%)
17 (2.7%)
Eligibility criteria Premature rupture of membranes Advanced preterm labor Indicated preterm delivery
Mean SD or absolute number (percentage) unless otherwise specified. Obese indicated a BMI of 30 kg/m2; nonobese indicated a BMI <30 kg/m2. BMI, body mass index; MgSO4, magnesium sulfate. McPherson. Neuroprotective magnesium sulfate and obesity. Am J Obstet Gynecol 2015.
were stable across intention-to-treat and as-treated analyses.
C OMMENT To our knowledge, based on a thorough literature search, this study is the first to
investigate the effect of maternal obesity on the effectiveness of antenatal magnesium sulfate for neuroprotection for pregnancies at high risk for preterm delivery. We observed that among atrisk women who received magnesium
1.e4 American Journal of Obstetrics & Gynecology MONTH 2015
sulfate, obese women did not have higher rates of infant cerebral palsy or death compared with nonobese women. Furthermore, obesity did not appear to modify the effect of magnesium sulfate on infant cerebral palsy or death in an analysis of the entire clinical trial population that was randomly allocated to either study drug or placebo. Because maternal obesity does not appear to alter the risk reduction of magnesium sulfate neuroprotective therapy, this suggests that escalated dosing is not required for obese patients to realize benefit. In 2008, Rouse et al5 evaluated the effect of magnesium sulfate for cerebral palsy prevention and found that moderate or severe cerebral palsy was lower in women in the magnesium sulfate group compared with placebo (1.9% vs 3.5%; RR, 0.6; 95% CI, 0.3e1.0). Although antenatal magnesium sulfate for neuroprotection in high-risk patients has become the standard of care, optimal dosing has not been defined in general or selected populations. In 2014, McPherson et al9 found that the duration of magnesium sulfate given for neuroprotection was not associated with the rates of cerebral palsy, but specific dosing recommendations could not be made, especially for a specialized population such as obese patients. There is evidence that obesity leads to increased rates of cerebral palsy secondary to increased susceptibility of fetal brain injury related to the inflammatory environment. In 2013, Crisham Janik et al2 performed a retrospective cohort study linking cases of childhood cerebral palsy to maternal obesity and found that obese women had 27% increase in the odds of having a child with cerebral palsy (95% CI, 1.1e1.5), and morbidly obese women had a >2.5-fold increase in the odds of having a child with cerebral palsy (95% CI, 1.8e3.7). In 2014, Pan et al4 found that morbidly obese women were at 3 times increased odds of having a child with cerebral palsy (95% CI, 1.1e8.4). Our study has several strengths. The prospective cohort is derived from a wellcharacterized, comprehensive multicenter clinical trial study population
Obstetrics
ajog.org
Research
TABLE 2
Outcomes of obese vs nonobese women receiving magnesium sulfate (n [ 806) Outcome
Obese (n [ 199)
Nonobese (n [ 607)
RR (95% CI)
aORa (95% CI)
Any cerebral palsy or death
28 (14.1%)
54 (8.9%)
1.60 (1.03e2.43)
1.27 (0.76e2.14)
Moderate to severe cerebral palsy or death
24 (12.1%)
44 (7.2%)
1.66 (1.04e2.66)
1.31 (0.75e2.28)
Any cerebral palsy
7 (3.5%)
17 (2.8%)
1.26 (0.52e2.98)
1.04 (0.42e2.59)
Moderate to severe cerebral palsy
3 (1.5%)
6 (0.9%)
1.53 (0.38e6.04)
1.27 (0.31e5.25)
21 (10.6%)
40 (6.6%)
1.60 (0.97e2.66)
1.22 (0.68e2.21)
b
Death
Obese indicates a BMI of 30 kg/m ; nonobese indicates a BMI <30 kg/m . 2
2
aOR, adjusted odds ratio; CI, confidence interval; RR, relative risk. a
Adjusted for gestational age at delivery in weeks; b Any cerebral palsy includes cerebral palsy that precedes death.
McPherson. Neuroprotective magnesium sulfate and obesity. Am J Obstet Gynecol 2015.
that has been previously validated. Information bias is minimized, given the prospective nature of the data collection. The retrospective design of this analysis, however, is a potential limitation because we are restricted to variables that exist in the established database and there are missing data, however, rarely greater than 5% within a single variable. One notable variable that was not available is gestational weight gain. Therefore, we were unable to assess the independent or additive effect of increased gestational weight gain on magnesium sulfate neuroprotective efficacy or neonatal risk. Notably, the study groups defined by maternal BMI differed in several baseline characteristics. Multivariable regression analysis, however, was used to adjust for these measured differences, minimizing
confounding and leaving only unmeasured potential residual bias. Although our main analysis indicates that maternal obesity does not affect the neuroprotective effect of antenatal magnesium sulfate, a subanalysis of morbidly obese women suggests that the risk for infant cerebral palsy may be increased in those who deliver after 28 gestational weeks. We hypothesize that the excess cerebral palsy risk associated with morbid obesity was observed only after 28 gestational weeks because prior to 28 weeks, the majority of risk is likely attributed to extreme prematurity rather than maternal characteristics. Owing to the small case numbers from stratification and the potential for type 1 error with multiple comparisons, we consider this observation as hypothesis generating and recommend that this
specific population be studied more rigorously before concluding that magnesium sulfate dosing must be increased in the morbidly obese patient at risk for preterm delivery. Based on the known effects of morbid obesity on drug levels in general and on the inflammatory milieu, it is biologically plausible that neuroprotective efficacy could be reduced. And given the increasing prevalence of morbid obesity compounded with our observations, this is an important subgroup to study further. In conclusion, we did not observe an excess of infant cerebral palsy among obese women who received antenatal magnesium sulfate. Therefore, we cannot recommend an escalated magnesium sulfate dosing for obese patients who meet criteria for antenatal neuroprotective therapy. -
TABLE 3
Outcomes of morbidly obese vs nonmorbidly obese women receiving magnesium sulfate (n [ 806) Outcome
Morbidly obese (n [ 31)
Nonmorbidly obese (n [ 775)
RR (95% CI)
aORa (95% CI)
Any cerebral palsy or death
7 (22.6%)
75 (9.7%)
2.33 (1.17e4.64)
2.24 (0.86e5.82)
Moderate to severe cerebral palsy or death
5 (16.1%)
63 (8.1%)
1.98 (0.85e4.58)
1.68 (0.68e4.87)
Any cerebral palsyb
3 (9.6%)
21 (2.7%)
3.5 (1.12e11.34)
3.28 (0.90e11.96)
Moderate to severe cerebral palsy
1 (3.2%)
8 (1.0%)
3.13 (0.40e24.22)
2.67 (0.32e22.61)
Death
4 (12.9%)
57 (7.34%)
1.76 (0.68e4.54)
1.39 (0.44e4.42)
Morbidly obese indicates a BMI of 40 kg/m2; nonmorbidly obese indicates a BMI <40 kg/m2. aOR, adjusted odds ratio; CI, confidence interval; RR, relative risk. a
Adjusted for gestational age at delivery; b Any cerebral palsy includes cerebral palsy that precedes death.
McPherson. Neuroprotective magnesium sulfate and obesity. Am J Obstet Gynecol 2015.
MONTH 2015 American Journal of Obstetrics & Gynecology
1.e5
Research
Obstetrics
ACKNOWLEDGMENT We acknowledge the Eunice Kennedy Shriver National Institute of Child Health and Human Development, the Maternal-Fetal Medicine Units Network, and the Protocol Subcommittee for making this database publicly available.
REFERENCES 1. American College of Obstetricians and Gynecologists. Obesity in pregnancy. ACOG Committee opinion no. 549. Obstet Gynecol 2013;121:213-7. 2. Crisham Janik MD, Newman TB, Cheng YW, Xing G, Gilbert WM, Wu YW. Maternal diagnosis
ajog.org of obesity and risk of cerebral palsy in the child. J Pediatr 2013;163:1307-12. 3. Bergehella V. Obstetric evidence based guidelines, 2nd ed. New York, NY: Informa Healthcare; 2012. 4. Pan C, Deroche CB, Mann JR, McDermott S, Hardin JW. Is pregnancy obesity associated with risk of cerebral palsy and epilepsy in children? J Child Neorol 2014;29: NP196-201. 5. Rouse DJ, Hirtz DG, Thom E, et al. A randomized, controlled trial of magnesium sulfate for the prevention of cerebral palsy. N Engl J Med 2008;359:895-905. 6. American College of Obstetricians and Gynecologists. Magnesium sulfate before anticipated preterm birth for neuroprotection. ACOG
1.e6 American Journal of Obstetrics & Gynecology MONTH 2015
Committee opinion no. 455. Obstet Gynecol 2010;116:669-71. 7. Sugimoto J, Romani A, Valenti-Torres A, et al. Magnesium decreases inflammatory cytokine production: a novel innate immunomodulatory mechanism. J Immunol 2012;188:6338-46. 8. Suzuki-Kaksaka H, Sugimoto J, Tetarbe M, Romani A, Ramirez Kitchen C, Berstein H. Magnesium sulfate increases intracellular magnesium reducing inflammatory cytokine release in neonates. Am J Reprod Immunol 2013;70: 213-20. 9. McPherson J, Rouse D, Grobman W, Palatnik A, Stamilio D. Association of duration of neuroprotective magnesium sulfate with neonatal and maternal outcomes. Obstet Gynecol 2014;124:749-55.