Abstract
Objective
To test the primary hypothesis that extremely preterm infants antenatally exposed to both magnesium sulfate and antenatal corticosteroids have a lower rate of severe neurodevelopmental impairment or death compared to infants exposed to antenatal corticosteroids alone.
Methods
This was a prospective observational study of infants (n=3,093) 22 0/7 to 26 6/7 weeks of gestational age born from 2011 to 2014 at NICHD Neonatal Research Network hospitals. The primary outcome was severe neurodevelopmental impairment or death at 18–26 months of corrected age follow-up based on exposure to antenatal corticosteroids and magnesium sulfate or antenatal corticosteroids alone. Secondary outcomes included components of severe neurodevelopmental impairment by exposure group and comparisons of severe neurodevelopmental impairment or death between infants exposed to both antenatal corticosteroids and magnesium sulfate to infants exposed to magnesium sulfate alone or to neither antenatal corticosteroids nor magnesium sulfate. Logistic regression models adjusted for background characteristics.
Results
Infants exposed to both antenatal corticosteroids and magnesium sulfate had a lower rate of severe neurodevelopmental impairment or death 813/2239, 36.3%) compared to infants exposed to antenatal corticosteroids alone (225/508, 44.3%; aOR 0.73; 95% CI 0.58–0.91), magnesium sulfate alone (47/89, 53%; aOR 0.49; 95% CI 0.29–0.82, or neither therapy (121/251; 48.2%; aOR 0.66, 95% CI 0.49–0.89). Similarly, infants exposed to both antenatal corticosteroids and magnesium sulfate had a lower rate of death compared to either or neither therapy but the rate of severe neurodevelopmental impairment among survivors did not differ between exposure groups.
Conclusion
In infants born between 22 0/7 to 26 6/7 weeks of gestation, exposure to both antenatal corticosteroids and magnesium sulfate was associated with lower rates of severe neurodevelopmental impairment or death and death compared to exposure to antenatal corticosteroids alone.
Clinical Trial Registration
Précis
In extremely preterm infants, exposure to both antenatal corticosteroids and magnesium sulfate was associated with less severe neurodevelopmental impairment or death compared to antenatal corticosteroid exposure alone
INTRODUCTION
Antenatal therapies offer a unique opportunity to improve neonatal outcomes in preterm infants. Antenatal corticosteroids and magnesium sulfate benefit preterm infants.1,2 In meta-analyses of large randomized controlled trials, antenatal magnesium sulfate reduced cerebral palsy but not the composite outcome of cerebral palsy or death due to a trend for higher mortality in the magnesium sulfate group.3–6 Antenatal corticosteroids reduced mortality, respiratory distress syndrome, necrotizing enterocolitis, and intracranial hemorrhage (ICH) in meta-analyses and have been associated with lower rates of adverse neurological outcomes in observational studies.2,7–9 As both antenatal corticosteroids and magnesium sulfate may offer neurological benefit, the combination of fetal exposure to both therapies may have additive value.
It is unlikely that trials comparing the effect of both treatments will be performed to determine if antenatal corticosteroids and magnesium sulfate in combination or alone are beneficial to preterm infants, as both treatments are recommended by the American College of Obstetricians and Gynecologists10,11 and the National Institute for Health and Care Excellence12 for infants born between 24 0/7 weeks and 33 6/7 weeks of gestation. This study used prospectively collected observational research data to compare the rate of severe neurodevelopmental impairment or death between infants exposed to both antenatal corticosteroids and magnesium sulfate and those exposed to antenatal corticosteroids alone. Our primary hypothesis was that in preterm infants with a gestational age between 22 0/7 and 26 6/7 weeks, there would be lower rates of death or severe neurodevelopmental impairment in infants exposed to both antenatal corticosteroids and magnesium sulfate compared to infants exposed to antenatal corticosteroids alone.
METHODS
We analyzed prospectively collected observational research data from the National Institute of Child Health and Human Development Neonatal Research Network (NRN) Generic Database and Follow-Up Study for infants born between January 1, 2011, and December 31, 2014 with initial analyses beginning in September 2018. Each participating center received Institutional Review Board approval for data collection. Magnesium sulfate exposure was first documented within the NRN in 2011, precluding an earlier window of analysis. Infants included in the study were 22 0/7 to 26 6/7 weeks of gestation based on the best estimate of gestational age in weeks and days, using a hierarchy of best obstetrical estimate (based on last menstrual period, obstetrical parameters, and/or prenatal ultrasound) over best neonatal estimate (including physical criteria and Ballard or Dubowitz examinations). Infants were excluded if born with a major congenital anomaly or if death occurred during the first 12 hours after birth without receiving delivery room resuscitation (no ventilation, intubation, or medications) to ensure results were not affected by postnatal restriction of care. Infants born from January 1, 2011 to June 30, 2012 received comprehensive neurodevelopmental follow-up at 18–22 months’ corrected age while those born from July 1, 2012, to December 31, 2014 received follow-up at 22–26 months’ corrected age. Neurodevelopmental follow-up assessments included the Bayley Scales of Infant and Toddler Development, 3rd edition (Bayley-III),13 administered by certified psychologists or developmental specialists. Certified examiners assessed for cerebral palsy by neurological examination including evaluation of muscle tone, movements, reflexes, protective reactions, and gross and fine motor function with annual certification practices as previously described.14,15 Severity of motor impairment was on the basis of the Gross Motor Function Classification System (GMFCS) levels.16,17 Cerebral palsy severity was defined by GMFCS level with mild as level 1, moderate as level 2 or 3, and severe as level 4 or 5.18
Antenatal corticosteroid exposure was defined as maternal receipt of either betamethasone or dexamethasone prior to her infant’s birth. Infants were considered exposed to magnesium sulfate if mothers received magnesium sulfate during the hospitalization for delivery (without further specification of indication for pre-eclampsia or fetal neuroprotection). Severe neurodevelopmental impairment was defined as one or more of the following: severe cerebral palsy (Gross Motor Function Classification System level 4–5), motor or cognitive composite score <70 on the Bayley-III exam, bilateral blindness (no useful vision in either eye), or bilateral severe functional hearing impairment with or without amplification. Grade 3–4 ICH was defined as ventricular enlargement with concurrent or prior blood in the ventricles or blood or echodensity in the parenchyma. Any ICH was defined as either of the aforementioned features of grade 3–4 ICH or blood or echodensity in the germinal matrix or ventricles.
This analysis sought to test the primary hypothesis that antenatal exposure to both magnesium sulfate and antenatal corticosteroids is associated with a lower rate of severe neurodevelopmental impairment or death compared to exposure to antenatal corticosteroids alone. Initial sample size and power estimates using existing NRN data indicated sufficient power for testing. Based on the frequency of the primary outcome of death or severe neurodevelopmental impairment according to follow-up report (43%), a comparison of infants exposed to both antenatal corticosteroids and magnesium sulfate (N=2,243) to infants exposed to antenatal corticosteroids alone (N=509) would have >99% power to detect a hypothesized odds ratio in the primary outcome of 0.53 (corresponding to a Cohen’s d of 0.35, assuming a two-sided alpha level and the inclusion of 10 control variables in the models). Secondary pre-specified contrasts included a comparison of infants exposed to magnesium sulfate but not antenatal corticosteroids, and those who were exposed to neither drug, with infants exposed to both antenatal corticosteroids and magnesium sulfate, thus yielding four exposure groups of interest: (1) antenatal corticosteroids and magnesium sulfate, (2) antenatal corticosteroids only, (3) antenatal magnesium sulfate only, and (4) no antenatal corticosteroids or magnesium sulfate. Pre-specified secondary outcomes included grade 3–4 ICH, death before discharge, grade 3–4 ICH or death, death before follow-up, severe neurodevelopmental impairment, moderate to severe cerebral palsy, and moderate to severe cerebral palsy or death. Additional, exploratory analyses examined rates of severe neurodevelopmental impairment or death, death by follow up, and severe neurodevelopmental impairment alone by gestational age (22–24 weeks of gestation and 25–26 weeks of gestation).
We conducted bivariate analyses to compare the baseline characteristics and in-hospital morbidities of infants of the four exposure groups to identify any systematic differences between the groups, using Chi-squared test for categorical variables, ANOVA for continuous normally distributed variables, and Kruskal-Wallis test for continuous skewed variables. These characteristics included gestational age, race, birth weight, small for gestational age, sex, and as surrogates for socioeconomic status, maternal education and no maternal private health insurance. In-hospital co-morbidity indicators included sepsis, meningitis, cardiac resuscitation, postnatal steroid use, indomethacin exposure within 24 hours after birth and necrotizing enterocolitis (modified Bell’s criteria,16 stage 2 or 3). Analyses of pre-specified outcomes and contrasts were performed using logistic regression models, adjusting for baseline characteristics that differed significantly across the groups (at p<0.15) and including study center as a random effect. Results of a global test of any differences in outcomes among the four exposure groups were also reported. Models were adjusted for year of birth to account for temporal trends as well as the change in timing of developmental follow-up (from 18–22 months to 22–26 months). Significant co-morbidities were not adjusted for in modeling, in order to retain focus on relationships between antenatal exposures and outcomes. All analyses utilized a 2-sided alpha level of 0.05, and were conducted with SAS software version 9.4. As an exploratory analysis, a test for interaction between exposure and gestational age was also assessed for all outcomes. If the test for interaction was close to significant at p<0.10, stratified models by ‘low’ versus ‘high’ gestational age were presented. Gestational age cut-offs were selected in an effort to retain a maximum sample size within each stratum.
RESULTS
From January 1, 2011,to December 31, 2014, 3,816 infants at 22 0/7 to 26 6/7 weeks of gestation were born at participating NRN sites. One hundred fifty (3.9%) infants had major congenital anomalies, 214 (5.6%) infants had missing data on antenatal corticosteroids or magnesium sulfate exposure, and 359 (9.4%) had missing data on the follow-up status at 2 years. Of the remaining 3,093 infants, 2,243 (72.5%) infants were exposed to both antenatal corticosteroids and magnesium sulfate, 509 (16.4%) infants were exposed to antenatal corticosteroids alone, 89 (2.9%) were exposed to magnesium sulfate alone, and 252 (8.1%) were exposed to neither antenatal corticosteroids nor magnesium sulfate. Baseline characteristics that differed between subgroups included gestational age, small for gestational age, education level, and maternal insurance status (Table I). Education level was not included in adjustment modeling as data were not available for 248/3093 (8%) infants, and it was highly correlated with insurance status (p<0.001) as a measure of socioeconomic status. Short term morbidities and early postnatal exposures are compared in Table II. Postnatal corticosteroid use and hypotension treatment differed between groups (Table II). For adjustment models for severe neurodevelopmental impairment or death, severe neurodevelopmental impairment alone, and death, data were available for 2981 infants, 2242 infants, and 2988 infants respectively.
Table I:
Demographic and Clinical Characteristics by Exposure Group
| Characteristic* | Antenatal Corticosteroids + Magnesium Sulfate (n=2243) | Antenatal Corticosteroids Only (n=509) | Magnesium Sulfate Only (n=89) | No Antenatal Corticosteroids or Magnesium Sulfate (n=252) | p-value |
|---|---|---|---|---|---|
| Birth weight, g M (IQR) | 720 (605–840) | 720 (590–850) | 720 (600–840) | 733 (601–850) | 0.86 |
| Gestational age M (IQR) | 25.3 (24.4–26.1) | 25.1 (24.3–26.0) | 25.1 (24.1–26.1) | 25.0 (23.8–25.9) | <0.001 |
| Male sex n (%) | 1193 (53.2) | 241 (47.5) | 49 (55.1) | 127 (50.4) | 0.11 |
| Nonwhite race n (%) | 1095 (50.2) | 248 (50.2) | 44 (50.0) | 142 (58.4) | 0.11 |
| Multiple births n (%) | 581 (25.9) | 145 (28.5) | 17 (19.1) | 67 (26.6) | 0.28 |
| SGA n (%) | 205 (9.2) | 43 (8.5) | 7 (7.9) | 6 (2.4) | <0.001 |
| Maternal education high school or less n (%) |
383 (18.2) | 100 (22.1) | 23 (30.3) | 62 (28.8) | <0.001 |
| No Maternal Private Insurance n (%) | 1371 (61.3) | 311 (61.8) | 71 (79.8) | 185 (73.4) | <0.001 |
SGA, small for gestational age.
Data are median (interquartile range) or n (%) unless otherwise specified. Data missing for <1% of each characteristic with the exception of nonwhite race (3%) and maternal education (8%)
Table II:
Co-morbidities and Interventions of Enrolled Patients by Antenatal Corticosteroid and Magnesium Sulfate Exposure
| Antenatal Corticosteroids +Magnesium Sulfate | Antenatal Corticosteroids Only | Magnesium Sulfate Only | No Antenatal Corticosteroids or Magnesium Sulfate | p-value | |
|---|---|---|---|---|---|
| Characteristic* | n (%) | n (%) | n (%) | n (%) | |
| Postnatal steroid use | 484 (23.3) | 98 (20.4) | 12 (14.6) | 35 (14.8) | 0.005 |
| Early onset culture positive sepsis | 53 (2.4) | 19 (3.7) | 1 (1.1) | 12 (4.8) | 0.057 |
| Late onset culture positive sepsis | 638 (29.8) | 127 (26.5) | 20 (26.3) | 50 (23.6) | 0.157 |
| Meningitis | 42 (1.9) | 5 (1) | 0 (0) | 1 (0.4) | 0.149 |
| Hypotension treatment | 719 (32.1) | 196 (38.5) | 37 (41.6) | 139 (55.2) | <0.001 |
| Indomethacin exposure† | 894 (39.9) | 217 (42.6) | 28 (31.5) | 104 (41.3) | 0.229 |
| Necrotizing Enterocolitis‡ | 249 (11.1) | 64 (12.6) | 10 (11.2) | 21 (8.3) | 0.378 |
| Physiologic BPD§ | 1115 (61.0) | 243 (64.0) | 33 (55.9) | 96 (58.2) | 0.459 |
BPD, bronchopulmonary dysplasia.
Data missing for <1% of each characteristic with the exception of postnatal steroid use (7%) and late onset culture positive sepsis (6%). Physiologic BPD is only assessed among survivors to 36 weeks’ PMA (n=2431), among whom data was missing for <1%.
Indicates indomethacin exposure within first 24 hours of life
Indicates proven NEC, with or without surgery
Defined at 36 weeks’ corrected age as positive pressure support, supplemental FiO2 > 0.30, or supplemental FiO2 ≤ 0.30 with failed room air challenge
Descriptive details on the distributions of outcomes among subgroups are presented in Table III, while Table IV presents results of adjustment modeling. Infants exposed to both antenatal corticosteroids and magnesium sulfate had a lower rate of severe neurodevelopmental impairment or death (813/2239, 36.3%) compared to infants exposed to antenatal corticosteroids only (225/508, 44.3%; aOR 0.73; 95% CI 0.58–0.91), as well as to the additional reference groups of magnesium sulfate only (47/89, 53%; aOR 0.49; 95% CI 0.29–0.82), or neither therapy (121/251; 48.2%; aOR 0.66, 95% CI 0.49–0.89) (Tables III and IV). Infants exposed to both antenatal corticosteroids and magnesium sulfate had a lower rate of death before follow-up (505/2243, 22.5%) compared to infants exposed to antenatal corticosteroids only (149/509, 29.3%; aOR 0.69; 95% CI 0.54–0.89), magnesium sulfate only (32/89, 36%; aOR 0.55; 95% CI 0.32–0.93), or neither therapy (93/252, 36.9%; aOR 0.51; 95% CI 0.38–0.70). The rate of severe neurodevelopmental impairment among survivors did not differ between exposure groups. An exploratory assessment of any interaction between exposure group and gestational age, indicated possible differential effects of exposure by gestational age on the outcome of neurodevelopmental impairment (p=0.04) or the primary outcome of death or severe neurodevelopmental impairment (p=0.06). In particular, infants born at 25–26 weeks of gestation exposed to both antenatal corticosteroids and magnesium sulfate had lower rates of severe neurodevelopmental impairment or death (368/1381, 26.7%) compared to infants exposed to antenatal corticosteroids alone (100/287, 34.8%; aOR 0.67; 95% CI 0.49–0.90) (Supplemental Table I-II and Table V), while no significant difference in rates was observed between these two exposure groups among infants born at 22–24 weeks of gestation.
Table III:
Neurodevelopmental Outcomes by Antenatal Corticosteroid and Magnesium Sulfate Exposure
| Antenatal Corticosteroids + Magnesium Sulfate | Antenatal Corticosteroids Only | Magnesium Sulfate Only | No Antenatal Corticosteroid or Magnesium Sulfate | |
|---|---|---|---|---|
| Outcome | n (%) | n (%) | n (%) | n (%) |
| Outcomes at Discharge* | ||||
| Grade 3–4 ICH | 369 (17.1) | 94 (19.4) | 28 (34.6) | 91 (38.6) |
| Any ICH† | 718 (33.2) | 168 (34.7) | 42 (51.9) | 134 (56.8) |
| Death | 475 (21.2) | 140 (27.5) | 30 (33.7) | 89 (35.3) |
| Any ICH or death | 998 (44.8) | 255 (50.3) | 56 (62.9) | 166 (65.9) |
| Any ICH in survivors | 523 (29.8) | 115 (31.3) | 26 (44.1) | 77 (47.2) |
| Grade 3–4 ICH or death | 719 (32.3) | 201 (39.6) | 45 (50.6) | 133 (52.8) |
| Grade 3–4 ICH in survivors | 244 (13.9) | 61 (16.6) | 15 (25.4) | 44 (27.0) |
| Outcomes at Follow-Up‡ | ||||
| Severe NDI§ or death | 813 (36.3) | 225 (44.3) | 47 (52.8) | 121 (48.2) |
| Severe NDI§ | 309 (17.9) | 76 (21.2) | 15 (26.3) | 28 (17.7) |
| Death | 505 (22.5) | 149 (29.3) | 32 (36.0) | 93 (36.9) |
| Severe cerebral palsy|| | 49 (2.8) | 12 (3.3) | 1 (1.8) | 6 (3.8) |
| Moderate to severe cerebral palsy¶ | 117 (6.7) | 35 (9.7) | 4 (7.1) | 13 (8.2) |
| Moderate to severe cerebral palsy¶ or death | 621 (27.7) | 184 (36.2) | 36 (40.9) | 106 (42.1) |
| Bayley-3 Cognitive Composite Score <70 | 203 (11.8) | 43 (12.1) | 9 (15.8) | 14 (8.9) |
| Bayley-3 Motor Composite Score <70 | 225 (13.1) | 59 (16.6) | 11 (19.3) | 22 (14.1) |
| Blindness# | 12 (0.7) | 1 (0.3) | 0 (0) | 0 (0) |
| Hearing impairment** | 51 (2.9) | 8 (2.2) | 1 (1.8) | 6 (3.8) |
ICH, intracranial hemorrhage; NDI, neurodevelopmental impairment.
Data are n (%). Data missing for <1% of each outcome with the exception of ICH (4%).
Indicates: blood/echodensity in the germinal matrix (Grade 1), or blood echodensity in the ventricle (Grade 2), or ventricular size enlarged and blood/echodensity in the ventricle (Grade 3), or blood/echodensity in the parenchyma (Grade 4)
Among survivors to discharge, data missing for <1% of each outcome with the exception of severe NDI (2%), severe cerebral palsy (2%), Bayley-3 Cognitive Composite Score<70 (3%), blindness (2%), and hearing impairment (2%). Reasons NDI or components of NDI could be missing (a) child died, (b) child not present at follow-up, (c) child did not have all of the information to determine NDI
Severe neurodevelopmental impairment defined as ≥ 1 of the following: severe cerebral palsy, gross motor function level ≥ 4–5, motor and/or cognitive composite score < 70 on Bayley Scales of Infant Development III, bilateral blindness (no useful vision in either eye), or bilateral hearing impairment even with amplification.
Gross Motor Function Classification System level 4–5
Gross Motor Function Classification System level 2–5
Indicates whether infant has bilateral blindness with no useful vision in either eye (the 2002+ definition)
Indicates whether infant has any hearing impairment even with amplification (2006+ definition)
Table IV.
Odds Ratios of Neurodevelopmental Outcomes by Antenatal Corticosteroid and Magnesium Sulfate Exposure
| Outcome | Antenatal Corticosteroids + Magnesium Sulfate (vs. Antenatal Corticosteroids Only) | Antenatal Corticosteroids + Magnesium Sulfate (vs. Magnesium Sulfate Only) | Antenatal Corticosteroids + Magnesium Sulfate (vs. No Antenatal Corticosteroids or Magnesium Sulfate) | |||
|---|---|---|---|---|---|---|
| Unadjusted OR (CI) | Adjusted OR (CI)* | Unadjusted OR (CI) | Adjusted OR (CI)* | Unadjusted OR (CI) | Adjusted OR (CI)* | |
| Outcomes at Discharge | ||||||
| Grade 3–4 ICH or death† | 0.70 (0.57, 0.86) | 0.77 (0.61–0.96) | 0.46 (0.29, 0.72) | 0.48 (0.30–0.79) | 0.40 (0.31, 0.52) | 0.44 (0.33–0.59) |
| Death† | 0.67 (0.53, 0.83) | 0.70 (0.55–0.89) | 0.49 (0.30, 0.80) | 0.57 (0.33–0.97) | 0.44 (0.33, 0.59) | 0.51 (0.37–0.70) |
| Grade 3–4 ICH† | 0.84 (0.65, 1.08) | 0.92 (0.70–1.21) | 0.41 (0.25, 0.68) | 0.41 (0.24–0.69) | 0.33 (0.25, 0.44) | 0.35 (0.26–0.48) |
| Outcomes at Follow-Up | ||||||
| Severe NDI‡ or death† | 0.69 (0.57, 0.85) | 0.73 (0.58–0.91) | 0.47 (0.30, 0.74) | 0.49 (0.29–0.82) | 0.57 (0.44, 0.75) | 0.66 (0.49–0.89) |
| Death† | 0.66 (0.52, 0.82) | 0.69 (0.54–0.89) | 0.47 (0.29, 0.77) | 0.55 (0.32–0.93) | 0.44 (0.34, 0.59) | 0.51 (0.38–0.70) |
| Severe NDI‡ | 0.82 (0.61, 1.09) | 0.81 (0.60–1.10) | 0.56 (0.29, 1.07) | 0.50 (0.25–0.98) | 1.01 (0.65, 1.55) | 0.98 (0.62–1.55) |
| Moderate to severe cerebral palsy or death† | 0.63 (0.51, 0.78) | 0.66 (0.53–0.84) | 0.51 (0.32, 0.82) | 0.55 (0.33–0.92) | 0.48 (0.36, 0.63) | 0.54 (0.40–0.73) |
| Moderate to severe cerebral palsy | 0.64 (0.42, 0.96) | 0.62 (0.40–0.95) | 0.90 (0.31, 2.64) | 0.78 (0.26–2.33) | 0.77 (0.42, 1.41) | 0.72 (0.38–1.34) |
| Severe cerebral palsy | 0.83 (0.43, 1.58) | 0.86 (0.44–1.70) | 1.55 (0.21, 11.52) | 1.48 (0.20–11.18) | 0.74 (0.31, 1.75) | 0.73 (0.30–1.79) |
| Bayley cognitive score <70 | 0.92 (0.64, 1.32) | 0.96 (0.65–1.42) | 0.63 (0.29, 1.39) | 0.55 (0.24–1.23) | 1.38 (0.78, 2.45) | 1.40 (0.76–2.54) |
| Bilateral hearing impairment | 1.67 (0.77, 3.60) | 1.70 (0.78–3.71) | 1.70 (0.21, 13.56) | 1.72 (0.21–14.02) | 0.84 (0.35, 2.02) | 0.84 (0.34–2.08) |
ICH, intracranial hemorrhage; NDI, neurodevelopmental impairment.
All models were simultaneously adjusted for: gestational age (square root transformation), sex, small for gestational age, maternal health insurance status, maternal race, follow-up window, and center (as a random effect). Outcome differences statistically significant if adjusted odds ratios between exposure groups exclude 1.
Significant global test for differences in outcome rates among groups: (1) ANS and MgSO4, (2) ANS only, (3) antenatal MgSO4 only, and (4) no ANS or MgSO4
Severe neurodevelopmental impairment defined as ≥ 1 of the following: severe cerebral palsy, gross motor function level ≥ 4–5, motor and/or cognitive composite score < 70 on Bayley Scales of Infant Development III, bilateral blindness (no useful vision in either eye), or bilateral hearing impairment even with amplification.
Table V.
Odds Ratios of Outcome by Gestational Age and Antenatal Corticosteroid and Magnesium Sulfate Exposure*
| Outcome | Antenatal Corticosteroids + Magnesium Sulfate (vs. Antenatal Corticosteroids Only) | Antenatal Corticosteroids + Magnesium Sulfate (vs. Magnesium Sulfate Only) | Antenatal Corticosteroids + Magnesium Sulfate (vs. No Antenatal Corticosteroids or Magnesium Sulfate) | |||
|---|---|---|---|---|---|---|
| Unadjusted OR (CI) | Adjusted OR (CI)* | Unadjusted OR (CI) | Adjusted OR (CI)* | Unadjusted OR (CI) | Adjusted OR (CI)* | |
| 22–24 Weeks’ Gestation | ||||||
| Severe NDI‡ or death† | 0.79 (0.58, 1.08) | 0.81 (0.58, 1.12) | 0.2 (0.08, 0.5) | 0.21 (0.08, 0.58) | 0.48 (0.31, 0.74) | 0.61 (0.39, 0.97) |
| Severe NDI†‡ | 0.96 (0.62, 1.48) | 0.98 (0.61, 1.55) | 0.2 (0.07, 0.56) | 0.19 (0.06, 0.60) | 0.88 (0.46, 1.69) | 1.01 (0.51, 2.00) |
| 25–26 Weeks’ Gestation | ||||||
| Severe NDI‡ or death† | 0.66 (0.5, 0.87) | 0.67 (0.49, 0.90) | 0.76 (0.41, 1.43) | 0.70 (0.36, 1.36) | 0.73 (0.5, 1.07) | 0.70 (0.47, 1.06) |
| Severe NDI†‡ | 0.74 (0.5, 1.1) | 0.69 (0.46, 1.03) | 1.19 (0.43, 3.29) | 1.00 (0.36, 2.81) | 1.1 (0.6, 2.02) | 0.93 (0.50, 1.73) |
NDI, neurodevelopmental impairment.
All models were simultaneously adjusted for: gestational age (square root transformation), sex, small for gestational age, maternal health insurance status, maternal race, follow-up window, and center (as a random effect). Centers with low frequencies of severe cerebral palsy (n<5) were pooled to counter the effects of small sample sizes. Outcome differences statistically significant if adjusted odds ratios between exposure groups exclude 1.
Significant global test for differences among groups: (1) ANS and MgSO4, (2) ANS only, (3) antenatal MgSO4 only, and (4) no ANS or MgSO4
Severe neurodevelopmental impairment defined as ≥ 1 of the following: severe cerebral palsy, gross motor function level ≥ 4–5, motor and/or cognitive composite score < 70 on Bayley Scales of Infant Development III, bilateral blindness (no useful vision in either eye), or bilateral hearing impairment even with amplification.
Rates of moderate to severe cerebral palsy did not differ between exposure groups (p=0.1479). The rate of grade 3–4 ICH did not differ between infants exposed to both antenatal corticosteroids and magnesium sulfate (369/2163, 17.1%) and infants exposed to antenatal corticosteroids only (94/484, 19.4%; aOR 0.92; 95% CI 0.70–1.21). Lower rates of grade 3–4 ICH were observed in infants exposed to antenatal corticosteroids and magnesium sulfate compared to infants exposed to magnesium sulfate alone (28/81, 31.4%; aOR 0.41; 95% CI 0.24–0.69) and infants exposed to neither therapy (91/236, 38.6%; aOR 0.35; 95% CI 0.26–0.48) (Tables III and IV).
DISCUSSION
Independent data from randomized controlled trials of antenatal corticosteroids and magnesium sulfate during preterm labor indicate that each therapy provides neurological benefit but data from antenatal corticosteroids trials on extremely preterm infants are limited. In the current study, outcomes after exposure to both antenatal corticosteroids and magnesium sulfate were analyzed compared to infants exposed to antenatal corticosteroids alone. Exposure to both antenatal corticosteroids and magnesium sulfate was associated with a lower rate of severe neurodevelopmental impairment or death compared to exposure to antenatal corticosteroids alone. Of the secondary outcomes investigated, infants receiving both antenatal therapies had lower rates of death without an increased rate of severe neurodevelopmental impairment in survivors compared to infants exposed to antenatal corticosteroids alone.
Multiple observational studies have analyzed the association between antenatal corticosteroids exposure and the composite of neurodevelopmental impairment or death, death alone, and different neurodevelopmental outcomes. A recent meta-analysis of largely observational studies (~2/3 of the patients were from the NRN study) and a few small randomized controlled trials reported a lower rate of severe disability in preterm infants exposed to antenatal corticosteroids compared to infants not exposed to antenatal corticosteroids (RR 0.79; 95% CI 0.73–0.85).17 Exposure to antenatal corticosteroids was associated with a lower rate of death in infants between 22 and 29 weeks of gestation compared to infants not exposed in an NRN study (10,972 patients)18 and infants born between 22 and 25 weeks of gestation in the Vermont Oxford Network (29,932 patients).19 Given these data for survival and neuroprotection in infants exposed to antenatal corticosteroids and additional evidence for neuroprotection in magnesium sulfate–exposed infants, the present study conducted analyses not only based on antenatal corticosteroids exposure, but additionally analyzed infants by magnesium sulfate exposure. Whereas exposure to both therapies was associated with a lower rate of death, exposure to magnesium sulfate in addition to antenatal corticosteroids was not associated with a lower rate of severe neurodevelopmental impairment or components of severe neurodevelopmental impairment including Bayley scores, bilateral hearing impairment, and cerebral palsy.
The benefit of antenatal corticosteroids exposure by gestational age has also been previously analyzed. In the NRN observational study of infants born at 22 to 25 weeks of gestation, exposure to antenatal corticosteroids was associated with a lower risk of neurodevelopmental impairment or death compared to unexposed infants (aOR 0.60; 95% CI 0.53–0.69).9 In the present study, differential effects by gestational age were observed for severe neurodevelopmental impairment or death by follow up and death by follow up alone, although these differences were in infants born at 25–26 weeks of gestation wherein exposure to both antenatal corticosteroids and magnesium sulfate had lower rates of severe neurodevelopmental impairment or death (368/1381, 26.7%) compared to infants exposed to antenatal corticosteroids alone (Tables V and Appendix 2), while no significant difference in rates was observed between these two exposure groups among infants born at 22–24 weeks of gestation. However, this possible differential effect should be interpreted with caution, given the decreased sample sizes in the lower gestation group (Appendix 3.
The therapeutic mechanism for neuroprotection in infants exposed to magnesium sulfate is unclear but may result from neuronal stabilization or anti-inflammatory properties.20 A meta-analysis of randomized controlled trials of magnesium sulfate exposure for any indication did not demonstrate an effect on the composite outcome of neurodevelopmental impairment or death (RR 1.00; 95% CI 0.91–1.11) and only reduced cerebral palsy alone (RR 0.64; 95% CI 0.44–0.92) due to a trend for higher mortality in the magnesium sulfate group. However, in the RCTs of magnesium sulfate for neuroprotection, only 25% of infants were born at <30 weeks of gestation and analyses of this subgroup did not show a difference in death or major neurologic disability (RR 1.04: CI 0.86–1.24).1 An individual participant data meta-analysis of these trials similarly found that there was no reduction in major neurosensory disability or death in infants born at <26 weeks of gestation (RR 1.05; 95% CI 0.95–1.16) and 26–27 weeks of gestation (RR 0.93; 95% CI 0.79–1.10).21 Given the limited number of extremely preterm infants included in these randomized trials, the present study provides additional evidence that antenatal magnesium sulfate exposure may improve outcomes in infants born at lower gestational ages. In contrast to the limited evidence from randomized trials at lower gestational ages that did not report a reduction in the composite outcomes of neurodevelopmental impairment or death in infants exposed to magnesium sulfate, in the present study, infants exposed to both antenatal corticosteroids and magnesium sulfate had a lower rate of severe neurodevelopmental impairment or death compared to infants exposed to antenatal corticosteroids alone.
Because ICH has been associated with poor neurodevelopmental outcomes in extremely preterm infants,22,23 this analysis compared rates of grade 3–4 ICH between exposure groups. The rate of grade 3–4 ICH did not differ between infants exposed to both antenatal corticosteroids and magnesium sulfate compared to infants exposed to antenatal corticosteroids alone. These findings further support data from randomized controlled trials showing benefit for antenatal corticosteroids but not for magnesium sulfate. A Cochrane review on the effects of antenatal corticosteroids on ICH in 21 randomized controlled trials demonstrated a significant decrease in both any ICH (RR 0.54; 95% CI 0.43 to 0.69) and severe ICH (RR 0.28; 95% CI 0.16 to 0.50).2 Multiple large cohort studies have also demonstrated lower rates of both ICH and severe ICH in infants exposed to antenatal corticosteroids.8,9 In contrast, the Cochrane review in which five randomized controlled trials of antenatal magnesium sulfate were analyzed (N=6,145) found no reduction in ICH (RR 0.96; 95% CI 0.86 to 1.08;) or grade 3–4 ICH (RR 0.83; 95% CI 0.62 to 1.13).1
Multiple studies have identified discordance in antenatal therapy exposure rates, wherein antenatal corticosteroids exposure rates exceed magnesium sulfate exposure rates. A large cohort study from the Canadian Neonatal Network database in infants born at <28 weeks of gestation born between 2011 and 2014 reported that 79% (1,788/2,260) of infants unexposed to magnesium sulfate were exposed to antenatal corticosteroids.24 In a cohort study of infants born in France during 2011 born at <27 weeks of gestation,25 78.5% of infants were exposed to antenatal corticosteroids compared to only 4% of infants exposed to magnesium sulfate. Only 13% of deliveries were emergent and of deliveries in which there was no antenatal therapy exposure, 26% of deliveries were after 24 hours (a sufficient interval of time for antenatal therapy exposure). A study within the Canadian Neonatal Network analyzed covariates associated with a higher rate of magnesium sulfate, which included antenatal corticosteroids exposure and maternal indication for delivery.26 In the present study, the reason fewer infants within the cohort were exposed to magnesium sulfate (75.4%) compared to the number of infants exposed to antenatal corticosteroids (89.0%) is unknown and the indication for and timing of magnesium sulfate exposure were not available. Importantly, a larger proportion of mothers unexposed to either therapy compared to both therapies received high school or less education or had no maternal private health insurance which may suggest health inequity as a driver for antenatal therapy exposure rates. It is possible that clinical factors (e.g. precipitous delivery and admission in advanced labor) or practice variations resulted in differential exposure to magnesium sulfate or antenatal corticosteroids, but the specific reasons for differential exposure remain unclear as extensive data details were not collected. However, if there is sufficient time to administer antenatal corticosteroids, there should similarly be sufficient time to administer magnesium sulfate. Given the lower rate of severe neurodevelopmental impairment or death in infants exposed to both antenatal corticosteroids and magnesium sulfate in the present study compared to infants exposed to antenatal corticosteroids alone, increasing the rates of magnesium sulfate exposure through quality improvement or other interventions may improve infant outcomes.27
The strengths of the current study are its large sample size with long term follow up, multicenter design, inclusion of lower gestational ages who are known to be at higher risk for neurodevelopmental impairment,28,29 and analysis of recent birth years. As an observational study with baseline differences between the groups, there is risk for selection bias and practice likely varied by center for which analyses adjusted. However, significant outcome differences remained between exposure groups even after adjustments for multiple clinical and sociodemographic differences. Another limitation of the present study’s analyses is that exposure to antenatal corticosteroids was defined as exposure to either a partial or complete course of antenatal corticosteroids. It is possible that many infants exposed to antenatal corticosteroids alone only received a partial course. Infants exposed to a complete course of antenatal corticosteroids have a lower risk of neurodevelopmental impairment or death than infants exposed to a partial course.30 Analysis of published randomized trials of magnesium sulfate dosing analyzing death and severe neurodevelopmental impairment by antenatal corticosteroids exposure would include many older infants at lower risk of severe neurodevelopmental impairment or death, as only 25% of the infants from these trials were <30 weeks of gestation.1 If limited to extremely preterm infants, additional post hoc analysis of randomized trials might further support our findings. In contrast, a pre-study power analysis determined that this study’s window of inclusion provided a sufficient sample size capable of delineating differences in severe neurodevelopmental impairment or death between exposure groups.
Infants born at 22 0/7 to 26 6/7 weeks of gestation who were exposed to both antenatal corticosteroids and magnesium sulfate had lower rates of severe neurodevelopmental impairment or death compared to infants exposed to antenatal corticosteroids alone, magnesium sulfate alone, or neither therapy. Compared to other exposure groups, infants exposed to both antenatal corticosteroids and magnesium sulfate had a lower rate of death with no difference in severe neurodevelopmental impairment among survivors. Increased rates of magnesium sulfate exposure through quality improvement or other interventions may improve severe neurodevelopmental impairment or death without increasing the risk of neurodevelopmental impairment in survivors.
Supplementary Material
Acknowledgments
Funding Source:
The National Institutes of Health and the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) (U10 HD21373, UG1 HD21364, UG1 HD21385, UG1 HD27851, UG1 HD27853, UG1 HD27856, UG1 HD27880,UG1 HD27904, UG1 HD34216, UG1 HD36790, UG1 HD40492, UG1 HD40689, UG1 HD53089, UG1 HD53109, UG1 HD68244, UG1 HD68270, UG1 HD68278, UG1 HD68263, UG1 HD68284; UG1 HD87226, UG1 HD87229) and the National Center for Advancing Translational Sciences (NCATS) (UL1 TR6, UL1 TR41, UL1 TR42, UL1 TR77, UL1 TR93, UL1 TR105, UL1 TR442, UL1 TR454, UL1 TR1117, provided grant support for the Neonatal Research Network, including for the Follow-up Study.
The National Institutes of Health and the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) provided grant support for the Neonatal Research Network, including for the Follow-up Study.
Data collected at participating sites of the NICHD Neonatal Research Network (NRN) were transmitted to RTI International, the data coordinating center (DCC) for the network, which stored, managed and analyzed the data for this study. On behalf of the NRN, Dr. Abhik Das (DCC Principal Investigator) and Mr. Scott McDonald (DCC Statistician) had full access to all the data in the study and take responsibility for the integrity of the data and accuracy of the data analysis.
Footnotes
Financial Disclosure
Waldemar Carlo disclosed that they are a consultant for Mednax, and they receive expenses reimbursed for travel to board meetings. Roy Heyne disclosed Catholic Health Professionals of Houston paid honorarium for an ethics talk that he gave to this nonprofit entity last year. He is nonpaid board member of nonprofit Catholic Physicians Guild of Dallas, nonpaid regional director of Catholic Medical Association, and a nonpaid board member and officer of Low Birth Weight Development Center in Dallas. The other authors did not report any potential conflicts of interest.
Each author has confirmed compliance with the journal’s requirements for authorship.
Authors’ Data Sharing Statement
Will individual participant data be available (including data dictionaries)? No.
What data in particular will be shared? Not available.
What other documents will be available? Not available.
When will data be available (start and end dates)? Not applicable.
By what access criteria will data be shared (including with whom, for what types of analyses, and by what mechanism)? Data reported in this paper may be requested through a data use agreement. Further details are available at https://neonatal.rti.org/index.cfm?fuseaction=DataRequest.Home
REFERENCES
- 1.Doyle LW, Crowther CA, Middleton P, Marret S, Rouse D. Magnesium sulphate for women at risk of preterm birth for neuroprotection of the fetus. Cochrane Database Syst Rev. 2009(1):CD004661. [DOI] [PubMed] [Google Scholar]
- 2.Roberts D, Dalziel S. Antenatal corticosteroids for accelerating fetal lung maturation for women at risk of preterm birth. The Cochrane database of systematic reviews. 2006(3):CD004454. [DOI] [PubMed] [Google Scholar]
- 3.Crowther CA, Hiller JE, Doyle LW, Haslam RR, Australasian Collaborative Trial of Magnesium Sulphate Collaborative G. Effect of magnesium sulfate given for neuroprotection before preterm birth: a randomized controlled trial. JAMA. 2003;290(20):2669–2676. [DOI] [PubMed] [Google Scholar]
- 4.Rouse DJ, Hirtz DG, Thom E, et al. A randomized, controlled trial of magnesium sulfate for the prevention of cerebral palsy. The New England journal of medicine. 2008;359(9):895–905. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Marret S, Marpeau L, Zupan-Simunek V, et al. Magnesium sulphate given before very-preterm birth to protect infant brain: the randomised controlled PREMAG trial. BJOG : an international journal of obstetrics and gynaecology. 2007;114(3):310–318. [DOI] [PubMed] [Google Scholar]
- 6.Costantine MM, Weiner SJ, Eunice Kennedy Shriver National Institute of Child H, Human Development Maternal-Fetal Medicine Units N. Effects of antenatal exposure to magnesium sulfate on neuroprotection and mortality in preterm infants: a meta-analysis. Obstetrics and Gynecology. 2009;114(2 Pt 1):354–364. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ambalavanan N, Van Meurs KP, Perritt R, et al. Predictors of death or bronchopulmonary dysplasia in preterm infants with respiratory failure. J Perinatol. 2008;28(6):420–426. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Mori R, Kusuda S, Fujimura M, Neonatal Research Network J. Antenatal corticosteroids promote survival of extremely preterm infants born at 22 to 23 weeks of gestation. The Journal of Pediatrics. 2011;159(1):110–114 e111. [DOI] [PubMed] [Google Scholar]
- 9.Carlo WA, McDonald SA, Fanaroff AA, et al. Association of antenatal corticosteroids with mortality and neurodevelopmental outcomes among infants born at 22 to 25 weeks’ gestation. JAMA. 2011;306(21):2348–2358. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.American College of Obstetricians and Gynecologists Committee on Obstetric Practice, Society for Maternal-Fetal Medicine. Committee Opinion No. 455: Magnesium sulfate before anticipated preterm birth for neuroprotection. Obstetrics and Gynecology. 2010;115(3):669–671. [DOI] [PubMed] [Google Scholar]
- 11.Committee on Obstetric Practice. Committee Opinion No. 713: Antenatal corticosteroid therapy for fetal maturation. Obstetrics and Gynecology. 2017;130(2):e102–e109. [DOI] [PubMed] [Google Scholar]
- 12.National Institute for Health and Care Excellence NICE Guidelines (NG25). Preterm labour and birth. 2015. www.nice.org/uk/guidance/ng25. [PubMed]
- 13.Bayley N Bayley Scales of Infant and Toddler Development-Third Edition. San Antonio, TX: Harcourt Assessment, Inc; 2006. [Google Scholar]
- 14.Gargus RA, Vohr BR, Tyson JE, et al. Unimpaired outcomes for extremely low birth weight infants at 18 to 22 months. Pediatrics. 2009;124(1):112–121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Newman JE, Bann CM, Vohr BR, et al. Improving the Neonatal Research Network annual certification for neurologic examination of the 18–22 month child. The Journal of Pediatrics. 2012;161(6):1041–1046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Bell MJ, Ternberg JL, Feigin RD, et al. Neonatal necrotizing enterocolitis. Therapeutic decisions based upon clinical staging. Ann Surg. 1978;187(1):1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Sotiriadis A, Tsiami A, Papatheodorou S, Baschat AA, Sarafidis K, Makrydimas G. Neurodevelopmental outcome after a single course of antenatal steroids in children born preterm: A systematic review and meta-analysis. Obstetrics and Gynecology. 2015;125(6):1385–1396. [DOI] [PubMed] [Google Scholar]
- 18.Travers CP, Carlo WA, McDonald SA, et al. Mortality and pulmonary outcomes of extremely preterm infants exposed to antenatal corticosteroids. Am J Obstet Gynecol. 2018;218(1):130 e131–130 e113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ehret DEY, Edwards EM, Greenberg LT, et al. Association of antenatal steroid exposure with survival among infants receiving postnatal life support at 22 to 25 weeks’ gestation. JAMA Netw Open. 2018;1(6):e183235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Costantine MM, Drever N. Antenatal exposure to magnesium sulfate and neuroprotection in preterm infants. Obstet Gynecol Clin North Am. 2011;38(2):351–366, xi. [DOI] [PubMed] [Google Scholar]
- 21.Crowther CA, Middleton PF, Voysey M, et al. Assessing the neuroprotective benefits for babies of antenatal magnesium sulphate: An individual participant data meta-analysis. PLoS Med. 2017;14(10):e1002398. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Futagi Y, Toribe Y, Ogawa K, Suzuki Y. Neurodevelopmental outcome in children with intraventricular hemorrhage. Pediatr Neurol. 2006;34(3):219–224. [DOI] [PubMed] [Google Scholar]
- 23.Sherlock RL, Anderson PJ, Doyle LW, Victorian Infant Collaborative Study G. Neurodevelopmental sequelae of intraventricular haemorrhage at 8 years of age in a regional cohort of ELBW/very preterm infants. Early Hum Dev. 2005;81(11):909–916. [DOI] [PubMed] [Google Scholar]
- 24.Shalabi M, Mohamed A, Lemyre B, et al. Antenatal Exposure to Magnesium Sulfate and Spontaneous Intestinal Perforation and Necrotizing Enterocolitis in Extremely Preterm Neonates. American Journal of Perinatology. 2017;34(12): 1227–1233. [DOI] [PubMed] [Google Scholar]
- 25.Diguisto C, Foix L’Helias L, Morgan AS, et al. Neonatal Outcomes in Extremely Preterm Newborns Admitted to Intensive Care after No Active Antenatal Management: A Population-Based Cohort Study. The Journal of Pediatrics. 2018;203:150–155. [DOI] [PubMed] [Google Scholar]
- 26.De Silva DA, Proctor L, von Dadelszen P, et al. Determinants of magnesium sulphate use in women hospitalized at <29 weeks with severe or non-severe pre-eclampsia. Plos One 2017;12(12):e0189966. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.De Silva DA, Synnes AR, Dadelszen P, et al. MAGnesium sulphate for fetal neuroprotection to prevent Cerebral Palsy (MAG-CP)-implementation of a national guideline in Canada. Implementation Science. 2018; 13(1):8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Pierrat V, Marchand-Martin L, Arnaud C, et al. Neurodevelopmental outcome at 2 years for preterm children born at 22 to 34 weeks’ gestation in France in 2011: EPIPAGE-2 cohort study. BMJ. 2017;358:j3448. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Adams-Chapman I, Heyne RJ, DeMauro SB, et al. Neurodevelopmental impairment among extremely preterm infants in the neonatal research network. Pediatrics. 2018;141(5). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Chawla S, Natarajan G, Shankaran S, et al. Association of neurodevelopmental outcomes and neonatal morbidities of extremely premature infants with differential exposure to antenatal steroids. JAMA Pediatrics. 2016;170(12):1164–1172. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.