Objective
The purpose of this study was to assess the relationship between neonatal respiratory morbidity and the interval between antenatal corticosteroids (ACS) administration and birth.
Study Design
We performed a retrospective cohort study among women who had received ACS and delivered at <34 weeks of gestation. We categorized these women in 4 groups: ACS-to-delivery interval of 0-7, 8-14, 15-21, and 22-28 days. Multivariable logistic regression analysis assessed the association between the ACS-to-delivery interval and neonatal respiratory morbidity.
Results
We included 254 neonates. Eighty-two neonates (32%) were intubated. In comparison with neonates with an ACS-to-delivery interval of 0-7 days, the risk for intubation was increased in all other groups (odds ratio [OR], 2.3; 95% confidence interval [CI], 1.1–5.4; OR, 5.6; 95% CI, 1.8–18; and OR, 4.8; 95% CI, 0.71–32, not statistically significant, respectively).
Conclusion
The effect of ACS decreases when the ACS-to-delivery interval exceeds 7 days. The first administration of ACS should be considered carefully.
Preterm delivery occurs in 7-10% of all pregnancies and is a major cause of infant death and morbidity. Respiratory distress syndrome (RDS) is often the most acute problem of the premature neonate because of insufficient surfactant production in the immature fetal lung. Antenatal corticosteroids (ACS) treatment, to enhance fetal lung maturity in pregnant women who are at risk for premature birth, was first introduced in 1972 by Liggins and Howie, who reported a significant reduction of RDS in premature neonates whose mothers had received ACS. Since this first randomized controlled trial, many studies have assessed the effectiveness of ACS in women who are at risk for preterm birth. Metaanalysis of these studies showed a significant reduction in the incidence of RDS and the risk of intraventricular hemorrhage (IVH), necrotizing enterocolitis (NEC), and neonatal death.
Although the effectiveness of ACS is beyond doubt, the optimal timing between the administration of the ACS course and premature birth remains uncertain. In a review, Crowley showed the most marked benefit in neonatal outcome when delivery occurred between 24 hours and 7 days after the initiation of corticosteroid treatment. Peaceman et al reported an increased need for short-term respiratory support among premature infants born >7 days after exposure to a single course of ACS; however, no significant differences were found in other measures of neonatal morbidity and death. Data that determine whether the effect of ACS really diminishes from 7 days after initiation of the treatment are scarce. Knowledge of the duration of effectiveness of ACS is of particular interest in the decision whether to repeat ACS administration, because the effects of a repeated dose are still debatable and because harmful effects on neonatal length, weight, and head circumference at birth have been reported. We therefore studied whether the effectiveness of a single complete course of ACS in the prevention of neonatal respiratory and composite neonatal morbidity depends on the time interval between ACS administration and delivery.
Materials and Methods
We performed a retrospective cohort study in 2 Dutch perinatal centers. All neonates who were born alive at a gestational age (GA) between 24 5/7 and 34 weeks in 2006 and whose mothers completed a single course of ACS (2 doses of 12-mg Celestone Chronodose [Merck Sharp & Dohme bv, Brussels, Belgium] intramuscularly with a 24-hour interval) were included. We excluded neonates whose mother did not complete a full course of ACS or whose mother had multiple courses of steroids. Neonates who died within 24 hours after birth because pediatric intervention was not desired because of extreme prematurity were excluded as well.
The maternal medical charts were reviewed for characteristics such as age, obstetric history, GA at initiation of corticosteroids, and delivery. Neonatal charts were reviewed for death and morbidity. Each neonate who was born from a multiple pregnancy was analyzed separately. If women were discharged to other hospitals before delivery, we contacted those hospitals to complete the missing data. A case was lost to follow up and excluded from further analysis if information concerning the neonatal admission was untraceable.
The primary outcome , severe neonatal respiratory morbidity, was defined as the need for intubation of the neonate at the neonatal intensive care unit. Neonates were intubated if they were in need of at least 40% oxygen. Secondary end points were RDS (graded in 4 stadia according to abnormalities on X-thorax: grade 1, slight reticular (slight granular) decrease in transparency of the lung; grade 2, soft decrease in transparency and air-bronchograms, which overlaps the cardiac borders; grade 3, unclear cardiac borders; grade 4, white lung), continuous positive airway pressure (CPAP; given if the neonate needed 21-40% oxygen), and chronic lung disease (CLD; diagnosed if the need for oxygen persisted from 28 days after birth).
For the analysis, patients were categorized in 4 groups: ACS-to-delivery interval from 0-7 days, from 8-14 days, from 15-21 days, and from 22-28 days. The interval was calculated from the day that the first dose of the ACS course was given. First, we constructed spline functions to visualize the association between the need for intubation and the ACS-to-delivery interval in 4 different GA groups: delivery from 26-27 6/7 weeks, from 28–29 6/7 weeks, from 30–31 6/7 weeks, and from 32–33 6/7 weeks. A spline function is the kind of estimate produced by a spline regression in which the slope varies for different ranges of the regressors. The association between the variables on the X- and Y-axes is not fixed by a prespecified formula; instead the line of the spline regression graft approaches the observed relationship in our study. We performed a multivariable logistic regression analysis to assess the relationship between the ACS-to-delivery interval and neonatal intubation. The association with the ACS-to-delivery intervals was further analyzed separately for the following outcomes: CPAP, RDS, and CLD. Finally, we used another multivariable logistic regression analysis to assess whether the interval between ACS and delivery also influenced composite neonatal morbidity. All logistic regression analyses were corrected for GA. Composite neonatal morbidity was defined as the presence of ≥1 of the following grades: CLD, IVH grades 3 and 4 (diagnosed by ultrasound: grade 3, hemorrhage with ventricle dilatation; grade 4, intraparenchymal hemorrhage), NEC, proven sepsis and periventricular leukomalacia grades 2, 3, and 4 (diagnosed by ultrasound: formation of cysts 4 weeks after the cerebral incident). Statistical analysis was performed with SPSS software (version 17.0; SPSS Inc, Chicago, IL).
Results
We identified 522 neonates, whose mothers had received ACS and/or delivered preterm in 2006. Of these neonates, 131 were excluded for being born before the ACS course was completed (of which 1 neonate was 1 of a twin pregnancy) or for receiving multiple courses. We excluded 101 neonates because they were born after a GA of 33 6/7 weeks and 22 neonates because they were born >28 days after the first dose of ACS. Data for 4 other neonates were not retraceable; therefore, they were excluded from the study. Of the remaining 264 neonates, 3 fetuses died before delivery (2 of them were 1 of a twin pregnancy); 5 neonates died during delivery, and 2 neonates died within 24 hours after birth because further pediatric intervention was not desired because of extreme prematurity. Thus, 254 neonates of 220 mothers were included in the analysis ( Figure 1 ).
The mean age of the mothers was 32 years (range, 17–44 years), with a mean body mass index of 24.7 kg/m 2 (range, 15.4–48.0 kg/m 2 ). Ninety-three women (42%) were multiparous, of whom 36 women (39%) had had a previous preterm delivery. In our cohort, 185 women (84%) were pregnant of a singleton. Thirty-three (15%) and 2 (1%) of the pregnancies concerned twins and triplets, respectively; 3 neonates who were part of a twin pregnancy were excluded (as described earlier) because of antepartum death (n = 2) and an incomplete ACS course (n = 1). The multiple pregnancies were spread out among the different groups of GA and ACS-to-delivery intervals. The mean GA at which the first dose of ACS was administrated was 29 0/7; the mean GA at delivery was 30 1/7 weeks ( Table 1 ).
| Maternal characteristic | Measurement |
|---|---|
| Age, y a | 31.61 (7–44) |
| Body mass index, kg/m 2a | 24.7 (15–48) |
| Smoking, n (%) | |
| Yes | 39 (18) |
| No | 100 (45) |
| Unknown | 81 (37) |
| Primiparous women, n (%) | 127 (58) |
| Pregnancies, n (%) | |
| Singleton pregnancies | 185 (84) |
| Twin pregnancies | 33 (15) |
| Triplet pregnancies | 2 (0.9) |
| History of preterm labor, n (%) | 36 (16) |
| Gestational age at 1st antenatal corticosteroid dose, wk a | 29 0/7 (24 3/7–33 2/7) |
| Gestational age at delivery, wk a | 30 1/7 (25 0/7–33 6/7) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
