Objective
We sought to assess whether betamethasone (BETA) <34 weeks reduces adverse outcomes in late preterm infants.
Study Design
We performed a retrospective cohort study of patients with spontaneous birth 34-36 6/7 weeks. We determined whether patients were exposed to preterm labor (PTL) <34 weeks and BETA and calculated the incidence of adverse respiratory and composite outcomes and neonatal intensive care unit admission. We used χ 2 analyses to determine associations between PTL+BETA and adverse outcomes, and Poisson regression to model cumulative incidence and control for confounders.
Results
We enrolled 700 mother-infant pairs. The 36-week PTL+BETA infants were at increased risk of respiratory outcome (incident risk ratio [IRR], 2.73; 95% confidence interval [CI], 1.37–5.45), neonatal intensive care unit admission (IRR, 2.01; 95% CI, 1.14–3.56), and composite outcome (IRR, 1.70; 95% CI, 1.08–2.68) compared to those without PTL+BETA. Chorioamnionitis was independently associated with all adverse outcomes.
Conclusion
We hypothesize that early PTL is a surrogate for intrauterine inflammation and is responsible for the observed adverse outcomes in those with PTL+BETA.
Of all infants born preterm (<37 weeks), approximately 71% are delivered during the late preterm period (34-36 6/7 weeks). Recently published data describing outcomes for this cohort consistently report that these infants experience significantly higher rates of adverse pulmonary outcomes, overall morbidity, hospital readmission, and death than their term counterparts. While the percentage of infants delivered during this time period has risen by 16% over the last 10 years, we have yet to develop a strategy to successfully improve outcomes in this cohort.
In contrast to late preterm infants, there is less ambiguity regarding management strategies to optimize outcomes for infants at risk of delivering <34 weeks. For these patients, the American College of Obstetricians and Gynecologists recommends the administration of either 2, 12-mg doses of betamethasone (BETA) given 24 hours apart or 4, 6-mg doses of dexamethasone given 12 hours apart to reduce the incidence of respiratory distress syndrome, intraventricular hemorrhage (IVH), necrotizing enterocolitis (NEC), and neonatal death.
Infants delivered <34 weeks experience improved outcomes when exposed to BETA in utero compared to their gestational age-matched peers who are not exposed to BETA. What remains unknown is whether late preterm infants exposed to BETA at <34 weeks have improved outcomes compared to late preterm infants not exposed to BETA. Therefore, our objective was to determine whether or not late preterm infants who had been exposed to antenatal corticosteroids <34 weeks of gestation experience more favorable respiratory and composite adverse outcomes than infants not exposed to corticosteroids in utero.
Materials and Methods
We performed an institutional review board–approved retrospective cohort study of women who delivered between 34-36 6/7 weeks’ gestational age due to spontaneous preterm birth (PTB) at a large, urban, tertiary care center from 2003 through 2008.
Patient identification was performed through a search by International Classification of Diseases, Ninth Revision codes for PTB. Medical records were reviewed to assess eligibility for the study. Women with a singleton pregnancy who delivered between 34-36 6/7 weeks secondary to either spontaneous preterm labor (PTL), defined as documented cervical change on sterile vaginal examination in the setting of regular uterine contractions, or preterm premature rupture of membranes (PPROM), defined as rupture of membranes prior to the onset of labor, were included. Patients with multiple gestation, major fetal congenital anomaly, intrauterine fetal demise, preeclampsia, insufficient neonatal information, and/or any indication for iatrogenic delivery were excluded.
Maternal demographic information; obstetric, medical, surgical, gynecologic, and social histories; antenatal diagnoses and medication; and gestational age at delivery were recorded through retrospective chart review. Whether a patient was admitted <34 weeks for PTL was also noted, with documentation of gestational age at that admission and whether or not corticosteroids and/or tocolysis were received.
Estimated gestational age at the time of delivery was established as follows. Patients with a sure last menstrual period (LMP) were dated by their LMP if this gestational age was confirmed by ultrasound dating within 7 days of first-trimester scan, 10 days of an early second-trimester scan, 14 days of a late second-trimester ultrasound, and 21 days of third-trimester scan. Patients whose LMP dating was inconsistent with their ultrasound dating or who were unsure of their LMP were dated by their earliest ultrasound. Obstetric dating was always used to assign gestational age.
Neonatal information recorded included mode of delivery, birth weight, Apgar scores, infant sex, length of neonatal intensive care unit (NICU) stay (days), and whether or not the infant had respiratory insufficiency (defined as requirement for continuous positive airway pressure [CPAP] and/or ventilator), hypoglycemia (blood glucose <50), apnea/bradycardia, presumed or culture-positive sepsis, IVH grades I-IV, seizures, NEC diagnosed clinically and/or radiographically, anemia requiring transfusion, hyperbilirubinemia requiring phototherapy ≥4 days, reflux requiring medication, and/or temperature instability (failure to wean from isolette or need to return to isolette). Whether or not the infant died while hospitalized, as well as day of death, was also noted. All diagnoses were determined by reviewing the documentation of the attending neonatologist or pediatrician who assessed each infant. We chose not to record subjective diagnoses listed in the chart such as “respiratory insufficiency” and “respiratory distress” given the heterogeneous and practitioner-dependent use of each of these terms. Neonatal information was limited to those events that occurred between delivery and discharge home from our level-3 NICU.
At our institution, there are standard practices for the management of PTL and PPROM between 24-33 6/7 weeks, which were unchanged over the study period. Patients who present with PTL receive 48 hours of magnesium sulfate tocolysis and a single course (2 doses) of BETA. If labor is successfully interrupted, the patient is ultimately discharged with close outpatient follow-up. Patients who present with PPROM receive 7 days of antibiotics (ampicillin and erythromycin) for latency as well as a single course (2 doses) of BETA. Patients with PPROM are hospitalized until delivery; if spontaneous delivery does not occur and/or the patient is not induced for chorioamnionitis, patients with PPROM are delivered at 34 weeks’ gestational age. Patients who present with either PTL or PPROM ≥34 weeks do not receive BETA, tocolysis, or antibiotics for latency. Patients with PPROM ≥34 weeks are delivered.
Since the primary aim of our study was to determine the effect of the administration of BETA on late preterm respiratory outcomes, we created a composite variable, RESP, to describe adverse neonatal respiratory outcomes. RESP included requirement of CPAP and/or ventilator. To further describe the association between neonatal outcomes and BETA exposure we created 2 additional composite variables, COMP and COMP+RESP, combining biologically related individual outcomes. COMP included infant death while hospitalized, admission to the NICU ≥8 days, and/or diagnosis of hypoglycemia, apnea/bradycardia, sepsis, IVH, seizures, NEC, anemia, hyperbilirubinemia, reflux, or temperature instability as defined above. COMP+RESP represented the union these 2 composite variables.
A priori sample size calculations were performed to determine how many mother-infant pairs would be needed to detect a 50% change in the prevalence of adverse respiratory outcomes between those infants exposed to BETA in utero compared with those not exposed. We estimated that the prevalence of RESP among infants born in the late preterm period at our institution is 18.8% based upon our previously published work. Assuming 80% power, a type I error of 5%, and an enrollment ratio of 1:1 between those with and without BETA exposure, we determined we would require 700 mother-infant pairs.
Significant associations between both individual and composite outcomes with exposure to BETA were determined using χ 2 and Fisher’s exact tests. Poisson regression was used to obtain estimates of the cumulative incidence, ie, risk, and incident risk ratio (IRR) of both individual and composite adverse neonatal outcomes and to control for confounders. Software (STATA, Version 10; StataCorp, College Station, TX) was used for data analysis.
Results
We reviewed 1984 mother-infant charts. We excluded 40.3% (n = 799) for gestational age at delivery <34 weeks, 8.5% (n = 168) for induction of labor for maternal or fetal indication, 8.0% (n = 159) for multiple gestation, 4.3% (n = 85) for major fetal malformation, 3.2% (n = 63) for gestational age ≥37 weeks, and 0.5% (n = 10) for insufficient neonatal information at the time of delivery. In accordance with our a priori sample size calculations, we analyzed 700 mother-infant pairs that met our inclusion and exclusion criteria.
Demographics of the patients included in the study are described in Table 1 . In our study population, all patients with PTL or PPROM <34 weeks received BETA. There were 148 patients who experienced PTL/PPROM <34 weeks and were administered BETA (PTL+BETA) representing 21.1% of the cohort studied. In all, 552 patients representing 78.9% of the cohort did not experience PTL/PPROM <34 weeks and were not administered BETA.
| Demographic variable | BETA (n = 148) n (%) | No PTL+BETA (n = 552) n (%) | P value a |
|---|---|---|---|
| Gestational age at delivery, wk | < .001 | ||
| 34 | 65 (43.9) | 117 (21.2) | |
| 35 | 40 (27.0) | 181 (32.8) | |
| 36 | 43 (29.1) | 254 (46.0) | |
| Maternal age, y | .04 | ||
| ≥35 | 8 (5.4) | 65 (11.8) | |
| <35 | 140 (94.6) | 487 (88.2) | |
| Mode of delivery | .69 | ||
| SVD | 121 (81.8) | 465 (84.2) | |
| Operative vaginal delivery | 6 (4.1) | 16 (2.9) | |
| Cesarean delivery | 21 (14.2) | 71 (12.9) | |
| Race | .41 | ||
| African American | 125 (84.5) | 450 (81.5) | |
| Other | 23 (15.5) | 102 (18.5) | |
| Prenatal care | .07 | ||
| No prenatal care | 10 (6.7) | 75 (13.6) | |
| Medicaid clinic | 104 (70.3) | 350 (63.4) | |
| Private insurance office | 34 (23.0) | 127 (23.0) | |
| Clinical chorioamnionitis | .63 | ||
| Yes | 9 (6.1) | 28 (5.1) | |
| No | 139 (93.9) | 524 (94.9) | |
| Histologic chorioamnionitis | .20 | ||
| Yes | 16 (10.8) | 56 (10.1) | |
| No | 39 (26.4) | 110 (19.9) | |
| Unknown | 93 (62.8) | 386 (69.9) |
a P values calculated by χ 2 or Fisher’s exact test where appropriate.
In our cohort, 2.7% (n = 19) received antenatal 17-hydroxyprogesterone and 17.4% (n = 122) received magnesium sulfate tocolysis. There was no association between administration of either 17-hydroxyprogesterone or magnesium sulfate tocolysis and neonatal outcome.
Although overall PTL+BETA infants were not at increased risk for RESP ( Table 2 ), gestational age at delivery did significantly modify the effect of PTL+BETA on RESP ( P = .04). Specifically, 36-week infants exposed to PTL+BETA had a >2-fold increase in the risk of RESP (IRR, 2.31; 95% confidence interval [CI], 1.15–4.65) compared to unexposed 36-week infants.
| Adverse outcome | PTL + BETA (n = 148) | No PTL+BETA (n = 552) | P value a |
|---|---|---|---|
| Respiratory composite | 28 (18.9%) | 79 (14.3%) | .17 |
| Composite (w/o respiratory) | 74 (50.0%) | 171 (31.0%) | < .001 |
| Composite + respiratory | 79 (53.4%) | 189 (34.2%) | < .001 |
| Any NICU stay | 74 (50.0%) | 168 (30.4%) | < .001 |
| Prolonged NICU stay | 43 (29.1%) | 89 (16.1%) | < .001 |
| Sepsis | 62 (41.9%) | 231 (41.8%) | .99 |
| IVH or seizure | 2 (1.4%) | 5 (0.9%) | .62 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
