A repeat course of betamethasone two or more weeks after PPROM did not increase risks of composite neonatal morbidity and neonatal mortality; there was a trend toward decreased respiratory morbidity considering neonates of all gestational ages.
Key findings include increase in composite neonatal morbidity associated with steroid administration in preterm PROM with latency greater than 2 weeks.
This study adds to what is already known by confirming that routine administration of steroids in PPROM, as a risk factor for preterm birth, is a legitimate choice.
This Rpubs paper is the unpublished full text of Obstetrics &
Gynecology 131():p 115S, May 2018
A second course of antenatal corticosteroids (repeat ACS) improves
composite neonatal morbidity in pregnancies at risk of preterm delivery.
PPROM has been an exclusion criterion in some studies of repeat ACS.
This study investigated whether repeat ACS is associated with decreased
neonatal morbidity in PPROM without association with increased
chorioamnionitis or neonatal hypoglycemia.
This was a retrospective cohort study of women from 2005 to 2017
admitted for PPROM between 23w0d and 33w6d and latent at least 14 days
after a two-dose initial course of ACS. Anomalous fetuses were excluded.
Neonates exposed to a repeat course of steroids were compared with those
who did not. The primary outcome was composite neonatal morbidity
including respiratory morbidity, necrotizing enterocolitis,
intraventricular hemorrhage, and death. Secondary outcomes were rates of
clinical and histologic chorioamnionitis, and neonatal
hypoglycemia.
182 women and 216 neonates were included. 77 women received standard
of care and 105 received repeat ACS. Maternal age, demographics, parity,
body mass index, and median gestational age at rupture, and median
gestational age at delivery were comparable between groups. Repeat ACS
was associated with decreased respiratory morbidity (p < 0.05).
Neonatal hypoglycemia and clinical and histologic chorioamnionitis were
similar between groups (p = 0.06, 0.14, 0.82 respectively). When
neonates less than 28 weeks at delivery were analyzed separately,
respiratory morbidity remained the only outcome at the level of
significance (p = 0.05).
Repeat ACS is associated with a decreased incidence of composite
neonatal morbidity and is not associated with increased rates of
chorioamnionitis and neonatal hypoglycemia. Further prospective studies
are needed.
Antenatal steroids, neonatal morbidity, neonatal mortality, preterm
birth, preterm premature rupture of membranes
Antenatal corticosteroids (ACS) administered for risk of preterm delivery are associated with decreased morbidity of prematurity if preterm birth occurs (1-7). Current guidelines recommend two doses of ACS administered 24 hours apart as a single “course” when a woman is at risk of preterm birth, with a repeat course if she remains at risk 1-2 weeks after the initial course (11). These recommendations stem from studies of repetitive dosing that showed growth restriction and neurological effects in animals (8-10), but no significant changes with a single repeat course (4, 7).
The initial studies supporting a repeat course excluded women with preterm premature rupture of membranes (PPROM) (1-5), because of a theoretical risk of chorioamnionitis due to the immunosuppressant effects of glucocorticoids. Although glucocorticoid use in pregnancy has been associated with spontaneous preterm delivery, fetal growth restriction (FGR), and gestational diabetes, no relationship with chorioamnionitis has been identified. The largest review of ACS did not find an increased risk of chorioamnionitis when all recipients are grouped together and it found insufficient evidence to conclude that women with PPROM behave differently as a subgroup (14). Two studies include women with PPROM (6-7) and did not find an increased incidence of chorioamnionitis with a repeat course, although women with PPROM were not analyzed separately.
Chorioamnionitis can also be associated with neonatal hypoglycemia may result from steroid administration. These risks, combined with a concern for chorioamnionitis associated with poorer neonatal outcomes (Suhas et al, Miyazaki et al), can make the decision to administer repeat ACS in PPROM challenging to clinicians.
This study set out to to assess the neonatal benefit of a repeat
course (set of two 12 milligram intramuscular doses) of antenatal
betamethasone given to women with preterm premature rupture of
membranes, and to assess the rates of of chorioamnionitis and neonatal
hypoglycemia associated with a repeat course of antenatal betamethasone
given to women with preterm premature rupture of membranes.
This was an investigator-initiated retrospective cohort study of women undergoing inpatient expectant management of PPROM. At our institution, women with clinical diagnosis of PPROM undergo an initial course of betamethasone (2 injections of 12.5 mg each at 24 hour intervals) as well a seven day course of antibiotics to promote latency. They undergo daily inpatient surveillance and are delivered if they develop preterm labor, clinical chorioamnionitis, placental abruption, abnormal fetal heart rate tracing. If they remain clinically stable, they are monitored on the inpatient service and delivered around 34 gestational weeks. If patients remain undelivered for 14 days following their initial course of antenatal corticosteroids, an additional course can be administered, at the discretion of the attending physician.
Institutional review board approval was obtained from the St. Louis University IRB.
Inclusion criteria were preterm premature rupture of membranes between 23w0d and 33w6d, latency greater than least 14 days, and recurrent or continued threat of preterm delivery 14 or more days after rupture and before 34 completed weeks. Women were excluded if they had active clinical chorioamnionitis at or after their 14 day latency (at the time of possible repeat course), fetuses were excluded if they were anomalous (non-anomalous twins of anomalous fetuses were included), and woman-fetus pairs were excluded if induction or scheduled C-section was required at 34w0d or greater.
The primary outcome was composite neonatal morbidity, including respiratory distress syndrome, intubation, necrotizing enterocolitis, grades III and IV intraventricular hemorrhage, and death. Secondary outcomes included incidence of clinical chorioamnionitis (CCA) defined by two of three parameters (maternal fever greater than 100.4 F twice or 101 F once, maternal heart rate greater than 100 beats per minute, fetal heart rate greater than 160 beats per minute, fundal tenderness, or white cell count greater than 30), histologic chorioamnionitis (HCA) defined broadly as inflammatory infiltrate within the fetal membranes, and neonatal hypoglycemia (defined as a blood glucose of under 40 mg/dL in term infants and under 30 mg/dL in preterm infants). In addition, demographic data, body mass index (BMI), complete medical and obstetrical history, tobacco status, comorbid short cervix or preterm labor, tocolytic use, ultrasound data, and delivery data were also collected. Neonatal sex, weight, Apgars, cord gases, disposition from delivery room, length of stay, and placental pathology were also abstracted from neonatal charts.
A power analysis using a power of 0.8, alpha of 0.05, and incidences of the primary outcome in a cross-sectional review and a study with a similar design, we expected a 64% incidence of the primary outcome in “No Repeat” group and 44% in “Repeat ACS” group (Manuck, Garite). This predicted that 100 neonates would be needed in each group.
Baseline differences in maternal characteristics and delivery
characteristics were determined using chi-square test for categorical
variables, student t tests for continuous variables with normal
distribution, and Mann-Whitney-Wilcoxon test for continuous variables
not determined as normal by the Shapiro-Wilke test. For uniformity in
reporting, p values were calculated using Mann-Whitney-Wilcoxon test for
continuous variables, as more continuous variables were not normal. The
primary outcome of composite neonatal morbidity/mortality was initially
studied using a chi-square test, and then examined with a multivariate
linear regression to assess the contribution of gestational age at
delivery to the observed effect.
Between January 2005 and December 2017, 2196 pregnancies were identified as complicated by PPROM in the electronic delivery record. Following review, 22 were found not to be complicated by PPROM but represented documentation errors. 376 were excluded as they were greater than or equal to 34 completed weeks at diagnosis of PPROM; 195 were excluded for gestational less than 24 completed weeks at diagnosis; 1095 were excluded for latency less than 14 days, 48 were excluded for never having received an initial dose of ACS, 1 was excluded for use of dexamethasone, 10 were excluded for having received repeat ACS prior to PPROM for another indication, 8 fetuses were anomalous, and 225 were excluded for lack of complete documentation (see Figure 1). The remaining 216 neonates were included for analysis. This included 34 twin pairs, 28 of which were complete sets and 6 of which were not (all six were excluded for anomalies). 119 neonates received repeat ACS and 97 neonates did not.
Maternal age, demographics, gravidity and parity, BMI, tobacco use, median gestational age at rupture, gestational age at delivery, and mode of delivery were comparable between groups (see Table 1). Women were delivered for similar reasons between groups. Latency was significantly different between groups, related to study design (Table 2).
The incidence of composite neonatal morbidity was 53% in the “No Repeat” group and 40% in the “Repeat ACS” group. Before performing multivariate logistic regression, there was no significant difference in the composite neonatal morbidity, but there was a significant difference in respiratory morbidity, largely driven by RDS. However, upon controlling for gestational age and race using logistic regression, there was only a trend toward significance in respiratory morbidity, and the only factor that remained significantly predictive of morbidity was gestational age (see Table 3).
Secondary outcomes including CCA and HCA were not significantly
different between groups and CCA as a reason for delivery was not
different between groups (p = 0.1, 0.8, and 0.6 respectively, see Tables
1 and 2), but a trend toward significance was noted in neonatal
hypoglycemia (p = 0.6, see Table 2).
| Characteristics | Repeat (n = 77) | No Repeat (n = 125) | p-value |
|---|---|---|---|
| Age (years) | 26.4 ± 6.0 | 27.3 ± 6.5 | 0.4 |
| Maternal BMI | 30.6 ± 7.5 | 32.6 ± 9.3 | 0.2 |
| Parity | 0.3 | ||
| 0 | 33 (42.9) | 35 (33.3) | |
| 1 | 18 (23.4) | 23 (21.9) | |
| ≥2 | 26 (33.8) | 47 (44.8) | |
| Race (%) | 0.9 | ||
| White non-Hispanic | 42 (55.3) | 56 (54.4) | |
| Black non-Hispanic | 32 (42.1) | 43 (41.8) | |
| Other (including Hispanic) | 2 (2.6) | 4 (3.9) | |
| Prior preterm birth | 20 (26.0) | 25 (23.8) | 0.9 |
| Tobacco use | 19 (24.7) | 30 (28.6) | 0.7 |
| Twin gestation | 22 (28.6) | 18 (17.1) | 0. |
| Reason for delivery | |||
| Labor | 39 (40.2) | 49 (41.2) | 0.7 |
| Planned induction | 49 (50.5) | 59 (49.6) | 0.4 |
| Chorioamnionitis | 4 (4.1) | 3 (2.5) | 0.6 |
| Abnormal fetal testing | 4 (4.1) | 8 (6.7) | 0.7 |
| Hypertensive disorder | 1 (1.0) | 0 | NC |
| Mode of delivery | 0.3 | ||
| Vaginal delivery | 49 (51.0) | 70 (58.8) | |
| Cesarean delivery | 47 (49.0) | 49 (41.2) | |
| Clinical chorioamnionitis | 12 (15.6) | 27 (25.7) | 0.1 |
| Histopathologic chorioamnionitis | 33 (42.9) | 48 (45.7) | 0.8 |
Age and BMI are shown as mean ± SD; remaining variables are shown as
n (%).
| Characteristics | Repeat (n = 97) | No Repeat (n = 119) | p-value |
|---|---|---|---|
| Gestational age at delivery (wk) | 32 ± 2.0 | 31 ± 3.0 | 0.2 |
| Latency | <0.0001 | ||
| 15-21 days | |||
| 22-28 days | 12 (12) | 57 (48) | |
| > 28 days | 20 (21) | 32 (27) | |
| Birth weight (grams) | 1675.6 ± 417 | 1675 ± 605 | 0.7 |
| 1-minute Apgar < 5 | 23 (24) | 38 (33) | 0.6 |
| 5-minute Apgar < 7 | 13 (13) | 25 (22) | 0.5 |
| Composite morbidity | 39 (40) | 63 (53) | 0.08 |
| Respiratory composite | 36 (37) | 61 (51) | 0.05 |
| Intraventricular hemorrhage | 6 (6) | 8 (7) | 1.0 |
| Necrotizing enterocolitis | 0 | 5 (4) | NC |
| Hypoglycemia | 14 (14) | 6 7 (6) | 0.06 |
| Neonatal death | 0 | 1 (1) | NC |
Age and BMI are shown as mean ± SD; remaining variables are shown as
n (%).
| Variable | Composite Morbidity | Respiratory Morbidity | p-value |
|---|---|---|---|
| Rescue ACS | 0.92 (0.8-1.0) | 0.91 (0.79-1.0) | |
| Gestational age at delivery (wk) | 0.91 (0.88-0.93) | 0.91 (0.89-0.94) | <0.001 |
| Black race | 0.93 (0.8-1.1) | 0.95 (0.83-1.1) | |
| Other non-White Race | 0.77 (0.45-1.1) | 0.81 (0.48-1.1) |
In summary, in this retrospective cohort of women with PPROM who received one course of ACS and remained undelivered for at least 2 weeks, a repeat ACS course was associated with similar neonatal morbidity, decreased respiratory morbidity, similar CCA and HCA rates, and a trend toward increased neonatal hypoglycemia.
The effects on neonatal morbidity obtained in pairwise comparisons is similar to the literature demonstrating benefit of ACS in neonates in general, and the effect on neonatal blood glucose is also similar. Our findings are also consistent with the two studies that found a neonatal benefit in women who received repeat ACS, although this study qualifies that the benefit associated with ACS pales in comparison with the effect of gestational age on the morbidities examined. Further, incidences of CCA and HCA are similar in our study as in the extant literature.
The study’s strengths include its limited clinical question and the use of logistic regression to model several determinants of preterm birth.
The study was limited by its retrospective design and by the failure to reach the necessary N of 100 in the “repeat ACS” group. Moreover, incidences of the primary outcome differed from the incidences used in calculation of the required sample size. Furthermore, differences in latency likely represent a selection bias, which may further confound the result obtained.
We conclude that a repeat course of ACS is not associated with
significant harm. There was a trend toward decreased respiratory
morbidity in the repeat ACS group. While gestational age at PPROM and at
delivery were similar between groups, logistic regression suggests that
gestational age has a different effect size, consistent with gestational
age having a greater effect on neonatal morbidity and mortality than ACS
administration in any group. Future research should be prospective and
ideally randomized to exclude selection bias.
Stutchfield P, Whitaker R, Russell I. Antenatal betamethasone and incidence of neonatal respiratory distress after elective caesarean section: pragmatic randomised trial. BMJ 2005 Sep 24;331(7518):662.
Wapner RJ et al. Single versus weekly courses of antenatal corticosteroids: evaluation of safety and efficacy. Am J Obstet Gynecol 2006;195:633-42.
Wapner RJ et al. Long-term outcomes after repeat doses of antenatal corticosteroids. N Engl J Med 357;12: 1190-1198.
Garite T, Kurtzman J, Maurel K, Clark, R. Impact of a ‘rescue course’ of antenatal corticosteroids: a multicenter randomized placebo-controlled trial. Am J Obstet Gynecol 2009;200:248.e1-e9.
Crowther CA, McKinlay CJD, Middleton P, Harding JE. Crowther CA, McKinlay CJD, Middleton P, Harding JE. Cochrane Database of Syst Rev 2015 Jul 5;(7):CD003935.
Crowther CA, Haslam RR, Hiller JE, Doyle LW, Robinson JS. Neonatal respiratory distress syndrome after repeat exposure to antenatal corticosteroids: a randomised controlled trial. Lancet 2006;367:1913-19.
Crowther CA, Doyle LW, Haslam RR. Hiller JE, Harding JE, Robinson JS. Outcomes at 2 years of age after repeat doses of antenatal corticosteroids. N Engl J Med 357;12:1179-1189.
Dunlop SA, Archer MA, Quinlivan JA, Beazley LD, Newnham JP. Repeated prenatal corticosteroids delay myelination in the ovine central nervous system. J Matern Fetal Med 1997 Nov-Dec;6(6):309-13.
Jobe AH, Newnham J, Willet K, Sly P, Ikegami M. Fetal versus maternal and gestational age effects of repetitive antenatal glucocorticoids. Pediatrics 1998;102:1116-1125.
Jobe AH, Wada N, Berry LM, Ikegami M, Ervin MG. Single and repetitive maternal glucocorticoid exposures reduce fetal growth in sheep. Am J Obstet Gynecol 1998;178:880-5.
Committee on Practice Bulletins-Obstetrics. ACOG Practice
Bulletin No. 188: Prelabor Rupture of Membranes. Obstet Gynecol. 2018
Jan;131(1):e1-e14. doi: 10.1097/AOG.0000000000002455. PMID: 29266075.
This manuscript is not peer-reviewed. I did not list co-authors given the informal nature of this publication, but Dr. Laura K. Vricella was the original PI, and our statistician was Collin Miller. Then-students who assisted are Drs. Shivani Parikh and Ashley Peterson, all then affiliated with St. Louis University. Be merciful, I drafted this as a resident and didn’t edit at all before posting.