Contents: Original Research
Preterm birth affects 9.6% of births in the United States and is the most frequent cause of neonatal morbidity and mortality.1 Antenatal corticosteroid administration has proven to be one of the most important interventions in reducing neonatal morbidity, particularly respiratory distress syndrome and intraventricular hemorrhage.2,3 Antenatal corticosteroids have the largest clinical benefit 2–7 days after initiation. However, many women deliver outside the optimal therapeutic window as a result of difficulty in predicting timing of delivery.
Numerous trials have examined the effects of repeated courses of antenatal corticosteroids in patients who continue to remain at risk for preterm delivery.4–7 Although weekly administration is not recommended, a single “rescue” course has been shown to decrease composite neonatal morbidity. Based on the results of one trial of rescue corticosteroids,8 the Maternal-Fetal Medicine Division at Women & Infants Hospital of Rhode Island implemented a guideline in April 2009 for administering a rescue course of antenatal corticosteroids. Subsequently, the critical points from the division guideline were incorporated into the hospital guideline on the use of antenatal corticosteroids and distributed among all practitioners on the internal hospital clinician resources website. Additionally, residents received education of the guidelines by the Maternal-Fetal Medicine Division members and were active in the management of the majority of patients in this study. Because several trials have demonstrated the neonatal benefits of corticosteroids,9 we chose to focus on practice pattern changes. In this study, we evaluated use (both appropriate and inappropriate) of rescue corticosteroids before and after the implementation of this guideline.
MATERIALS AND METHODS
This was a before–after retrospective cohort study performed at Women & Infants Hospital of Rhode Island, a freestanding women's hospital with approximately 8,900 births per annum. To ascertain women eligible for this study, we queried our all-inclusive pharmacy records to identify women who received antenatal corticosteroids for fetal maturation (12 mg betamethasone intramuscular every 24 hours for two doses) between April 2008 and April 2013, that is, 1 year before and 4 years after guideline implementation. Because our years of data abstraction were not calendar years, we defined each April to April year by the beginning year (ie, April 2012 to April 2013 was defined as 2012). Women were included in our study if they received one or both doses of their initial course of antenatal corticosteroids and delivered at Women & Infants Hospital of Rhode Island. The charts of these women were then reviewed and complete data abstraction was performed on women who received or were eligible to receive a rescue course of corticosteroids. Women who were transferred to our facility after completion of their initial course of corticosteroids were not captured in this cohort. Women were eligible to receive rescue dose corticosteroids if they were deemed to be at continued high risk of preterm delivery in the next week and met the following five criteria based on a prior trial8: 1) less than 33 weeks of gestation, 2) intact membranes, 3) no signs of infection, 4) receipt of a first course of corticosteroids at least 14 days prior, and 5) initiation of a first course of corticosteroids at less than 30 weeks of gestation. Women were excluded from the study if they delivered precipitously, thereby not allowing time for the first dose of the rescue course to be administered. For the purposes of analysis, we applied postguideline implementation criteria to the women in the preguideline year, although no formal guideline was in place during that year. We did this to establish baseline use, and the characteristics associated with this use, to allow comparison after the guideline was implemented.
Two of the authors abstracted medical records using standardized data collection forms including basic demographic information, pre-existing maternal medical conditions, indications for corticosteroids, delivery information, and prenatal care provider (maternal–fetal medicine, resident clinic, private, or other). At Women & Infants Hospital of Rhode Island, the maternal–fetal medicine service assumes care for most high-risk patients including transfers from outside institutions and provides prenatal care for approximately 600 patients annually. Approximately 2,500 women per year are delivered on the resident service and the remaining majority is delivered by the private services.
We assumed a baseline rate of rescue corticosteroid administration of 5% before the release of the maternal–fetal medicine guideline and an estimated increase of 8% per year until 2012. Therefore, to detect at least a 40% absolute increase (β=0.02; two-tailed α=0.05) in the administration of rescue corticosteroids from 2009 through 2012, 12 patients per comparison year were required. Data analysis was performed with SAS 9.2 and Stata 13.1. Patient characteristics and other demographic variables were analyzed using χ2 and Fisher exact test for categorical variables and Student t test for continuous variables. Nonparametric data were analyzed using the Wilcoxon rank-sum test. Rates of rescue dose corticosteroid administration over time were compared using the Cochrane-Armitage trend test. Predictors of receipt of rescue corticosteroids were analyzed in multivariable logistic regression analysis to adjust for confounders. This study was formally approved as a retrospective chart review by the institutional review board of Women & Infants Hospital of Rhode Island, protocol number 13-0082.
During the study period of 2008–2012, 2,528 women received a first course of antenatal corticosteroids and delivered at Women & Infants Hospital of Rhode Island. After exclusion of women who delivered rapidly precluding time for administration of rescue corticosteroids (n=5), 142 (5.6%) women were eligible for rescue corticosteroids, of whom 73% (n=103) received one or two doses of rescue corticosteroids (Fig. 1). Eligible women were similar with regard to age, body mass index, race, and gestational age before (n=22) and after the guideline (n=120) (Table 1). The most common indications for rescue corticosteroid administration preguideline and postguideline were preterm labor, growth restriction, and nonreassuring fetal heart tracing. The average time from receipt of the first dose of rescue corticosteroids to delivery was 15.2±17.0 days. Only 25 of 103 (24.3%) of those who received a rescue course delivered within 2–7 days of receipt of the first rescue dose (Table 2). The rate of appropriate administration of one or both doses of a rescue course among eligible women increased from 18.2% (95% confidence interval [CI] 5–40%) in 2008 to 65.4% (95% CI 44–83%) in 2009, 93.5% (95% CI 79–99%) in 2010, 96.1% (95% CI 80–99%) in 2011, and 75.7% (95% CI 59–88%) in 2012 (P<.001) (Fig. 2).
An additional 41 of 2,381 women (1.7%) received rescue corticosteroids for which they were not eligible. The rate of inappropriate administration also rose from 0.4% (95% CI 0.04–1.4%) in 2008 to 1.9% (95% CI 0.9–3.5%) in 2009, 1.8% (95% CI 0.8–3.6%) in 2010, 2.4% (95% CI 1.2–4.3%) in 2011, and 2.2% (95% CI 1.1–4.0%) in 2012 (P=.03). In the postguideline period, there was high level of adherence to four of the five eligibility criteria: less than 33 weeks of gestation (99%), no signs of infection (99%), receipt of the first course of corticosteroids at least 14 days prior (96%), and receipt of the first course of corticosteroids at less than 30 weeks of gestation (94%). In contrast, intact membranes, a requirement for eligibility, were present in only 85% of those who received rescue corticosteroids (Table 3). Thus, among ineligible recipients of rescue corticosteroids, ruptured membranes were the most common condition precluding their eligibility (22/41 [54%]).
Multivariable logistic regression to assess differences in prenatal care provider type for appropriate administration of rescue corticosteroids demonstrated no statistically significant differences (Table 4). Multivariable logistic regression to assess for factors associated with appropriate receipt of corticosteroids demonstrated no association with demographic characteristics, parity, or indication for corticosteroid use. Clinical conditions that were not associated with rescue dose administration included hypertensive disorders (adjusted odds ratio [OR] 0.28, 95% CI 0.03–2.23), multifetal gestation (adjusted OR 0.78, 95% CI 0.23–2.62), and fetal growth restriction (adjusted OR 1.41, 95% CI 0.37–5.48; Table 5).
This report describes the uptake of a guideline for rescue corticosteroid administration for fetal maturation at a large freestanding women's hospital. The uptake was rapid with a change from 18% (4/22) of eligible women receiving a rescue dose preguideline to 65% (17/26), 93% (29/31), 96% (25/26), and 76% (28/37) within the 4 years of guideline implementation, respectively. We observed a small, nonstatistically significant decline in rescue corticosteroid use the last year of the study; however, the 95% CIs for each year after implementation of the guideline all crossed one another. This increase in the use of rescue corticosteroids was seen across all prenatal care providers. However, almost one third of women who received rescue corticosteroids in the postguideline period were not eligible according to the guideline, and only one fourth delivered in the 2- to 7-day window after receipt of the first rescue dose. Why rescue corticosteroids were not administered when they were indicated is not something that we could definitively determine, because this information was not recorded in the medical record. Provision of rescue corticosteroids to women with ruptured membranes was the single most common reason that the guideline was not followed. We suspect that the gestational age range for eligibility was the paramount factor in clinicians' decision to administer a rescue course, and membrane status was simply overlooked. Another reason why that guideline may have failed includes insufficient ongoing health care provider education about the protocol.
The impetus for this investigation was the dearth of data on the use (and misuse) of rescue corticosteroids in actual clinical practice as opposed to their use as part of a research protocol. In fact, in a PubMed search (performed October 23, 2016) using the terms “rescue,” “antenatal,” “corticosteroids” and “protocol,” we found one study that evaluated the accuracy of the timing of corticosteroid administration after a rescue protocol was implemented.10 Makhija et al sought to compare the “accuracy of antenatal corticosteroid administration within the optimal window” (between 48 hours and 7 days before delivery) before and after the implementation of a rescue corticosteroid protocol. They reported no change in the percentage of women who delivered within the optimal window before or after guideline implementation (26.5% compared with 28.5%; P=.41) and with rates very similar to our observed rate of 24.3% among eligible women. As has been recently shown in population representative data from Canada, after controlling for confounding factors, delivery outside the optimal window after an initial course of corticosteroids is associated with worse neonatal outcomes than delivery within the window.11
Our study has several limitations. One limitation is that we are unable to differentiate how much influence our guideline, specifically, had on institutional practice changes. However, practice in our hospital tends to be fairly uniform. The majority of patients subject to these guidelines had residents involved in their care, and, conversely, almost all private physicians in the hospital serve as rotating voluntary attending faculty on the resident service. This bidirectional coverage contributes to uniformity of care, especially because at Women & Infants Hospital of Rhode Island there is a single maternal–fetal medicine service with standard practice guidelines, and all residents rotate on this service. It is also a single-site analysis of a relatively small cohort of women. Another potential limitation was our inclusion of women who were deemed to be at risk for delivery in the next week, a judgment made by the physician who was often incorrect, as emphasized previously. One last limitation is that women who received their entire initial course of corticosteroids at an outside institution were not included in our analysis. The majority of women, however, who are transferred received their first injection at the outside hospital but would have been captured in our study after they received their second injection at Women & Infants Hospital of Rhode Island. In our experience, when corticosteroids for fetal maturation are initiated at one of our referring hospitals, the patient is usually expeditiously transferred to us.
In summary, although overall rescue corticosteroid use increased rapidly after the implementation of an institutional guideline, almost one third of all rescue corticosteroid recipients in our cohort were ineligible according to the guideline, most commonly because they had ruptured membranes, and only one fourth delivered within 2–7 days of the first rescue dose. Optimization of both initial and rescue dose corticosteroid administration will require more accurate prediction of preterm birth timing.
1. Hamilton BE, Martin JA, Osterman MJ, Curtin SC, Matthews TJ. Births: final data for 2014. Natl Vital Stat Rep 2015;64:1–64.
2. Liggins GC, Howie RN. A controlled trial of antepartum glucocorticoid treatment for prevention of the respiratory distress syndrome in premature infants. Pediatrics 1972;50:515–25.
3. National Institutes of Health Consensus Development Panel. Antenatal corticosteroids revisited: repeat courses—National Institutes of health Consensus Development Conference Statement, August 17–18, 2000. Obstet Gynecol 2001;98:144–50.
4. Crowther CA, McKinlay CJ, Middleton P, Harding JE. Repeat doses of prenatal corticosteroids for women at risk of preterm birth for improving neonatal health outcomes. The Cochrane Database of Systematic Reviews 2011, Issue 6. Art. No.: CD003935. DOI: 10.1002/14651858.CD003935.pub4.
5. Crowther CA, Haslam RR, Hiller JE, Doyle LW, Robinson JS; Australasian Collaborative Trial of Repeat Doses of Steroids (ACTORDS) Study Group. Neonatal respiratory distress syndrome after repeat exposure to antenatal corticosteroids: a randomised controlled trial. Lancet 2006;367:1913–9.
6. Guinn DA, Atkinson MW, Sullivan L, Lee M, MacGregor S, Parilla BV, et al. Single vs weekly courses of antenatal corticosteroids for women at risk of preterm delivery: a randomized controlled trial. JAMA 2001;286:1581–7.
7. Wapner RJ, Sorokin Y, Thom EA, Johnson F, Dudley DJ, Spong CY, et al. Single versus weekly courses of antenatal corticosteroids: evaluation of safety and efficacy. Am J Obstet Gynecol 2006;195:633–42.
8. Garite TJ, Kurtzman J, Maurel K, Clark R; Obstetrix Collaborative Research Network. Impact of a ‘rescue course’ of antenatal corticosteroids: a multicenter randomized placebo-controlled trial. Am J Obstet Gynecol 2009;200:248.e1–9.
9. 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, Issue 3. Art No.: CD004454. DOI: 10.1002/14651858.CD004454.pub2.
10. Makhija NK, Tronnes AA, Dunlap BS, Schulkin J, Lannon SM. Antenatal corticosteroid timing: accuracy after the introduction of a rescue course protocol. Am J Obstet Gynecol 2016;214:120.e1–6.
© 2017 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
11. Melamed N, Shah J, Soraisham A, Yoon EW, Lee SK, Shah PS, et al. Association between antenatal corticosteroid administration-to-birth interval and outcomes of preterm Neonates. Obstet Gynecol 2015;125:1377–84.