- Email: [email protected]
Antenatal corticosteroid timing: accuracy after the introduction of a rescue course protocol
Accepted Manuscript Antenatal Corticosteroid Timing: Accuracy after the Introduction of a Rescue Course Protocol Neeta K. Makhija, MD, Ashlie A. Tronnes, MD, Benjamin S. Dunlap, Jay Schulkin, PhD, Sophia M. Lannon, MD, MPH PII:
S0002-9378(15)00869-8
DOI:
10.1016/j.ajog.2015.08.018
Reference:
YMOB 10582
To appear in:
American Journal of Obstetrics and Gynecology
Received Date: 20 May 2015 Revised Date:
10 July 2015
Accepted Date: 10 August 2015
Please cite this article as: Makhija NK, Tronnes AA, Dunlap BS, Schulkin J, Lannon SM, Antenatal Corticosteroid Timing: Accuracy after the Introduction of a Rescue Course Protocol, American Journal of Obstetrics and Gynecology (2015), doi: 10.1016/j.ajog.2015.08.018. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Antenatal Corticosteroid Timing: Accuracy after the Introduction of a Rescue Course Protocol
RI PT
Neeta K MAKHIJA, MD1, Ashlie A TRONNES MD 1,, Benjamin S DUNLAP2 , Jay SCHULKIN PhD3, Sophia M LANNON, MD, MPH1 1 University
of Washington School of Medicine, Department of Obstetrics and Gynecology of Washington School of Medicine, Department of Medicine 3 American College of Obstetricians and Gynecologists
The authors of this study report no conflict of interest.
SC
2 University
TE D
Corresponding Author: Neeta K Makhija, MD [email protected] (484) 629 - 5879
M AN U
An abstract describing this work was presented at the 35th Annual Pregnancy Meeting of the Society for Maternal-Fetal Medicine, San Diego, California. February 6, 2015.
AC C
EP
University of Washington Ob/GYN Box 356460 1959 NE Pacific Street Seattle, WA, 98195-6460
Word Count: Abstract: 472 Total manuscript: 2,295
Please include Table 2 in the print version of the manuscript
ACCEPTED MANUSCRIPT 2 Condensation: Introduction of rescue antenatal corticosteroids did not impact timing of the first course, nor did it increase total patients delivering within any optimal window. <24> Short version of title: Rescue Course Antenatal Corticosteroid Timing
RI PT
Reference words: antenatal corticosteroids, antenatal corticosteroids timing, preterm birth prediction, rescue antenatal corticosteroids
Abstract
SC
Background: Antenatal corticosteroid (ACS) administration is a critical fetal intervention and utilization of a rescue protocol is now standard practice. Rescue ACS may improve
M AN U
overall accuracy of ACS administration timing, but this observation and its effect on the initial course is unknown.
Objective: We sought to compare the accuracy of ACS administration before and after the
TE D
implementation of a rescue ACS protocol.
Study Design: We performed a retrospective cohort study of patients who received minimum one dose of ACS between 2006-2012 at the University of Washington Medical
EP
Center (UWMC) using the UWMC Pharmacy Database. For inclusion, subjects were required to be admitted, receive initial ACS course between 24 to 34 weeks gestational age
AC C
and deliver at UWMC. We designated two groups based on when rescue ACS became standard practice at UWMC: before rescue ACS (2006-2008) and after rescue ACS (20092012). Primary outcome was delivery within any optimal ACS window defined as 48 hours to 7 days after the first dose or third dose. We also compared delivery within the optimal window of the initial and rescue ACS courses independently and assessed ACS timing by indication for delivery. Chi squared and independent sample t-test were used to compare
ACCEPTED MANUSCRIPT 3 results. Results: Between 2006 and 2012, 1356 women met inclusion criteria, 601 before and 755
RI PT
after rescue ACS. The study groups demonstrated similar demographics with the exception of more Caucasians in the group after rescue ACS (47% vs. 60%, p<0.01) and also delivered at comparable gestational ages (32.7 wks. vs. 32.6 wks, p=0.59). Availability of a second
SC
course did not increase total subjects delivering within any optimal window (26.5% vs. 28.5%, p=0.41). Frequency of delivery within initial course optimal window did not change
M AN U
after introduction of rescue course protocol (26.1% vs. 26.4%, p=0.92). Similarly, of the 73 subjects who received rescue ACS, 24.7% delivered in the optimal window of the second course. Delivery within optimal window varied by indication for ACS, with highest accuracy among maternal indications (41.2% in any optimal window), followed by PPROM (32.1%).
TE D
Lowest administration accuracy was among women with cervical shortening and advanced cervical dilation with only 2.8% and 6.3% delivering within the optimal window, respectively. Furthermore, for women with cervical shortening, the mean gestational age of
EP
delivery was 35.1 weeks and the median interval from ACS administration to delivery was 55 days (IQR 34-72 days).
AC C
Conclusions: The opportunity for a second course of ACS did not improve the number of women delivering within any optimal ACS window. Administration timing was similar for the initial course and the rescue course, with approximately one quarter of women delivering within the optimal ACS window. These findings likely reflect the few circumstances where rescue ACS is useful and the poor predictability of preterm birth.
ACCEPTED MANUSCRIPT 4 Future focus should be aimed at tools to predict the timing of preterm birth in order to
AC C
EP
TE D
M AN U
SC
RI PT
optimize ACS administration.
ACCEPTED MANUSCRIPT 5 Introduction Antenatal corticosteroids (ACS) are an important intervention used to improve
RI PT
neonatal outcomes after preterm birth. Since the landmark study by Liggins and Howie in 1972,1 several studies confirm that treatment with ACS is associated with an overall
reduction in neonatal death, respiratory distress syndrome and intensive care admissions.2
SC
ACS offer the greatest benefit to the fetus if the infant is delivered 24-48 hours after initial dose and within seven days of the administration.2, 3 Benefits decrease after this optimal
M AN U
window.4-7 However, prediction of preterm birth is challenging, especially within a narrow timeframe. While one strategy to improve timing of administration is to provide repeated courses of ACS, studies demonstrate that multiple courses carry fetal risks.8-11 More recent research demonstrates that a single rescue course of ACS may also improve neonatal
TE D
outcomes when delivered within the optimal window.12, 13 In 2011 and 2012, the American College of Obstetricians and Gynecologists (ACOG) endorsed administering a single course of rescue steroids for women who remain at risk for preterm delivery, commonly known as
EP
the rescue protocol.14
Studies demonstrate that the majority of women receiving initial ACS deliver prior
AC C
to 34 weeks gestation, but accuracy of administration within the optimal window is low and varies by indication for delivery. 15-20 The accuracy of rescue ACS administration remains unknown. Recently, studies demonstrating combined increased utilization and inaccurate timing of ACS have raised concern for long-term harmful effects on infants born outside of the optimal window.17, 21 Furthermore, few data report on the long-term harm of a single rescue course of ACS. For these reasons ensuring that the availability of rescue course ACS increases the number of women receiving ACS within any optimal window
ACCEPTED MANUSCRIPT 6 remains a priority. Our objective is to compare the accuracy of ACS administration before and after the availability of a rescue course protocol and to determine whether rescue ACS
RI PT
increases the proportion of women delivering within any optimal ACS window.
Materials and Methods
We conducted a retrospective cohort study of women receiving antenatal
SC
corticosteroids, comparing the proportion of patients who delivered in the optimal window of the initial ACS course before and after the availability of rescue ACS. We identified
M AN U
subjects using the inpatient hospital pharmacy database, capturing all women admitted to the labor and delivery unit. Women were included if they were ordered at least one dose of ACS between January 2006 and December 2012 at the University of Washington Medical Center, a level three-referral hospital. Data was then linked to the University of Washington
TE D
Perinatal Database for additional demographic data. A single primary researcher then performed a review of each chart to confirm and obtain additional data. The Human Subjects Division at the University of Washington approved this study (IRB # 44589, Date
EP
of Approval 2/28/2015).
AC C
The cohort included women admitted to labor and delivery and receiving at least one dose of betamethasone 12mg IM, which is the ACS used almost exclusively at our and referring institutions. Patients were excluded if they received betamethasone at less than 24 0/7 weeks or greater than 33 6/7 weeks gestational age, delivered at an outside institution or we could not confirm administration of ACS by chart review. Some women received their first dose of ACS prior to transfer to our institution, and they were included if ACS administration was confirmed.
ACCEPTED MANUSCRIPT 7 Two groups were identified according to the year of initial ACS course administration. In 2009, the Maternal Fetal Medicine group at the University of Washington Medical Center developed a consensus policy supporting the use of a single rescue course
RI PT
of ACS. The first group received the initial ACS course between 2006 and 2008, which was before implementation of our rescue steroid protocol, and the second group received initial ACS course between 2009 and 2012, which was after wide spread use of the rescue course
SC
protocol.
We compared maternal, pregnancy and delivery characteristics between the groups
M AN U
before and after the availability of rescue steroids. The potential indications for administration of ACS were categorized as: preterm labor, preterm premature rupture of membranes, antenatal cervical shortening, advanced cervical dilation, vaginal bleeding, maternal factors and fetal factors. Preterm labor was defined as regular, painful
TE D
contractions with cervical change. Antenatal cervical shortening was used to describe women who were asymptomatic though found to have a cervical length < 2 cm on transvaginal ultrasound. Many of these women had additional risk factors for preterm
EP
birth. Advanced cervical dilation was defined as dilation > 2 cm and patients were only
AC C
given this diagnosis if they did not meet criteria for preterm labor. If patients had more than one indication, the investigator assigned a primary indication. We compared the proportion of patients delivering within the optimal window before and after the availability of a rescue course protocol. The primary outcome was delivery within optimal window of either the initial ACS course or the rescue ACS course. The optimal window for ACS administration is defined as delivery between 48 hours and 7 days after the first dose for the initial course and between
ACCEPTED MANUSCRIPT 8 48 hours and 7 days after the third dose for the rescue course. Although data suggest that steroid benefit can occur after 24 hours from the first dose, our institution routinely considers maximum benefit at 48 hours and uses this time point in clinical decision-making
RI PT
and delivery planning. The secondary outcomes included delivery within optimal window of the rescue course and delivery within the optimal window of the initial ACS course.
Additionally, we assessed delivery timing and delivery characteristics by indication for
SC
administration.
Sample size was determined by the availability of subjects receiving ACS within the
M AN U
study time period. With our sample size of 1356, the assumption that 30% of women received ACS within the optimal window, alpha of 0.05 and power of 80%, the minimal effect size detectable is 7%. Statistical analysis was completed using STATA software version 11.1. Pearson chi-squared tests were used to compare categorical variables and
TE D
independent sample t-tests were used to compare continuous variables. For all analyses a
Results
EP
two-sided significance level of <0.05 was considered statistically significant.
AC C
We identified 1809 pregnant women for whom a pharmacy order was placed for betamethasone within the study period. After exclusions for not receiving ACS, delivering at an outside hospital, or gestational age, the final study population was 1356 subjects (Figure 1). Of the total cohort, 601 received ACS in 2006-2008 before availability of rescue protocol and 755 received ACS in 2009-2012 after availability of rescue protocol. Of note, similar proportions of patients were excluded from each group secondary to delivery at an outside institution (20.2% before rescue ACS vs. 18% after rescue ACS, p= 0.24). However,
ACCEPTED MANUSCRIPT 9 women included in the study were on average 1.7 years older than those excluded for outside hospital delivery, p<0.01. Additional demographic data on the excluded group was not accessible.
RI PT
Table 1 reveals the maternal characteristics of our population. Study groups were similar with the exception of racial distribution. There was a high representation of
multiple gestations, transfer to higher level of care, hypertension, diabetes, and those with
SC
other medical comorbidities in our study group. As anticipated, there was a high
proportion of preterm deliveries. The study groups delivered at similar gestational ages
M AN U
with approximately 62% of the population delivering less than 34 weeks gestation. The primary outcome of delivery within the optimal window of either the initial or rescue ACS course, did not change after the availability of rescue ACS (26.5% before rescue ACS vs. 28.5% after rescue ACS, p=0.41, (Figure 2C). Considering interval to delivery as a
TE D
continuous variable did not impact our results (23.1 days before rescue ACS, vs. 23.7 days after rescue ACS, p=0.66). Furthermore, when examining accuracy of administration over the study years, ACS utilization and accuracy appeared stable over time (Figure 3). Finally,
EP
when stratified by gestational age, there was no statistical difference in the proportion of
AC C
patients who delivered in any optimal window before and after the availability of rescue steroids (24 - 28 wks: 43.5% before rescue ACS vs. 51.6% after rescue ACS, p=0.5; 28 - 32 wks: 35.7% vs. 42.6%, p=0.42; 32 – 34 wks: 39.4% vs. 36.57%, p=0.84).
The availability of a rescue course ACS also did not change the timing of the initial
course administration (26.1% before rescue ACS vs. 26.4% after rescue ACS, p=0.93, (Figure 2A). Similarly, only 24.7% of patients who received rescue ACS delivered within the
ACCEPTED MANUSCRIPT 10
optimal window of the rescue course (Figure 2B). Of note, 787 (58%) of all women in the study cohort delivered beyond seven days after the initial ACS dose. Delivery within the optimal window differed by indication for ACS (Table 2).
RI PT
Accuracy of administration was highest for maternal indications and lowest when given for cervical shortening. Mean gestational age of administration and of delivery is also noted in Table 2. Notably, mean gestational age of women who received initial ACS for cervical
SC
shortening was 27.2 weeks. However, mean gestational age at delivery was 35.1 weeks and the median time to delivery for women with cervical shortening was 55 days (interquartile
M AN U
range of 34-72), which was longer than any other indication.
Women receiving rescue dose steroids did not necessarily receive them for the same indication for which they received the initial course (See Table 2). Although these two populations were not directly compared, distribution of delivery within the optimal
Comment
TE D
window and median time to delivery was similar for initial course and rescue course.
EP
This study demonstrated that the availability of a rescue ACS course did not
AC C
improve the overall accuracy of ACS timing within any optimal window. The proportion of women delivering within the optimal window for the rescue ACS course was similar to the initial course, both of which were timed optimally approximately 25% of the time. These findings likely reflect the few circumstances where rescue ACS is useful and the poor predictability of preterm birth. Little data is available regarding the accuracy of rescue ACS timing. A number of prior studies have looked at accuracy of timing of initial ACS administration and
ACCEPTED MANUSCRIPT 11
demonstrate accuracy between 20 and 48%.15, 16, 18-20 As one would anticipate, accuracy of administration is higher in studies that focus on indicated preterm birth as the health care provider often dictates delivery timing.15 However, two studies that included patients with
RI PT
both labor and non-labor indications for administration found higher rates of accuracy between 41% to 45%.18, 19 Only one study restricting inclusion to only patients with
spontaneous preterm birth found overall lower accuracy than the current study (20%).16
SC
When limiting our cohort to women with a labor indication for ACS (preterm labor, PPROM, cervical shortening, ACD and vaginal bleeding) 16.8% of women delivered during any
M AN U
optimal window.
Our findings are consistent with prior studies that show ACS is most optimally timed in patients with maternal indications for administration and least accurately timed when vaginal bleeding is the indication.15, 18-20 Unlike prior reports the least accurately
TE D
timed indications for ACS administration in the current study is cervical shortening and advanced cervical dilation. Previous studies have not specifically identified cervical shortening as an indication for ACS at our institution. However, this was a clearly
EP
documented indication for administration of ACS. As mentioned above, this was defined as
AC C
asymptomatic, incidental or progressive shortening of cervical length with cervical length less than 2cm.
We found that when steroids were administered for the initial indication of cervical
shortening, accuracy in predicting delivery in the optimal window was low (2.8% delivered in the optimal window and over 97% of these patients delivered after seven days from initial ACS). Further analysis revealed that 62.5% of people with cervical shortening delivered at greater than 34 weeks. However, excluding patients with cervical shortening
ACCEPTED MANUSCRIPT 12
from the analysis, did not result in a large change in the proportion of women delivering in the optimal window (28% excluding cervical shortening compared to 26% including cervical shortening).
RI PT
This is a large cohort study of patients receiving ACS at a single institution.
Although our primary outcome sample size was large, the use of rescue steroids is low and thus the number of people who received the rescue course was small. We purposely
SC
excluded patients delivering at less than 23 6/7 weeks to avoid changes in the gestational age of resuscitation during the study years. A large portion of patients was transferred
M AN U
from outside facilities, and at times, the decision to give ACS was made prior to transfer. While transfers may influence the accuracy of ACS timing, it also adds generalizability to the study population. Those that delivered at outside hospitals were lost to follow up which could have an impact on our results with respect to delivery timing and rescue ACS
TE D
administration. While this is unlikely due to a similar distribution of excluded subjects in both groups, we expect loss to follow-up to bias findings towards fewer patients delivering within the optimal window, since many of the outside hospitals do not deliver infants less
EP
than 34 weeks gestational age.
AC C
The use of antenatal corticosteroids is increasing and timing is suboptimal, leaving more than half of patients who receive ACS to deliver after 35 weeks and risk unnecessary exposure in utero.17 The current study demonstrated a high overall preterm birth risk with 63% delivering prior to 34 weeks gestation but a low optimal ACS use. With concern for the overutilization of ACS, strategies to optimize the timing of administration, such as withholding steroids for an indication of isolated antenatal cervical shortening, may decrease unnecessary exposure.
ACCEPTED MANUSCRIPT 13
This study reiterates the difficulty in predicting preterm birth. Future focus should be aimed at tools to predict the timing of preterm birth in order to optimize ACS administration. With nearly 60% of women delivering after the optimal window for both
RI PT
initial course and rescue course ACS, one must consider the impact on ongoing pregnancies
AC C
EP
TE D
M AN U
SC
and the need for data regarding harm.
ACCEPTED MANUSCRIPT 14
Acknowledgements
RI PT
The authors thank Lisa S Ray at the University of Washington for her assistance in building the data set. We thank Dr. Melissa Schiff for programming assistance, and Jan Hamanishi for figure design.
References
5.
6.
7.
8.
9.
10.
11.
SC
M AN U
4.
TE D
3.
EP
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. Crowley PA. Antenatal corticosteroid therapy: a meta-analysis of the randomized trials, 1972 to 1994. Am J Obstet Gynecol 1995;173:322-35. Ring AM, Garland JS, Stafeil BR, Carr MH, Peckman GS, Pircon RA. The effect of a prolonged time interval between antenatal corticosteroid administration and delivery on outcomes in preterm neonates: a cohort study. Am J Obstet Gynecol 2007;196:457.e1-6. McEvoy C, Schilling D, Spitale P, Peters D, O'Malley J, Durand M. Decreased respiratory compliance in infants less than or equal to 32 weeks' gestation, delivered more than 7 days after antenatal steroid therapy. Pediatrics 2008;121:e1032-8. Kuk JY, An JJ, Cha HH, et al. Optimal time interval between a single course of antenatal corticosteroids and delivery for reduction of respiratory distress syndrome in preterm twins. Am J Obstet Gynecol 2013;209:256.e1-7. Peaceman AM, Bajaj K, Kumar P, Grobman WA. The interval between a single course of antenatal steroids and delivery and its association with neonatal outcomes. Am J Obstet Gynecol 2005;193:1165-9. Wilms FF, Vis JY, Pattinaja DA, et al. Relationship between the time interval from antenatal corticosteroid administration until preterm birth and the occurrence of respiratory morbidity. Am J Obstet Gynecol 2011;205:49.e1-7. Murphy KE, Hannah ME, Willan AR, et al. Multiple courses of antenatal corticosteroids for preterm birth (MACS): a randomised controlled trial. Lancet 2008;372:2143-51. Wapner RJ, Sorokin Y, Thom EA, et al. Single versus weekly courses of antenatal corticosteroids: evaluation of safety and efficacy. Am J Obstet Gynecol 2006;195:633-42. Guinn DA, Atkinson MW, Sullivan L, 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. French NP, Hagan R, Evans SF, Godfrey M, Newnham JP. Repeated antenatal corticosteroids: size at birth and subsequent development. Am J Obstet Gynecol 1999;180:114-21.
AC C
1.
ACCEPTED MANUSCRIPT 15
17.
18.
19. 20.
21.
RI PT
SC
16.
M AN U
15.
TE D
14.
EP
13.
Garite TJ, 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.e1-9. Vermillion ST, Bland ML, Soper DE. Effectiveness of a rescue dose of antenatal betamethasone after an initial single course. Am J Obstet Gynecol 2001;185:1086-9. ACOG Committee Opinion No. 475: antenatal corticosteroid therapy for fetal maturation. Obstet Gynecol 2011;117:422-4. Adams TM, Kinzler WL, Chavez MR, Vintzileos AM. The timing of administration of antenatal corticosteroids in women with indicated preterm birth. Am J Obstet Gynecol 2014. Adams TM, Kinzler WL, Chavez MR, Fazzari MJ, Vintzileos AM. Practice patterns in the timing of antenatal corticosteroids for fetal lung maturity. J Matern Fetal Neonatal Med 2014:1-4. Razaz N, Skoll A, Fahey J, Allen VM, Joseph KS. Trends in optimal, suboptimal, and questionably appropriate receipt of antenatal corticosteroid prophylaxis. Obstet Gynecol 2015;125:288-96. Boesveld M, Heida KY, Oudijk MA, Brouwers HA, Koenen SV, Kwee A. Evaluation of antenatal corticosteroid prescribing patterns among 984 women at risk for preterm delivery. J Matern Fetal Neonatal Med 2014;27:516-9. Vis JY, Wilms FF, Kuin RA, et al. Time to delivery after the first course of antenatal corticosteroids: a cohort study. Am J Perinatol 2011;28:683-8. Boesveld M, Oudijk MA, Koenen SV, et al. Evaluation of strategies regarding management of imminent preterm delivery before 32 weeks of gestation: a regional cohort study among 1375 women in the Netherlands. Am J Obstet Gynecol 2015;212:348 e1-7. Goldenberg RL, McClure EM. Appropriate use of antenatal corticosteroid prophylaxis. Obstet Gynecol 2015;125:285-7.
AC C
12.
ACCEPTED MANUSCRIPT 16 Table 1. Demographic and pregnancy characteristics between groups before and after availability
of rescue course protocol: 2006-2008 vs. 2009-2012 Group 2: After
Rescue Availability,
Rescue Availability,
N (%)
N (%)
N=601
N=755
Maternal Age (years) <20
48 (8) 418 (69.5)
35+
135 (22.5)
Race
528 (70.0) 171 (22.6)
75 (12.5)
100 (13.2)
Caucasian
283 (47.1)
453 (60.0)
TE D
Black
Asian
0.97
56 (7.4)
M AN U
20-34
p-value
RI PT
Group 1: Before
SC
Maternal Characteristics
<0.01
17 (2.8)
17 (2.3)
73 (12.2)
90 (119)
8 (1.3)
11 (1.5)
29 (19.3)
40 (5.8)
116 (4.8)
44 (5.3)
Prior Preterm1
112 (18.6)
145 (19.2)
0.74
Prior Cesarean
105 (17.5)
124 (16.9)
0.61
Multiple gestation
88 (14.6)
110 (14.6)
0.97
Transfer from referral center
134 (22.3)
200 (26.5)
0.07
Hypertension
93 (15.5)
123 (16.3)
0.68
Hispanic
Pacific Islander
AC C
Other
EP
Native American
ACCEPTED MANUSCRIPT 17 Diabetes
57 (9.5)
66 (8.7)
0.64
Other major medical comorbidity2
34 (5.6)
50 (6.6)
0.46
367 (62.1)
464 (62.9)
0.10
Delivery <34 weeks
Preterm delivery defined as any delivery after 20 weeks gestation
2
Major medical comorbidity including cardiac disease, renal disease, pulmonary disease,
AC C
EP
TE D
M AN U
SC
autoimmune disease, malignancy
RI PT
1
ACCEPTED MANUSCRIPT 18 Table 2. Delivery Timing and Characteristics based on Indication for Antenatal Corticosteroids
Indication Initial ACS
Delivery
Delivery
Delivery
GA1 of
GA at
Interval to
<48h
48h-7d
>7d
ACS 2
Delivery
delivery (d)
N (%)
N (%)
N (%)
Mean (SD)
Mean (SD)
Median (IQR)
N=1359
RI PT
Total
297
57 (19.2)
53 (17.9)
187 (62.9)
29.6 (3.1)
33.3 (4.5)
18 (3-43)
PPROM3
246
58 (23.6)
79 (32.1)
109 (44.3)
29.3 (3.2)
31.0 (3.2)
6 (2-15)
Antenatal cervical
106
0
3 (2.8)
103 (97.2)
27.2 (2.4)
35.1 (3.8)
55 (34-72)
Advanced dilation
32
1 (3.1)
Vaginal bleeding5
133
Maternal factors6
M AN U
shortening4
SC
Preterm labor
29 (90.6)
28.1 (3.2)
32.9 (5.1)
31 (18-45)
9 (6.8)
22 (16.5)
102 (76.7)
28.4 (3.1)
33.5 (4.6)
28 (9-54)
352
63 (17.9)
145 (41.2)
144 (40.9)
29.4 (2.7)
32.1 (3.5)
5 (2-23)
Fetal factors7
190
25 (13.1)
52 (27.4)
113 (59.5)
29.7 (2.7)
32.9 (3.4)
15 (4-37)
Rescue ACS8
N = 73
Preterm labor
22
PPROM
10
7 (31.8)
6 (27.3)
9 (40.9)
30.2 (2.0)
32.3 (3.2)
19 (3-43)
3 (30.0)
3 (30.0)
4 (40.0)
29.9 (1.8)
31.1 (2.2)
4 (1-9)
0
0
0
0
0
0
EP
Antenatal cervical
TE D
2 (6.3)
0
AC C
shortening
Advanced dilation
5
0
1 (20.0)
4 (80.0)
31.4 (1.1)
34.7 (2.6)
25 (14-33)
Vaginal bleeding
6
0
0
6 (100.0)
30.8 (1.6)
35.2 (1.6)
29 (28-40)
Maternal factors
8
5 (62.5)
1 (12.5)
2 (25.0)
30.4 (1.6)
31.6 (2.5)
2 (1-7)
Fetal factors
22
6 (27.3)
7 (31.8)
31.0 (2.0)
32.5 (2.4)
6 (2-18)
9 (40.9)
ACCEPTED MANUSCRIPT 19
Figure Legends Figure 1. Study population flowchart. Caption: ACS, antenatal corticosteroids.
RI PT
Figure 2. Delivery within the optimal window of antenatal corticosteroids before and after availability of a rescue course
SC
Caption: Delivery within the optimal window of antenatal corticosteroids for A) the initial course, B) the rescue course, and C) any antenatal course. Optimal window defined as delivery >48 hours and <7 days after administration of ACS.
AC C
EP
TE D
M AN U
Figure 3. Accuracy of antenatal corticosteroid administration in any optimal window over the study years
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S0002-9378(15)00869-8
DOI:
10.1016/j.ajog.2015.08.018
Reference:
YMOB 10582
To appear in:
American Journal of Obstetrics and Gynecology
Received Date: 20 May 2015 Revised Date:
10 July 2015
Accepted Date: 10 August 2015
Please cite this article as: Makhija NK, Tronnes AA, Dunlap BS, Schulkin J, Lannon SM, Antenatal Corticosteroid Timing: Accuracy after the Introduction of a Rescue Course Protocol, American Journal of Obstetrics and Gynecology (2015), doi: 10.1016/j.ajog.2015.08.018. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Antenatal Corticosteroid Timing: Accuracy after the Introduction of a Rescue Course Protocol
RI PT
Neeta K MAKHIJA, MD1, Ashlie A TRONNES MD 1,, Benjamin S DUNLAP2 , Jay SCHULKIN PhD3, Sophia M LANNON, MD, MPH1 1 University
of Washington School of Medicine, Department of Obstetrics and Gynecology of Washington School of Medicine, Department of Medicine 3 American College of Obstetricians and Gynecologists
The authors of this study report no conflict of interest.
SC
2 University
TE D
Corresponding Author: Neeta K Makhija, MD [email protected] (484) 629 - 5879
M AN U
An abstract describing this work was presented at the 35th Annual Pregnancy Meeting of the Society for Maternal-Fetal Medicine, San Diego, California. February 6, 2015.
AC C
EP
University of Washington Ob/GYN Box 356460 1959 NE Pacific Street Seattle, WA, 98195-6460
Word Count: Abstract: 472 Total manuscript: 2,295
Please include Table 2 in the print version of the manuscript
ACCEPTED MANUSCRIPT 2 Condensation: Introduction of rescue antenatal corticosteroids did not impact timing of the first course, nor did it increase total patients delivering within any optimal window. <24> Short version of title: Rescue Course Antenatal Corticosteroid Timing
RI PT
Reference words: antenatal corticosteroids, antenatal corticosteroids timing, preterm birth prediction, rescue antenatal corticosteroids
Abstract
SC
Background: Antenatal corticosteroid (ACS) administration is a critical fetal intervention and utilization of a rescue protocol is now standard practice. Rescue ACS may improve
M AN U
overall accuracy of ACS administration timing, but this observation and its effect on the initial course is unknown.
Objective: We sought to compare the accuracy of ACS administration before and after the
TE D
implementation of a rescue ACS protocol.
Study Design: We performed a retrospective cohort study of patients who received minimum one dose of ACS between 2006-2012 at the University of Washington Medical
EP
Center (UWMC) using the UWMC Pharmacy Database. For inclusion, subjects were required to be admitted, receive initial ACS course between 24 to 34 weeks gestational age
AC C
and deliver at UWMC. We designated two groups based on when rescue ACS became standard practice at UWMC: before rescue ACS (2006-2008) and after rescue ACS (20092012). Primary outcome was delivery within any optimal ACS window defined as 48 hours to 7 days after the first dose or third dose. We also compared delivery within the optimal window of the initial and rescue ACS courses independently and assessed ACS timing by indication for delivery. Chi squared and independent sample t-test were used to compare
ACCEPTED MANUSCRIPT 3 results. Results: Between 2006 and 2012, 1356 women met inclusion criteria, 601 before and 755
RI PT
after rescue ACS. The study groups demonstrated similar demographics with the exception of more Caucasians in the group after rescue ACS (47% vs. 60%, p<0.01) and also delivered at comparable gestational ages (32.7 wks. vs. 32.6 wks, p=0.59). Availability of a second
SC
course did not increase total subjects delivering within any optimal window (26.5% vs. 28.5%, p=0.41). Frequency of delivery within initial course optimal window did not change
M AN U
after introduction of rescue course protocol (26.1% vs. 26.4%, p=0.92). Similarly, of the 73 subjects who received rescue ACS, 24.7% delivered in the optimal window of the second course. Delivery within optimal window varied by indication for ACS, with highest accuracy among maternal indications (41.2% in any optimal window), followed by PPROM (32.1%).
TE D
Lowest administration accuracy was among women with cervical shortening and advanced cervical dilation with only 2.8% and 6.3% delivering within the optimal window, respectively. Furthermore, for women with cervical shortening, the mean gestational age of
EP
delivery was 35.1 weeks and the median interval from ACS administration to delivery was 55 days (IQR 34-72 days).
AC C
Conclusions: The opportunity for a second course of ACS did not improve the number of women delivering within any optimal ACS window. Administration timing was similar for the initial course and the rescue course, with approximately one quarter of women delivering within the optimal ACS window. These findings likely reflect the few circumstances where rescue ACS is useful and the poor predictability of preterm birth.
ACCEPTED MANUSCRIPT 4 Future focus should be aimed at tools to predict the timing of preterm birth in order to
AC C
EP
TE D
M AN U
SC
RI PT
optimize ACS administration.
ACCEPTED MANUSCRIPT 5 Introduction Antenatal corticosteroids (ACS) are an important intervention used to improve
RI PT
neonatal outcomes after preterm birth. Since the landmark study by Liggins and Howie in 1972,1 several studies confirm that treatment with ACS is associated with an overall
reduction in neonatal death, respiratory distress syndrome and intensive care admissions.2
SC
ACS offer the greatest benefit to the fetus if the infant is delivered 24-48 hours after initial dose and within seven days of the administration.2, 3 Benefits decrease after this optimal
M AN U
window.4-7 However, prediction of preterm birth is challenging, especially within a narrow timeframe. While one strategy to improve timing of administration is to provide repeated courses of ACS, studies demonstrate that multiple courses carry fetal risks.8-11 More recent research demonstrates that a single rescue course of ACS may also improve neonatal
TE D
outcomes when delivered within the optimal window.12, 13 In 2011 and 2012, the American College of Obstetricians and Gynecologists (ACOG) endorsed administering a single course of rescue steroids for women who remain at risk for preterm delivery, commonly known as
EP
the rescue protocol.14
Studies demonstrate that the majority of women receiving initial ACS deliver prior
AC C
to 34 weeks gestation, but accuracy of administration within the optimal window is low and varies by indication for delivery. 15-20 The accuracy of rescue ACS administration remains unknown. Recently, studies demonstrating combined increased utilization and inaccurate timing of ACS have raised concern for long-term harmful effects on infants born outside of the optimal window.17, 21 Furthermore, few data report on the long-term harm of a single rescue course of ACS. For these reasons ensuring that the availability of rescue course ACS increases the number of women receiving ACS within any optimal window
ACCEPTED MANUSCRIPT 6 remains a priority. Our objective is to compare the accuracy of ACS administration before and after the availability of a rescue course protocol and to determine whether rescue ACS
RI PT
increases the proportion of women delivering within any optimal ACS window.
Materials and Methods
We conducted a retrospective cohort study of women receiving antenatal
SC
corticosteroids, comparing the proportion of patients who delivered in the optimal window of the initial ACS course before and after the availability of rescue ACS. We identified
M AN U
subjects using the inpatient hospital pharmacy database, capturing all women admitted to the labor and delivery unit. Women were included if they were ordered at least one dose of ACS between January 2006 and December 2012 at the University of Washington Medical Center, a level three-referral hospital. Data was then linked to the University of Washington
TE D
Perinatal Database for additional demographic data. A single primary researcher then performed a review of each chart to confirm and obtain additional data. The Human Subjects Division at the University of Washington approved this study (IRB # 44589, Date
EP
of Approval 2/28/2015).
AC C
The cohort included women admitted to labor and delivery and receiving at least one dose of betamethasone 12mg IM, which is the ACS used almost exclusively at our and referring institutions. Patients were excluded if they received betamethasone at less than 24 0/7 weeks or greater than 33 6/7 weeks gestational age, delivered at an outside institution or we could not confirm administration of ACS by chart review. Some women received their first dose of ACS prior to transfer to our institution, and they were included if ACS administration was confirmed.
ACCEPTED MANUSCRIPT 7 Two groups were identified according to the year of initial ACS course administration. In 2009, the Maternal Fetal Medicine group at the University of Washington Medical Center developed a consensus policy supporting the use of a single rescue course
RI PT
of ACS. The first group received the initial ACS course between 2006 and 2008, which was before implementation of our rescue steroid protocol, and the second group received initial ACS course between 2009 and 2012, which was after wide spread use of the rescue course
SC
protocol.
We compared maternal, pregnancy and delivery characteristics between the groups
M AN U
before and after the availability of rescue steroids. The potential indications for administration of ACS were categorized as: preterm labor, preterm premature rupture of membranes, antenatal cervical shortening, advanced cervical dilation, vaginal bleeding, maternal factors and fetal factors. Preterm labor was defined as regular, painful
TE D
contractions with cervical change. Antenatal cervical shortening was used to describe women who were asymptomatic though found to have a cervical length < 2 cm on transvaginal ultrasound. Many of these women had additional risk factors for preterm
EP
birth. Advanced cervical dilation was defined as dilation > 2 cm and patients were only
AC C
given this diagnosis if they did not meet criteria for preterm labor. If patients had more than one indication, the investigator assigned a primary indication. We compared the proportion of patients delivering within the optimal window before and after the availability of a rescue course protocol. The primary outcome was delivery within optimal window of either the initial ACS course or the rescue ACS course. The optimal window for ACS administration is defined as delivery between 48 hours and 7 days after the first dose for the initial course and between
ACCEPTED MANUSCRIPT 8 48 hours and 7 days after the third dose for the rescue course. Although data suggest that steroid benefit can occur after 24 hours from the first dose, our institution routinely considers maximum benefit at 48 hours and uses this time point in clinical decision-making
RI PT
and delivery planning. The secondary outcomes included delivery within optimal window of the rescue course and delivery within the optimal window of the initial ACS course.
Additionally, we assessed delivery timing and delivery characteristics by indication for
SC
administration.
Sample size was determined by the availability of subjects receiving ACS within the
M AN U
study time period. With our sample size of 1356, the assumption that 30% of women received ACS within the optimal window, alpha of 0.05 and power of 80%, the minimal effect size detectable is 7%. Statistical analysis was completed using STATA software version 11.1. Pearson chi-squared tests were used to compare categorical variables and
TE D
independent sample t-tests were used to compare continuous variables. For all analyses a
Results
EP
two-sided significance level of <0.05 was considered statistically significant.
AC C
We identified 1809 pregnant women for whom a pharmacy order was placed for betamethasone within the study period. After exclusions for not receiving ACS, delivering at an outside hospital, or gestational age, the final study population was 1356 subjects (Figure 1). Of the total cohort, 601 received ACS in 2006-2008 before availability of rescue protocol and 755 received ACS in 2009-2012 after availability of rescue protocol. Of note, similar proportions of patients were excluded from each group secondary to delivery at an outside institution (20.2% before rescue ACS vs. 18% after rescue ACS, p= 0.24). However,
ACCEPTED MANUSCRIPT 9 women included in the study were on average 1.7 years older than those excluded for outside hospital delivery, p<0.01. Additional demographic data on the excluded group was not accessible.
RI PT
Table 1 reveals the maternal characteristics of our population. Study groups were similar with the exception of racial distribution. There was a high representation of
multiple gestations, transfer to higher level of care, hypertension, diabetes, and those with
SC
other medical comorbidities in our study group. As anticipated, there was a high
proportion of preterm deliveries. The study groups delivered at similar gestational ages
M AN U
with approximately 62% of the population delivering less than 34 weeks gestation. The primary outcome of delivery within the optimal window of either the initial or rescue ACS course, did not change after the availability of rescue ACS (26.5% before rescue ACS vs. 28.5% after rescue ACS, p=0.41, (Figure 2C). Considering interval to delivery as a
TE D
continuous variable did not impact our results (23.1 days before rescue ACS, vs. 23.7 days after rescue ACS, p=0.66). Furthermore, when examining accuracy of administration over the study years, ACS utilization and accuracy appeared stable over time (Figure 3). Finally,
EP
when stratified by gestational age, there was no statistical difference in the proportion of
AC C
patients who delivered in any optimal window before and after the availability of rescue steroids (24 - 28 wks: 43.5% before rescue ACS vs. 51.6% after rescue ACS, p=0.5; 28 - 32 wks: 35.7% vs. 42.6%, p=0.42; 32 – 34 wks: 39.4% vs. 36.57%, p=0.84).
The availability of a rescue course ACS also did not change the timing of the initial
course administration (26.1% before rescue ACS vs. 26.4% after rescue ACS, p=0.93, (Figure 2A). Similarly, only 24.7% of patients who received rescue ACS delivered within the
ACCEPTED MANUSCRIPT 10
optimal window of the rescue course (Figure 2B). Of note, 787 (58%) of all women in the study cohort delivered beyond seven days after the initial ACS dose. Delivery within the optimal window differed by indication for ACS (Table 2).
RI PT
Accuracy of administration was highest for maternal indications and lowest when given for cervical shortening. Mean gestational age of administration and of delivery is also noted in Table 2. Notably, mean gestational age of women who received initial ACS for cervical
SC
shortening was 27.2 weeks. However, mean gestational age at delivery was 35.1 weeks and the median time to delivery for women with cervical shortening was 55 days (interquartile
M AN U
range of 34-72), which was longer than any other indication.
Women receiving rescue dose steroids did not necessarily receive them for the same indication for which they received the initial course (See Table 2). Although these two populations were not directly compared, distribution of delivery within the optimal
Comment
TE D
window and median time to delivery was similar for initial course and rescue course.
EP
This study demonstrated that the availability of a rescue ACS course did not
AC C
improve the overall accuracy of ACS timing within any optimal window. The proportion of women delivering within the optimal window for the rescue ACS course was similar to the initial course, both of which were timed optimally approximately 25% of the time. These findings likely reflect the few circumstances where rescue ACS is useful and the poor predictability of preterm birth. Little data is available regarding the accuracy of rescue ACS timing. A number of prior studies have looked at accuracy of timing of initial ACS administration and
ACCEPTED MANUSCRIPT 11
demonstrate accuracy between 20 and 48%.15, 16, 18-20 As one would anticipate, accuracy of administration is higher in studies that focus on indicated preterm birth as the health care provider often dictates delivery timing.15 However, two studies that included patients with
RI PT
both labor and non-labor indications for administration found higher rates of accuracy between 41% to 45%.18, 19 Only one study restricting inclusion to only patients with
spontaneous preterm birth found overall lower accuracy than the current study (20%).16
SC
When limiting our cohort to women with a labor indication for ACS (preterm labor, PPROM, cervical shortening, ACD and vaginal bleeding) 16.8% of women delivered during any
M AN U
optimal window.
Our findings are consistent with prior studies that show ACS is most optimally timed in patients with maternal indications for administration and least accurately timed when vaginal bleeding is the indication.15, 18-20 Unlike prior reports the least accurately
TE D
timed indications for ACS administration in the current study is cervical shortening and advanced cervical dilation. Previous studies have not specifically identified cervical shortening as an indication for ACS at our institution. However, this was a clearly
EP
documented indication for administration of ACS. As mentioned above, this was defined as
AC C
asymptomatic, incidental or progressive shortening of cervical length with cervical length less than 2cm.
We found that when steroids were administered for the initial indication of cervical
shortening, accuracy in predicting delivery in the optimal window was low (2.8% delivered in the optimal window and over 97% of these patients delivered after seven days from initial ACS). Further analysis revealed that 62.5% of people with cervical shortening delivered at greater than 34 weeks. However, excluding patients with cervical shortening
ACCEPTED MANUSCRIPT 12
from the analysis, did not result in a large change in the proportion of women delivering in the optimal window (28% excluding cervical shortening compared to 26% including cervical shortening).
RI PT
This is a large cohort study of patients receiving ACS at a single institution.
Although our primary outcome sample size was large, the use of rescue steroids is low and thus the number of people who received the rescue course was small. We purposely
SC
excluded patients delivering at less than 23 6/7 weeks to avoid changes in the gestational age of resuscitation during the study years. A large portion of patients was transferred
M AN U
from outside facilities, and at times, the decision to give ACS was made prior to transfer. While transfers may influence the accuracy of ACS timing, it also adds generalizability to the study population. Those that delivered at outside hospitals were lost to follow up which could have an impact on our results with respect to delivery timing and rescue ACS
TE D
administration. While this is unlikely due to a similar distribution of excluded subjects in both groups, we expect loss to follow-up to bias findings towards fewer patients delivering within the optimal window, since many of the outside hospitals do not deliver infants less
EP
than 34 weeks gestational age.
AC C
The use of antenatal corticosteroids is increasing and timing is suboptimal, leaving more than half of patients who receive ACS to deliver after 35 weeks and risk unnecessary exposure in utero.17 The current study demonstrated a high overall preterm birth risk with 63% delivering prior to 34 weeks gestation but a low optimal ACS use. With concern for the overutilization of ACS, strategies to optimize the timing of administration, such as withholding steroids for an indication of isolated antenatal cervical shortening, may decrease unnecessary exposure.
ACCEPTED MANUSCRIPT 13
This study reiterates the difficulty in predicting preterm birth. Future focus should be aimed at tools to predict the timing of preterm birth in order to optimize ACS administration. With nearly 60% of women delivering after the optimal window for both
RI PT
initial course and rescue course ACS, one must consider the impact on ongoing pregnancies
AC C
EP
TE D
M AN U
SC
and the need for data regarding harm.
ACCEPTED MANUSCRIPT 14
Acknowledgements
RI PT
The authors thank Lisa S Ray at the University of Washington for her assistance in building the data set. We thank Dr. Melissa Schiff for programming assistance, and Jan Hamanishi for figure design.
References
5.
6.
7.
8.
9.
10.
11.
SC
M AN U
4.
TE D
3.
EP
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. Crowley PA. Antenatal corticosteroid therapy: a meta-analysis of the randomized trials, 1972 to 1994. Am J Obstet Gynecol 1995;173:322-35. Ring AM, Garland JS, Stafeil BR, Carr MH, Peckman GS, Pircon RA. The effect of a prolonged time interval between antenatal corticosteroid administration and delivery on outcomes in preterm neonates: a cohort study. Am J Obstet Gynecol 2007;196:457.e1-6. McEvoy C, Schilling D, Spitale P, Peters D, O'Malley J, Durand M. Decreased respiratory compliance in infants less than or equal to 32 weeks' gestation, delivered more than 7 days after antenatal steroid therapy. Pediatrics 2008;121:e1032-8. Kuk JY, An JJ, Cha HH, et al. Optimal time interval between a single course of antenatal corticosteroids and delivery for reduction of respiratory distress syndrome in preterm twins. Am J Obstet Gynecol 2013;209:256.e1-7. Peaceman AM, Bajaj K, Kumar P, Grobman WA. The interval between a single course of antenatal steroids and delivery and its association with neonatal outcomes. Am J Obstet Gynecol 2005;193:1165-9. Wilms FF, Vis JY, Pattinaja DA, et al. Relationship between the time interval from antenatal corticosteroid administration until preterm birth and the occurrence of respiratory morbidity. Am J Obstet Gynecol 2011;205:49.e1-7. Murphy KE, Hannah ME, Willan AR, et al. Multiple courses of antenatal corticosteroids for preterm birth (MACS): a randomised controlled trial. Lancet 2008;372:2143-51. Wapner RJ, Sorokin Y, Thom EA, et al. Single versus weekly courses of antenatal corticosteroids: evaluation of safety and efficacy. Am J Obstet Gynecol 2006;195:633-42. Guinn DA, Atkinson MW, Sullivan L, 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. French NP, Hagan R, Evans SF, Godfrey M, Newnham JP. Repeated antenatal corticosteroids: size at birth and subsequent development. Am J Obstet Gynecol 1999;180:114-21.
AC C
1.
ACCEPTED MANUSCRIPT 15
17.
18.
19. 20.
21.
RI PT
SC
16.
M AN U
15.
TE D
14.
EP
13.
Garite TJ, 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.e1-9. Vermillion ST, Bland ML, Soper DE. Effectiveness of a rescue dose of antenatal betamethasone after an initial single course. Am J Obstet Gynecol 2001;185:1086-9. ACOG Committee Opinion No. 475: antenatal corticosteroid therapy for fetal maturation. Obstet Gynecol 2011;117:422-4. Adams TM, Kinzler WL, Chavez MR, Vintzileos AM. The timing of administration of antenatal corticosteroids in women with indicated preterm birth. Am J Obstet Gynecol 2014. Adams TM, Kinzler WL, Chavez MR, Fazzari MJ, Vintzileos AM. Practice patterns in the timing of antenatal corticosteroids for fetal lung maturity. J Matern Fetal Neonatal Med 2014:1-4. Razaz N, Skoll A, Fahey J, Allen VM, Joseph KS. Trends in optimal, suboptimal, and questionably appropriate receipt of antenatal corticosteroid prophylaxis. Obstet Gynecol 2015;125:288-96. Boesveld M, Heida KY, Oudijk MA, Brouwers HA, Koenen SV, Kwee A. Evaluation of antenatal corticosteroid prescribing patterns among 984 women at risk for preterm delivery. J Matern Fetal Neonatal Med 2014;27:516-9. Vis JY, Wilms FF, Kuin RA, et al. Time to delivery after the first course of antenatal corticosteroids: a cohort study. Am J Perinatol 2011;28:683-8. Boesveld M, Oudijk MA, Koenen SV, et al. Evaluation of strategies regarding management of imminent preterm delivery before 32 weeks of gestation: a regional cohort study among 1375 women in the Netherlands. Am J Obstet Gynecol 2015;212:348 e1-7. Goldenberg RL, McClure EM. Appropriate use of antenatal corticosteroid prophylaxis. Obstet Gynecol 2015;125:285-7.
AC C
12.
ACCEPTED MANUSCRIPT 16 Table 1. Demographic and pregnancy characteristics between groups before and after availability
of rescue course protocol: 2006-2008 vs. 2009-2012 Group 2: After
Rescue Availability,
Rescue Availability,
N (%)
N (%)
N=601
N=755
Maternal Age (years) <20
48 (8) 418 (69.5)
35+
135 (22.5)
Race
528 (70.0) 171 (22.6)
75 (12.5)
100 (13.2)
Caucasian
283 (47.1)
453 (60.0)
TE D
Black
Asian
0.97
56 (7.4)
M AN U
20-34
p-value
RI PT
Group 1: Before
SC
Maternal Characteristics
<0.01
17 (2.8)
17 (2.3)
73 (12.2)
90 (119)
8 (1.3)
11 (1.5)
29 (19.3)
40 (5.8)
116 (4.8)
44 (5.3)
Prior Preterm1
112 (18.6)
145 (19.2)
0.74
Prior Cesarean
105 (17.5)
124 (16.9)
0.61
Multiple gestation
88 (14.6)
110 (14.6)
0.97
Transfer from referral center
134 (22.3)
200 (26.5)
0.07
Hypertension
93 (15.5)
123 (16.3)
0.68
Hispanic
Pacific Islander
AC C
Other
EP
Native American
ACCEPTED MANUSCRIPT 17 Diabetes
57 (9.5)
66 (8.7)
0.64
Other major medical comorbidity2
34 (5.6)
50 (6.6)
0.46
367 (62.1)
464 (62.9)
0.10
Delivery <34 weeks
Preterm delivery defined as any delivery after 20 weeks gestation
2
Major medical comorbidity including cardiac disease, renal disease, pulmonary disease,
AC C
EP
TE D
M AN U
SC
autoimmune disease, malignancy
RI PT
1
ACCEPTED MANUSCRIPT 18 Table 2. Delivery Timing and Characteristics based on Indication for Antenatal Corticosteroids
Indication Initial ACS
Delivery
Delivery
Delivery
GA1 of
GA at
Interval to
<48h
48h-7d
>7d
ACS 2
Delivery
delivery (d)
N (%)
N (%)
N (%)
Mean (SD)
Mean (SD)
Median (IQR)
N=1359
RI PT
Total
297
57 (19.2)
53 (17.9)
187 (62.9)
29.6 (3.1)
33.3 (4.5)
18 (3-43)
PPROM3
246
58 (23.6)
79 (32.1)
109 (44.3)
29.3 (3.2)
31.0 (3.2)
6 (2-15)
Antenatal cervical
106
0
3 (2.8)
103 (97.2)
27.2 (2.4)
35.1 (3.8)
55 (34-72)
Advanced dilation
32
1 (3.1)
Vaginal bleeding5
133
Maternal factors6
M AN U
shortening4
SC
Preterm labor
29 (90.6)
28.1 (3.2)
32.9 (5.1)
31 (18-45)
9 (6.8)
22 (16.5)
102 (76.7)
28.4 (3.1)
33.5 (4.6)
28 (9-54)
352
63 (17.9)
145 (41.2)
144 (40.9)
29.4 (2.7)
32.1 (3.5)
5 (2-23)
Fetal factors7
190
25 (13.1)
52 (27.4)
113 (59.5)
29.7 (2.7)
32.9 (3.4)
15 (4-37)
Rescue ACS8
N = 73
Preterm labor
22
PPROM
10
7 (31.8)
6 (27.3)
9 (40.9)
30.2 (2.0)
32.3 (3.2)
19 (3-43)
3 (30.0)
3 (30.0)
4 (40.0)
29.9 (1.8)
31.1 (2.2)
4 (1-9)
0
0
0
0
0
0
EP
Antenatal cervical
TE D
2 (6.3)
0
AC C
shortening
Advanced dilation
5
0
1 (20.0)
4 (80.0)
31.4 (1.1)
34.7 (2.6)
25 (14-33)
Vaginal bleeding
6
0
0
6 (100.0)
30.8 (1.6)
35.2 (1.6)
29 (28-40)
Maternal factors
8
5 (62.5)
1 (12.5)
2 (25.0)
30.4 (1.6)
31.6 (2.5)
2 (1-7)
Fetal factors
22
6 (27.3)
7 (31.8)
31.0 (2.0)
32.5 (2.4)
6 (2-18)
9 (40.9)
ACCEPTED MANUSCRIPT 19
Figure Legends Figure 1. Study population flowchart. Caption: ACS, antenatal corticosteroids.
RI PT
Figure 2. Delivery within the optimal window of antenatal corticosteroids before and after availability of a rescue course
SC
Caption: Delivery within the optimal window of antenatal corticosteroids for A) the initial course, B) the rescue course, and C) any antenatal course. Optimal window defined as delivery >48 hours and <7 days after administration of ACS.
AC C
EP
TE D
M AN U
Figure 3. Accuracy of antenatal corticosteroid administration in any optimal window over the study years
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT