- Email: [email protected]
Assessing the potential impact of extending antenatal steroids to the late preterm period
Accepted Manuscript Assessing the Potential Impact of Extending Antenatal Steroids to the Late Preterm Period Vivienne Souter, MD, Ellen Kauffman, MD, Ms Alice J. Marshall, BS, MFA, Jodie G. Katon, PhD, MS PII:
S0002-9378(17)30558-6
DOI:
10.1016/j.ajog.2017.04.029
Reference:
YMOB 11633
To appear in:
American Journal of Obstetrics and Gynecology
Received Date: 3 March 2017 Accepted Date: 17 April 2017
Please cite this article as: Souter V, Kauffman E, Marshall MAJ, Katon JG, Assessing the Potential Impact of Extending Antenatal Steroids to the Late Preterm Period, American Journal of Obstetrics and Gynecology (2017), doi: 10.1016/j.ajog.2017.04.029. 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 1 Assessing the Potential Impact of Extending Antenatal Steroids to the Late Preterm
2
Period
3
Vivienne SOUTER, MD
4
Foundation for Health Care Quality, Seattle, WA
5
Ellen KAUFFMAN, MD
6
Foundation for Health Care Quality, Seattle, WA
7
Ms Alice J. MARSHALL, BS, MFA
8
Foundation for Health Care Quality, Seattle, WA
9
Jodie G KATON, PhD, MS
M AN U
SC
RI PT
1
US Department of Veterans Affairs (VA), Health Services Research and Development
11
Center of Innovation for Veteran-Centered and Value Driven Care, VA Puget Sound
12
Healthcare System, Seattle, WA
13
Disclosure Statement: The authors report no conflict of interest.
14
Funding Source: The Foundation for Health Care Quality is a 501(c)(3) non-profit
15
organization supported by membership dues from the participants in its programs. There
16
was no other funding for this study.
17
Paper presentation information: This work was presented at The Pregnancy Meeting,
18
The 37th Annual Meeting of the SMFM, Las Vegas, Jan 23-28, 2017.
19
Disclaimer for Dr. Katon: The views expressed in this article are those of the authors
20
and do not necessarily reflect the position or policy of the Department of Veterans
21
Affairs.
AC C
EP
TE D
10
ACCEPTED MANUSCRIPT 2 Corresponding Author: Vivienne SOUTER, MD
23
Institution: Obstetrics Clinical Outcomes Assessment Program, Foundation for Health
24
Care Quality, 705 Second Avenue, Suite 410, Seattle, WA 98104
25
Telephone: (206) 375 0234 (cell)
26
Email: [email protected]
28
Word Count: Abstract – 377. Main text – 2974.
30
Table for Print Copy: Table 2
31
M AN U
29
SC
27
RI PT
22
32
Condensation:
33
steroids, most newborn respiratory complications occur in those delivering before 36
34
weeks.
38
Potential impact of late preterm antenatal steroid guidelines
EP
37
Short Title:
AC C
36
TE D
35
Among late preterm pregnancies potentially eligible for antenatal
ACCEPTED MANUSCRIPT 3 Abstract:
40
Background: In 2016 guidance statements were issued by the Society for Maternal Fetal
41
Medicine and by the American College of Obstetricians and Gynecologists about
42
extending antenatal steroid use to selected late preterm singleton pregnancies.
43
Objective: To review antenatal steroid use prior to the 2016 guidance statements and
44
assess the potential impact of these.
45
Study Design This cohort study used chart-abstracted data from singleton deliveries from
46
January 1, 2012 to March 31, 2016 at 12 centers participating in the Obstetrics Clinical
47
Outcomes Assessment Program, a quality initiative in Washington State. Pregnancies
48
with missing gestation at delivery, fetal anomalies, or antepartum demise were excluded.
49
Antenatal steroid use prior to the 2016 guidance was evaluated based on the percentage
50
of early preterm deliveries (23+0-33+6 weeks) and the percentage of all pregnancies that
51
received antenatal steroids. Newborn complication rates were calculated for late preterm
52
deliveries (34+0+0-36+6 weeks), grouped by whether they would be potentially eligible or
53
ineligible for antenatal steroids based on the 2016 guidance statements.
54
Results The opportunity for antenatal steroids was missed in 21.8% (226/1034) of early
55
preterm deliveries and of all those who received antenatal steroids, 32.2% (614/1908)
56
delivered at term. Seventy-four percent of preterm deliveries (n=2942) were in the late
57
preterm period. Eighty percent (n=2363) of late preterm deliveries were potentially
58
eligible for antenatal steroids and 60% of these (n=1411) delivered at 36 weeks. The rate
59
of respiratory complications in newborns delivering at 34 and 35 weeks was higher in the
60
group potentially eligible for late preterm antenatal steroids compared to those in the
61
ineligible group. Of those delivering at 36 weeks, no differences were detected in
AC C
EP
TE D
M AN U
SC
RI PT
39
ACCEPTED MANUSCRIPT 4 prevalence of respiratory complications by potential eligibility for antenatal steroids;
63
however, compared with the ineligible group, those potentially eligible had a lower risk
64
of neonatal intensive care unit admission (p<0.001). More than 2/3 (69%; 171/248) of
65
newborn respiratory complications among late preterm deliveries potentially eligible for
66
antenatal steroids occurred in those delivering at 34-35 weeks. The highest rate of
67
respiratory complications was in those ineligible for antenatal steroids due to pre-
68
pregnancy diabetes or chorioamnionitis, regardless of gestational age at delivery.
69
Conclusions Careful consideration of which pregnancies should receive late preterm
70
antenatal steroids and how to identify these pregnancies is important to optimize benefits
71
and mitigate potential risks of this intervention.
M AN U
SC
RI PT
62
AC C
EP
TE D
72 73 74
ACCEPTED MANUSCRIPT 5 75
Introduction: In 2016, a randomized controlled trial reported that a course of late preterm
77
betamethasone (34+0-36+6 weeks) given to women with singleton pregnancies considered
78
at risk for delivery before 37 weeks, was associated with a significant decrease in the risk
79
of newborn respiratory complications.1 These results prompted statements from the
80
American College of Obstetrics and Gynecology (ACOG) and the Society for Maternal
81
Fetal Medicine (SMFM) in April 2016 and two subsequent guidance documents
82
published later in 2016 supporting antenatal steroids for selected women with late
83
preterm (34+0 -36+6 weeks) singleton pregnancies considered to be at risk for preterm
84
delivery.2,3
85
with chorioamnionitis, and those with pre-pregnancy diabetes.
M AN U
SC
RI PT
76
Excluded from these recommendations were multiple pregnancies, those
Nationally, approximately 70% of preterm births and 8% of all deliveries occur at
87
34-36 weeks.4 Therefore, the new guidance statements have the potential to expose a
88
much larger percentage of the pregnant population to antenatal steroids. For antenatal
89
steroid exposure at <34 weeks of gestation, there are clear benefits for those born
90
prematurely and no obvious adverse effects on long-term follow up after a single course
91
of antenatal steroids.5 However, in the last two years a number of researchers have raised
92
concerns about the potential for negative as well as positive effects of antenatal steroids
93
in general
94
particular there is a paucity of data on long-term outcomes following antenatal steroid
95
exposure at >34 weeks.12 Given the large number of pregnancies potentially affected by
96
this change in practice, the short- and long-term risk benefit ratio is an important
97
consideration. Our goal was to review antenatal steroid use prior to the 2016 ACOG and
AC C
EP
TE D
86
6-9
and specifically about the risk-benefit ratio for late preterm steroids.10-12 In
ACCEPTED MANUSCRIPT 6 98
SMFM statements and to assess the potential impact of implementing the 2016
99
recommendations.
100
AC C
EP
TE D
M AN U
SC
RI PT
101
ACCEPTED MANUSCRIPT 7 102
Study Design This retrospective cohort study used chart-abstracted data from deliveries at 12
104
centers participating in the Obstetrics Clinical Outcomes Assessment Program (OB
105
COAP), a quality initiative in Washington State. These centers all had Level II or III
106
neonatal units and included both urban and rural settings. Centers contributing data for all
107
or part of the study period (January 1, 2012-March 31, 2016) were included in the study.
108
The analysis was restricted to singleton pregnancies delivering between 23+0 and 42+6
109
weeks of gestation. Deliveries with missing gestational age at delivery, fetal anomalies,
110
or antepartum demise were excluded.
M AN U
SC
RI PT
103
OB COAP captures consecutive deliveries at participating centers and records a
112
wide range of variables including those related to maternal demographic characteristics,
113
pre-pregnancy health, pregnancy complications, labor course, delivery, and postnatal
114
outcomes for mothers and newborns. Abstracted data are entered into a cloud-based,
115
standardized data tool. Data undergo real-time quality checks performed both at the site
116
and aggregate level. Monthly web meetings and unlimited access to OB COAP staff for
117
education and support are available. Audit of OB COAP data against billing records with
118
a minimum of 90% agreement is required for the program.
EP
TE D
111
The Western Institutional Review Board determined in 2015 that OB COAP is not
120
engaged in human subjects research and therefore is exempt from institutional review
121
board review.
122
AC C
119
Early preterm birth was defined as delivery between 23+0 and 33+6 weeks and late
123
preterm birth as delivery between 34+0 and 36+6 weeks of gestation. Pregnancies that
124
delivered in the late preterm period were divided into two groups based on the 2016
125
ACOG / SMFM guidance: those that would be considered potentially eligible and those
ACCEPTED MANUSCRIPT 8 that would be considered ineligible for late preterm antenatal steroids. Those considered
127
to be ineligible for antenatal steroids were further divided into those that would be
128
considered ineligible due to having already received antenatal steroids and those that had
129
not received earlier steroids but were ineligible due to chorioamnionitis or pre-pregnancy
130
diabetes. Data on maternal antibiotic treatment for chorioamnionitis were available only
131
for 2015 and 2016.
RI PT
126
Newborn outcomes evaluated included respiratory complications, need for
133
resuscitation at delivery, glucose instability, and admission to the neonatal intensive care
134
unit (NICU). Other descriptive variables included maternal demographics and pregnancy,
135
delivery, and newborn characteristics.
M AN U
SC
132
Descriptive statistics included the rate of preterm delivery and late preterm
137
delivery, the percentage of all deliveries that received antenatal steroids, and the
138
percentage that received antenatal steroids and subsequently delivered late preterm or at
139
term. We also report the percentage of early preterm deliveries where the opportunity for
140
antenatal steroids was missed (those delivering before 34 weeks of gestation that did not
141
receive antenatal steroids).
EP
TE D
136
Maternal demographic, pregnancy, delivery, and newborn characteristics are
143
reported for late preterm deliveries by potential eligibility for antenatal steroids. Newborn
144
outcomes among late preterm deliveries are reported by potential eligibility for antenatal
145
steroids (eligible versus ineligible) and gestational week at delivery (34, 35, 36 weeks).
146
Frequency of outcomes between late preterm deliveries potentially eligible or ineligible
147
for antenatal steroids was compared using chi-square tests. As deliveries ineligible for
148
antenatal steroids represented a heterogeneous group in terms of obstetric risk factors, we
AC C
142
ACCEPTED MANUSCRIPT 9 also qualitatively compared newborn outcomes between potential eligibility and
150
ineligibility for the late preterm antenatal steroids, subdividing the ineligible group into
151
those who were ineligible due to prior steroids and those who were ineligible due to
152
chorioamnionitis or pre-pregnancy diabetes.
153 154
AC C
EP
TE D
M AN U
SC
155
RI PT
149
ACCEPTED MANUSCRIPT 10 156
Results A total of 62300 singleton deliveries from 12 centers were entered into the
158
database during the study period. Of these, 60190 deliveries (97%) met the inclusion
159
criteria. The prevalence of preterm delivery (23+0-36+6 weeks) was 6.6% (n=3976).
160
Seventy-four percent of preterm deliveries were in the late preterm period (n=2942),
161
representing 4.9% of all singleton deliveries.
RI PT
157
Three percent (n=1908) of all pregnancies received antenatal steroids at some
163
point. Of the 1908 pregnancies exposed to antenatal steroids, 42.3% (n=808) delivered in
164
the early preterm period, 25.5% (n=486) in the late preterm period, and 32.2% (n=614) at
165
term. Among the 1034 early preterm deliveries the opportunity for antenatal steroids was
166
missed in 21.9% (n=226).
M AN U
SC
162
Based on the 2016 guidance, 80.3% (n=2363) of late preterm deliveries would be
168
considered eligible for antenatal steroids if late preterm birth was anticipated with enough
169
time to administer steroids. Of the late preterm population, 16.3% (480/2942) were
170
ineligible due to having already received antenatal steroids and 3.4% (99/2942) were
171
ineligible due to pre-pregnancy diabetes or chorioamnionitis, Those potentially eligible
172
for antenatal steroids tended to be younger, were less likely to be African American Non
173
Hispanic, had a lower prevalence of pre-pregnancy hypertension, and had a lower
174
prevalence of pre-eclampsia/gestational hypertension, compared with those who were
175
ineligible (Table 1).
176
potentially eligible for antenatal steroids were more likely to have a vaginal delivery and
177
to deliver at 36 weeks of gestation.
AC C
EP
TE D
167
Additionally, compared with those who were ineligible, those
ACCEPTED MANUSCRIPT 11 Newborn respiratory complications were recorded in 10.6% (312/2942) of all late
179
preterm deliveries in the study population and 79.5% (248/312) of these occurred in
180
pregnancies potentially eligible for antenatal steroids. Examining gestational week at
181
delivery among newborns with respiratory complications by eligibility for antenatal
182
steroids revealed that more than two thirds (69.0%; 171/248) of newborn respiratory
183
complications in those potentially eligible for antenatal steroids occurred among those
184
delivering at 34-35 weeks (Figure 1).
SC
RI PT
178
Of pregnancies delivering at 34 and 35 weeks of gestation, those potentially
186
eligible for antenatal steroids had a higher rate of newborn respiratory complications than
187
those who were ineligible (p< 0.05) (Table 2). For deliveries at 36 weeks, there was no
188
statistically significant difference in newborn respiratory complications between those
189
potentially eligible for antenatal steroids and those who were ineligible. No differences in
190
resuscitation at delivery or glucose instability were detected by potential eligibility for
191
antenatal steroids. Compared to those ineligible for antenatal steroids, those who were
192
potentially eligible had a lower rate of NICU admission regardless of gestational week at
193
delivery (Table 2).
EP
TE D
M AN U
185
Subdividing those ineligible for steroids by reason for ineligibility (prior receipt
195
of steroids, pre-pregnancy diabetes or chorioamnionitis) and comparing them to those
196
who were potentially eligible, slightly different patterns emerged. At nearly all
197
gestational weeks, those ineligible for antenatal steroids due to pre-pregnancy diabetes or
198
chorioamnionitis had the highest prevalence of all included outcomes (Table 3). Among
199
newborns delivering at 34 and 35 weeks, compared with those who were ineligible due to
200
prior receipt of antenatal steroids, those who were potentially eligible had a higher
AC C
194
ACCEPTED MANUSCRIPT 12 201
prevalence of respiratory complications.
202
among those delivering at 36 weeks. A similar pattern was observed for resuscitation at
203
delivery. Consistent patterns were less clear for glucose instability and NICU admission.
AC C
EP
TE D
M AN U
SC
RI PT
However, this difference was not apparent
ACCEPTED MANUSCRIPT 13 204
Comment Our study showed that antenatal steroid use for early preterm delivery is still
206
suboptimal. In examining late preterm deliveries we found that those potentially
207
ineligible based on pre-pregnancy diabetes or chorioamnionitis were at the highest risk
208
for respiratory complications regardless of gestational week at delivery. Further, while
209
the majority of late preterm deliveries among those potentially eligible for antenatal
210
steroids occurred at 36 weeks, the majority of respiratory complications in the steroid
211
eligible group occurred among those delivering at 34-35 weeks of gestation.
SC
RI PT
205
The ALPS (Antenatal Late Preterm Steroids) trial demonstrated statistically
213
significant benefits of antenatal steroids for late preterm delivery, but there remain a
214
number of issues surrounding implementation of the findings, which our study
215
illustrates.1
M AN U
212
First is how accurately preterm delivery can be predicted in the general population
217
in order to minimize steroid exposure for those who ultimately deliver at term.13 Our
218
findings highlight how difficult this is to achieve for early preterm deliveries as nearly
219
60% of singleton pregnancies receiving antenatal steroids in our study population
220
delivered at >34 weeks, and one third delivered at term. Conversely the opportunity for
221
appropriate use of steroids was missed in approximately 1 in 5 early preterm deliveries.
222
Similarly in the ALPS trial only 11% of pregnancies screened for the trial were eligible
223
and even then 16% delivered at term.1 Of the potentially late preterm steroid eligible
224
group, we do not know how many may receive steroids prior to preterm delivery nor how
225
many more pregnancies may be exposed to late preterm steroids and ultimately deliver at
226
> 37 weeks. In addition to those that ultimately deliver at term, there is the possibility of
AC C
EP
TE D
216
ACCEPTED MANUSCRIPT 14 “treatment creep” to include other groups such as a multiple pregnancies, pre-pregnancy
228
diabetics, and scheduled cesarean deliveries before 39 weeks of gestation.14 This could
229
further increase steroid exposure to the pregnant population. Without adherence to strict
230
criteria for determining risk for preterm delivery, extending antenatal steroid use to the
231
late preterm population could lead to a substantially larger population of newborns being
232
unnecessarily exposed to antenatal steroids.
RI PT
227
The second issue relates to which pregnancies are most likely to benefit from late
234
preterm steroids. In our study, the highest rate of newborn complications in late preterm
235
deliveries was in the group that had pre-pregnancy diabetes or chorioamnionitis and had
236
not received earlier steroids. While urgent delivery for overt chorioamnionitis likely
237
precludes receipt of late preterm steroids, the situation with pre-pregnancy diabetes is
238
more complicated. Pre-pregnancy diabetes is not considered an absolute contraindication
239
to antenatal steroids but the use of antenatal steroids can pose a challenge for glucose
240
control and may increase risks for exacerbation of neonatal hypoglycemia.15 However,
241
antenatal steroids, if effective, may significantly improve newborn respiratory outcomes
242
in this high-risk group.
EP
TE D
M AN U
SC
233
The third issue relates to the upper limit for eligibility for antenatal steroids. Fetal
244
respiratory maturation continues throughout the late prenatal period. More than half of
245
the potentially steroid eligible late preterm deliveries in our study were at 36 weeks,
246
when the number needed to treat is likely higher than the 35 reported in the ALPS trial.
247
Additionally, by 36 weeks of gestation newborn respiratory outcomes in our study were
248
similar in those pregnancies that were potentially eligible for antenatal steroids and those
249
that were ineligible, raising the question as to whether both or neither of these groups
AC C
243
ACCEPTED MANUSCRIPT 15 should receive antenatal steroids. Our findings suggest that restricting antenatal steroids
251
for late preterm pregnancies to those expected to deliver at 34-35 weeks could reduce the
252
target population for late preterm antenatal steroids by half (1.6% versus 4.0% of total
253
deliveries) while still capturing the majority (69%) of newborn respiratory complications
254
in this group.
255
Institute for Health and Clinical Excellence (NICE), to consider antenatal steroids up to
256
35+6 weeks of gestation.16
RI PT
250
SC
This finding supports the recommendation from the UK’s National
The last issue relates to the unknown risks of late preterm steroid exposure,
258
particularly long-term neurodevelopment. There are very limited data relating to long-
259
term follow-up after exposure to antenatal steroids after 34 weeks of gestation and there
260
is a growing body of literature highlighting the effects of antenatal steroids, some of
261
which have the potential for harm.7 Dalziel et al followed up 192 adults (87 exposed to
262
betamethasone in utero and 105 exposed to placebo) from a randomized controlled trial
263
and reported no obvious adverse health or neurodevelopmental outcomes following
264
preterm antenatal steroids.17 However, two other long-term follow-up studies of children
265
raised the possibility of adverse neurodevelopmental effects associated with exposure to
266
antenatal steroids later in pregnancy.18,19 An additional potential for harm relates to the
267
60% increase in neonatal hypoglycemia reported in the ALPS trial. While this may
268
appear to be a fairly minor short-term trade off for improved respiratory outcomes, it is
269
unclear if this side effect is benign in the long-term. Two recent studies have produced
270
conflicting results: one reporting transient mild neonatal hypoglycemia to be associated
271
with no increase in adverse neurodevelopmental outcomes at 2 years of age and another
272
reporting lower achievement test scores at 10 years. 20,21 The majority of those eligible
AC C
EP
TE D
M AN U
257
ACCEPTED MANUSCRIPT 16 for late preterm antenatal steroids delivered at 36 weeks of gestation, when their risk of
274
glucose instability was lowest. This raises questions regarding the increased risk of
275
hypoglycemia and the added burden on the healthcare system as well as potential long-
276
term outcomes.
RI PT
273
Strengths of our study include the large study population, the depth of clinical
278
information, and the ability to evaluate outcomes in the late preterm delivery population
279
prior to the SMFM and ACOG statements. Limitations of our study include that data on
280
chorioamnionitis were not available throughout the entire study period. Additionally, the
281
study did not explore whether late preterm deliveries were identified as at risk for preterm
282
birth prenatally nor whether there would have been adequate time to administer steroids
283
before birth. We thereby likely overestimated the potentially antenatal steroid eligible
284
group. Conversely, we likely underestimated the number of pregnancies potentially
285
eligible for late preterm antenatal steroids since it is probable that a portion of term
286
deliveries were considered at high risk for late preterm delivery. The timing of earlier
287
antenatal steroids was also not known nor was the number of earlier courses. Therefore,
288
some women may have received multiple doses or additional doses after 34 weeks,
289
although anecdotally adherence to a 34 week cutoff was the typical practice in our area
290
prior to 2016. Mode of delivery is also a risk factor for respiratory complications and the
291
Royal College of Obstetricians and Gynecologists in the UK recommends antenatal
292
steroids for planned cesarean delivery up to 38+6 weeks.22
293
stratification by mode of delivery (planned cesarean, cesarean in labor, and vaginal birth)
294
in addition to gestational week at delivery resulted in such small numbers in some cells,
295
that meaningful assessment of the influence of mode of delivery was precluded. Our
AC C
EP
TE D
M AN U
SC
277
However, in our study,
ACCEPTED MANUSCRIPT 17 study also had less strictly defined outcome measures compared to the ALPS trial, so
297
“glucose instability” was based on clinical documentation of this issue rather than
298
specific blood glucose levels and “respiratory complications” were not subdivided into
299
specific diagnoses. Finally, we did not evaluate the proportion of singleton pregnancies
300
with pre-labor ruptured membranes in the late preterm period for which use of late
301
preterm antenatal steroids remains controversial. While the 2016 SMFM and ACOG
302
statements about late preterm antenatal steroids did not specifically address such use in
303
the setting of pre-labor rupture of the membranes, this group made up 20% of the
304
population in the ALPS trial on which these statements were based.
M AN U
SC
RI PT
296
In recognition of the benefits of late preterm antenatal steroids professional
306
organizations have issued statements supporting the use of late preterm steroids in a
307
select group of pregnant women considered at high risk of delivery before 37 weeks of
308
gestation. However, the implementation raises questions including how to accurately
309
predict late preterm birth, who should receive late preterm antenatal steroids based on
310
pregnancy complications and gestational age, what the economic and health implications
311
may be, and the effect of antenatal steroids (either directly or indirectly through
312
hypoglycemia) on long-term outcomes.
EP
Given the difficulty in accurately predicting preterm delivery and the lack of
AC C
313
TE D
305
314
information about the long-term neurodevelopmental outcomes associated with late
315
preterm steroids, this intervention should be targeted to those most likely to gain benefit.
316
One of the limitations of the ALPS trial is that it evaluated outcomes for those who
317
received steroids throughout the late preterm period. However, the rate of respiratory
318
complications is substantially lower at 36 compared to 34 weeks of gestation.
ACCEPTED MANUSCRIPT 18 Consequently the estimated number needed to treat will change across the late preterm
320
period and is likely considerably higher at 36 weeks than it is at 34 weeks. In addition,
321
although the ALPS trial was not statistically powered to evaluate the effect of
322
betamethasone by gestational week, a subgroup analysis of those recruited at 36 weeks,
323
reported little to no difference in the rate of the primary outcome or of severe respiratory
324
outcomes in the placebo and betamethasone groups (primary outcome: placebo 7.1% vs
325
control 7.1%; severe respiratory outcomes (placebo 3.8% vs control 2.9%).
SC
RI PT
319
Our findings suggest that restricting antenatal steroids to eligible pregnancies at
327
34-35 weeks may capture most of the respiratory complications among late preterm
328
deliveries while minimizing antenatal steroid exposure at the population level. Careful
329
consideration is needed to ensure that guidelines target those pregnancies that could
330
benefit most from late preterm steroids while minimizing unnecessary exposure. Future
331
monitoring of the effect of late preterm antenatal steroids is essential to evaluate short-
332
and long-term outcomes of this change in practice.
AC C
EP
TE D
M AN U
326
ACCEPTED MANUSCRIPT 19 References
334
1 Gyamfi-Bannerman C, Thom EA, Blackwell SC, Tita ATN, Reddy UM, Saade GR,
335
Rouse DJ, McKenna DS, Clark EAS, Thorp JM Jr., Chien EK, Peaceman AM, Gibbs RS,
336
Swamy GK, Norton ME, Casey BM, Caritis SN, Tolosa JE, Sorokin Y, VanDorsten JP,
337
and Jain L, for the NICHD Maternal–Fetal Medicine Units Network* Antenatal
338
Betamethasone for Women at Risk for Late Preterm Delivery N Engl J Med
339
2016;374:1311-20.
340
2 Antenatal corticosteroid therapy for fetal maturation. Committee Opinion No. 677.
341
American College of Obstetricians and Gynecologists. Obstet Gynecol 2016;128:e187–
342
94.
343
3 Implementation of the use of antenatal corticosteroids in the late preterm birth period
344
in women at risk for preterm delivery American Journal of Obstetrics & Gynecology
345
2016; Volume 215:B13 - B15.
346
4 Martin JA, Hamilton BE, Osterman MJ, Curtin SC, Matthews TJ. Births: final data for
347
2013. Natl Vital Stat Rep 2015;64:1-65.
TE D
M AN U
SC
RI PT
333
5 Antenatal Corticosteroid Clinical Practice Guidelines Panel. Antenatal
349
corticosteroids given to women prior to birth to improve fetal, infant, child and adult
350
health: Clinical Practice Guidelines. 2015. Liggins Institute, The University of Auckland,
351
New Zealand.
352
6 Goldenberg and McClure Use of Antenatal Corticosteroid Prophylaxis. Obstetrics &
353
Gynecology 2015;125: 285-287.
354
7 Kemp, MW, Newnham JP, Challis JG, Jobe AH, and Stock SJ. The clinical use of
355
corticosteroids in pregnancy. Human Reproduction Update, 2016;22,:240–259.
AC C
EP
348
ACCEPTED MANUSCRIPT 20 356
8 Vidaeff AC, Belfort MA, Steer PJ. Antenatal corticosteroids: a time for more careful
357
scrutiny of the indications? BJOG 2016;123:1067-1069.
358
9 Ridout A, Watson H, Shennan A. Antenatal corticosteroids in perspective: rationalising current practice. BJOG 2016:123:1070.
360
10 Antenatal Betamethasone for Women at Risk for Late Preterm Delivery.
361
Redlich A, Bottger R, Costa SD. New England Journal of Medicine. 375:486, 2016 Aug
362
04. [Comment. Letter]
363
11 Antenatal Betamethasone for Women at Risk for Late Preterm Delivery.
364
Smith GC. N Engl J Med 375:486, 2016 Aug 04.
M AN U
SC
RI PT
359
[Comment. Letter]
366
12 Kamath-Rayne BD, Rozance PJ, Goldenberg RL, Jobe AH. Antenatal
367
corticosteroids beyond 34 weeks gestation: What do we do now? American Journal of
368
Obstetrics & Gynecology 2016;215 :423 – 430.
TE D
365
13 Profit J, Goldstein BA, Tamaresis J, Kan P, Lee HC. Regional Variation in
370
Antenatal Corticosteroid Use: A Network-Level Quality Improvement Study. Pediatrics
371
2015;135:2014-2177.
372
14 Stutchfield P, Whitaker R, Russell I. Antenatal betamethasone and incidence of
373
neonatal respiratory distress after elective caesarean section: pragmatic randomised trial.
374
BMJ 2005;331:662.
AC C
375
EP
369
15 National Institute for Health and Clinical Excellence. NICE
Clinical
376
Guideline 63: Diabetes in pregnancy. Management of diabetes and its complications from
377
pre-conception to the postnatal period. London: NICE; 2008.
378
16 National Institute for Health and Clinical Excellence. NICE Clinical Guideline
ACCEPTED MANUSCRIPT 21 (NG25) : Preterm Labor and Birth. London: NICE; 2015.
380
17 Dalziel SR, Lim VK, Lambert A, McCarthy D, Parag V, Rodgers A, Harding JE.
381
Antenatal Exposure To Betamethasone: Psychological Functioning And Health Related
382
Quality Of Life 31 Years After Inclusion In Randomised Controlled Trial. British
383
Medical Journal 2005;331:665-668.
384
18 Stutchfield PR, Whitaker R, Gliddon AE, Hobson L, Kotecha S, Doull IJ.
385
Behavioural, educational and respiratory outcomes of antenatal betamethasone for term
386
caesarean section (ASTECS trial). Arch Dis Child Fetal Neonatal Ed 2013;98:F195-200.
387
19 Asztalos et al. Association between gestational age at birth, antenatal corticosteroids,
388
and outcomes at 5 years: multiple courses of antenatal corticosteroids for preterm birth
389
study at 5 years of age (MACS-5). BMC Pregnancy and Childbirth 2014; 14:272.
390
20 McKinlay, Christopher JD, Jane M. Alsweiler, Judith M. Ansell, Nicola S. Anstice, J.
391
Geoffrey Chase, Gregory D. Gamble, Deborah L. Harris et al. Neonatal glycemia and
392
neurodevelopmental outcomes at 2 years. N Engl J Med 2015;373:1507-18.
393
21 Kaiser, J R, Bai, S, Gibson N, Holland G, Lin, T M, Swearingen, C J, El Hassan, NO.
394
Association between transient newborn hypoglycemia and fourth-grade achievement test
395
proficiency: a population-based study. JAMA pediatrics 2015;169:913-921.
396
22 Royal College of Obstetricians and Gynaecologists. Antenatal Corticosteroids to
397
Reduce Neonatal Morbidity and Mortality. Green-top Guideline No. 7 October 2010 .
398
23 Supplement to: Gyamfi-Bannerman C, Thom EA, Blackwell SC, et al. Antenatal
399
betamethasone for women at risk for late preterm delivery. N Engl J Med 2016;374:1311-
400
20.
401
AC C
EP
TE D
M AN U
SC
RI PT
379
ACCEPTED MANUSCRIPT 22 402
AC C
EP
TE D
M AN U
SC
RI PT
Table 1. Characteristics of those with late preterm deliveries by potential eligibility for antenatal steroids Ineligible Eligible (N=579) (N=2,363) Maternal demographics n % n % Maternal age at admission <20 24 4.1 107 4.5 20-34 373 64.4 1621 68.6 ≥35 182 31.4 635 26.9 a Maternal race/ethnicity 223 50.7 1023 50.2 White, NH African American, NH 23 5.2 77 3.8 Hispanic 73 16.6 389 19.1 Asian/Pacific Islander 81 18.4 384 18.9 b 7 1.6 53 2.6 AIAN Other 33 7.5 111 5.4 Private insurance No 370 63.9 1449 61.3 Yes 209 36.1 914 38.7 c <25 258 48.2 1148 51.9 Pre-pregnancy BMI 25-29 127 23.7 536 24.2 ≥30 150 28.0 529 23.9 Pregnancy characteristics Pre-pregnancy Htd No 538 92.9 2271 96.1 Yes 41 7.1 92 3.9 Pre-pregnancy diabetes No 476 82.2 2363 100.0 Yes 103 17.8 0 0.0 Nulliparous No 351 60.6 1330 56.3 Yes 228 39.4 1033 43.7 e No 515 88.9 2056 87.0 GDM Yes 64 11.1 307 13.0 f Pre-eclampsia/gestational Ht No 451 77.9 1964 83.1 Yes 128 22.1 399 16.9 Substance abuse No 572 98.8 2323 98.3 Yes 7 1.2 40 1.7 Nicotine use No 562 97.1 2296 97.2 Yes 17 2.9 67 2.8 Delivery characteristics Vaginal delivery No 322 55.6 805 34.1 Yes 257 44.4 1558 65.9 Cesarean with labor No 517 89.3 2194 92.8 Yes 62 10.7 169 7.2 Cesarean without labor No 319 55.1 1727 73.1 Yes 260 44.9 636 26.9 Gestational week on delivery 34 207 35.8 315 13.3
ACCEPTED MANUSCRIPT 23
Low birth weight (<2,500 g)
175 197
30.2 34.0
637 1411
27.0 59.7
Male Female No Yes
335 243 298 281
58.0 42.0 51.5 48.5
1293 1056 1614 749
55.0 45.0 68.3 31.7
a
RI PT
Newborn characteristics Gender
35 36
NH Non-Hispanic AIAN American Indian/Alaska Native c BMI Body mass index (kg/m2) Pre-pregnancy Htd – pre-pregnancy hypertension e GDM Gestational diabetes mellitus f Pre-eclampsia/gestational Ht – pre-eclampsia or gestational hypertension
AC C
EP
TE D
M AN U
SC
b
ACCEPTED MANUSCRIPT
24
14.9 (95)
0.22
11.6 (74)
0.48 <0.001
5.2 (33) 45.1 (287)
EP AC C
8.6 (15)
0.03
5.5 (77)
8.1 (16)
0.13
8.6 (15)
0.25
7.3 (103)
10.7 (21)
0.10
8.6 (15) 62.3 (109)
0.09 <0.001
3.7 (52) 20.1 (283)
5.1 (10) 33.0 (65)
0.34 <0.001
M AN U
0.02
TE D
Newborn 24.1 (76) 15.9 (33) respiratory complications Resuscitation at 16.5 (52) 12.6 (26) delivery Glucose instability 7.9 (25) 6.3 (13) NICU admission 87.0 (274) 95.2 (197)
SC
RI PT
Table 2. Newborn complications by gestational week on delivery and eligibility for antenatal steroids among late preterm deliveries Gestational week 34 35 36 on delivery Eligible Ineligible p Eligible Ineligible p Eligible Ineligible p (N=315) (N=207) (N=637) (N=175) (N=1411) (N=197) % (n) % (n) % (n) % (n) % (n) % (n)
ACCEPTED MANUSCRIPT
25
EP
Inelig-prior steroids = ineligible for late preterm steroids due to prior receipt of antenatal steroids in this pregnancy. Inelig-chorio/diab = ineligible due to pre-pregnancy diabetes or chorioamnionitis
b
AC C
a
TE D
M AN U
SC
RI PT
Table 3. Newborn complications by gestational week on delivery by eligibility and reason for ineligibility for antenatal steroids among late preterm deliveries Gestational week 34 weeks 35 weeks 36 weeks on delivery a b Eligible Eligible Inelig-prior IneligEligible Inelig-prior IneligIneligIneligsteroids chorio/diab (N=1411) (N=315) steroids chorio/diab prior chorio/diab (N=637) (N=149) (N=26) (N=144) (N=53) steroids (N=20) (N=187) % % % % % % % % % Newborn 24.1 13.9 35.0 14.9 7.4 15.4 5.5 4.9 17.0 respiratory complications Resuscitation at 16.5 11.8 20.0 11.6 8.7 7.7 7.3 6.3 22.6 delivery Glucose instability 7.9 4.8 20.0 5.2 6.0 23.1 3.7 0.7 17.0 NICU admission 87.0 96.3 85.0 45.1 61.7 65.4 20.1 27.8 47.2
ACCEPTED MANUSCRIPT
26
RI PT
Figure Legend
AC C
EP
TE D
M AN U
SC
Figure 1. Distribution of weeks at delivery among late preterm deliveries with respiratory complications by eligibility for antenatal steroid
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S0002-9378(17)30558-6
DOI:
10.1016/j.ajog.2017.04.029
Reference:
YMOB 11633
To appear in:
American Journal of Obstetrics and Gynecology
Received Date: 3 March 2017 Accepted Date: 17 April 2017
Please cite this article as: Souter V, Kauffman E, Marshall MAJ, Katon JG, Assessing the Potential Impact of Extending Antenatal Steroids to the Late Preterm Period, American Journal of Obstetrics and Gynecology (2017), doi: 10.1016/j.ajog.2017.04.029. 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 1 Assessing the Potential Impact of Extending Antenatal Steroids to the Late Preterm
2
Period
3
Vivienne SOUTER, MD
4
Foundation for Health Care Quality, Seattle, WA
5
Ellen KAUFFMAN, MD
6
Foundation for Health Care Quality, Seattle, WA
7
Ms Alice J. MARSHALL, BS, MFA
8
Foundation for Health Care Quality, Seattle, WA
9
Jodie G KATON, PhD, MS
M AN U
SC
RI PT
1
US Department of Veterans Affairs (VA), Health Services Research and Development
11
Center of Innovation for Veteran-Centered and Value Driven Care, VA Puget Sound
12
Healthcare System, Seattle, WA
13
Disclosure Statement: The authors report no conflict of interest.
14
Funding Source: The Foundation for Health Care Quality is a 501(c)(3) non-profit
15
organization supported by membership dues from the participants in its programs. There
16
was no other funding for this study.
17
Paper presentation information: This work was presented at The Pregnancy Meeting,
18
The 37th Annual Meeting of the SMFM, Las Vegas, Jan 23-28, 2017.
19
Disclaimer for Dr. Katon: The views expressed in this article are those of the authors
20
and do not necessarily reflect the position or policy of the Department of Veterans
21
Affairs.
AC C
EP
TE D
10
ACCEPTED MANUSCRIPT 2 Corresponding Author: Vivienne SOUTER, MD
23
Institution: Obstetrics Clinical Outcomes Assessment Program, Foundation for Health
24
Care Quality, 705 Second Avenue, Suite 410, Seattle, WA 98104
25
Telephone: (206) 375 0234 (cell)
26
Email: [email protected]
28
Word Count: Abstract – 377. Main text – 2974.
30
Table for Print Copy: Table 2
31
M AN U
29
SC
27
RI PT
22
32
Condensation:
33
steroids, most newborn respiratory complications occur in those delivering before 36
34
weeks.
38
Potential impact of late preterm antenatal steroid guidelines
EP
37
Short Title:
AC C
36
TE D
35
Among late preterm pregnancies potentially eligible for antenatal
ACCEPTED MANUSCRIPT 3 Abstract:
40
Background: In 2016 guidance statements were issued by the Society for Maternal Fetal
41
Medicine and by the American College of Obstetricians and Gynecologists about
42
extending antenatal steroid use to selected late preterm singleton pregnancies.
43
Objective: To review antenatal steroid use prior to the 2016 guidance statements and
44
assess the potential impact of these.
45
Study Design This cohort study used chart-abstracted data from singleton deliveries from
46
January 1, 2012 to March 31, 2016 at 12 centers participating in the Obstetrics Clinical
47
Outcomes Assessment Program, a quality initiative in Washington State. Pregnancies
48
with missing gestation at delivery, fetal anomalies, or antepartum demise were excluded.
49
Antenatal steroid use prior to the 2016 guidance was evaluated based on the percentage
50
of early preterm deliveries (23+0-33+6 weeks) and the percentage of all pregnancies that
51
received antenatal steroids. Newborn complication rates were calculated for late preterm
52
deliveries (34+0+0-36+6 weeks), grouped by whether they would be potentially eligible or
53
ineligible for antenatal steroids based on the 2016 guidance statements.
54
Results The opportunity for antenatal steroids was missed in 21.8% (226/1034) of early
55
preterm deliveries and of all those who received antenatal steroids, 32.2% (614/1908)
56
delivered at term. Seventy-four percent of preterm deliveries (n=2942) were in the late
57
preterm period. Eighty percent (n=2363) of late preterm deliveries were potentially
58
eligible for antenatal steroids and 60% of these (n=1411) delivered at 36 weeks. The rate
59
of respiratory complications in newborns delivering at 34 and 35 weeks was higher in the
60
group potentially eligible for late preterm antenatal steroids compared to those in the
61
ineligible group. Of those delivering at 36 weeks, no differences were detected in
AC C
EP
TE D
M AN U
SC
RI PT
39
ACCEPTED MANUSCRIPT 4 prevalence of respiratory complications by potential eligibility for antenatal steroids;
63
however, compared with the ineligible group, those potentially eligible had a lower risk
64
of neonatal intensive care unit admission (p<0.001). More than 2/3 (69%; 171/248) of
65
newborn respiratory complications among late preterm deliveries potentially eligible for
66
antenatal steroids occurred in those delivering at 34-35 weeks. The highest rate of
67
respiratory complications was in those ineligible for antenatal steroids due to pre-
68
pregnancy diabetes or chorioamnionitis, regardless of gestational age at delivery.
69
Conclusions Careful consideration of which pregnancies should receive late preterm
70
antenatal steroids and how to identify these pregnancies is important to optimize benefits
71
and mitigate potential risks of this intervention.
M AN U
SC
RI PT
62
AC C
EP
TE D
72 73 74
ACCEPTED MANUSCRIPT 5 75
Introduction: In 2016, a randomized controlled trial reported that a course of late preterm
77
betamethasone (34+0-36+6 weeks) given to women with singleton pregnancies considered
78
at risk for delivery before 37 weeks, was associated with a significant decrease in the risk
79
of newborn respiratory complications.1 These results prompted statements from the
80
American College of Obstetrics and Gynecology (ACOG) and the Society for Maternal
81
Fetal Medicine (SMFM) in April 2016 and two subsequent guidance documents
82
published later in 2016 supporting antenatal steroids for selected women with late
83
preterm (34+0 -36+6 weeks) singleton pregnancies considered to be at risk for preterm
84
delivery.2,3
85
with chorioamnionitis, and those with pre-pregnancy diabetes.
M AN U
SC
RI PT
76
Excluded from these recommendations were multiple pregnancies, those
Nationally, approximately 70% of preterm births and 8% of all deliveries occur at
87
34-36 weeks.4 Therefore, the new guidance statements have the potential to expose a
88
much larger percentage of the pregnant population to antenatal steroids. For antenatal
89
steroid exposure at <34 weeks of gestation, there are clear benefits for those born
90
prematurely and no obvious adverse effects on long-term follow up after a single course
91
of antenatal steroids.5 However, in the last two years a number of researchers have raised
92
concerns about the potential for negative as well as positive effects of antenatal steroids
93
in general
94
particular there is a paucity of data on long-term outcomes following antenatal steroid
95
exposure at >34 weeks.12 Given the large number of pregnancies potentially affected by
96
this change in practice, the short- and long-term risk benefit ratio is an important
97
consideration. Our goal was to review antenatal steroid use prior to the 2016 ACOG and
AC C
EP
TE D
86
6-9
and specifically about the risk-benefit ratio for late preterm steroids.10-12 In
ACCEPTED MANUSCRIPT 6 98
SMFM statements and to assess the potential impact of implementing the 2016
99
recommendations.
100
AC C
EP
TE D
M AN U
SC
RI PT
101
ACCEPTED MANUSCRIPT 7 102
Study Design This retrospective cohort study used chart-abstracted data from deliveries at 12
104
centers participating in the Obstetrics Clinical Outcomes Assessment Program (OB
105
COAP), a quality initiative in Washington State. These centers all had Level II or III
106
neonatal units and included both urban and rural settings. Centers contributing data for all
107
or part of the study period (January 1, 2012-March 31, 2016) were included in the study.
108
The analysis was restricted to singleton pregnancies delivering between 23+0 and 42+6
109
weeks of gestation. Deliveries with missing gestational age at delivery, fetal anomalies,
110
or antepartum demise were excluded.
M AN U
SC
RI PT
103
OB COAP captures consecutive deliveries at participating centers and records a
112
wide range of variables including those related to maternal demographic characteristics,
113
pre-pregnancy health, pregnancy complications, labor course, delivery, and postnatal
114
outcomes for mothers and newborns. Abstracted data are entered into a cloud-based,
115
standardized data tool. Data undergo real-time quality checks performed both at the site
116
and aggregate level. Monthly web meetings and unlimited access to OB COAP staff for
117
education and support are available. Audit of OB COAP data against billing records with
118
a minimum of 90% agreement is required for the program.
EP
TE D
111
The Western Institutional Review Board determined in 2015 that OB COAP is not
120
engaged in human subjects research and therefore is exempt from institutional review
121
board review.
122
AC C
119
Early preterm birth was defined as delivery between 23+0 and 33+6 weeks and late
123
preterm birth as delivery between 34+0 and 36+6 weeks of gestation. Pregnancies that
124
delivered in the late preterm period were divided into two groups based on the 2016
125
ACOG / SMFM guidance: those that would be considered potentially eligible and those
ACCEPTED MANUSCRIPT 8 that would be considered ineligible for late preterm antenatal steroids. Those considered
127
to be ineligible for antenatal steroids were further divided into those that would be
128
considered ineligible due to having already received antenatal steroids and those that had
129
not received earlier steroids but were ineligible due to chorioamnionitis or pre-pregnancy
130
diabetes. Data on maternal antibiotic treatment for chorioamnionitis were available only
131
for 2015 and 2016.
RI PT
126
Newborn outcomes evaluated included respiratory complications, need for
133
resuscitation at delivery, glucose instability, and admission to the neonatal intensive care
134
unit (NICU). Other descriptive variables included maternal demographics and pregnancy,
135
delivery, and newborn characteristics.
M AN U
SC
132
Descriptive statistics included the rate of preterm delivery and late preterm
137
delivery, the percentage of all deliveries that received antenatal steroids, and the
138
percentage that received antenatal steroids and subsequently delivered late preterm or at
139
term. We also report the percentage of early preterm deliveries where the opportunity for
140
antenatal steroids was missed (those delivering before 34 weeks of gestation that did not
141
receive antenatal steroids).
EP
TE D
136
Maternal demographic, pregnancy, delivery, and newborn characteristics are
143
reported for late preterm deliveries by potential eligibility for antenatal steroids. Newborn
144
outcomes among late preterm deliveries are reported by potential eligibility for antenatal
145
steroids (eligible versus ineligible) and gestational week at delivery (34, 35, 36 weeks).
146
Frequency of outcomes between late preterm deliveries potentially eligible or ineligible
147
for antenatal steroids was compared using chi-square tests. As deliveries ineligible for
148
antenatal steroids represented a heterogeneous group in terms of obstetric risk factors, we
AC C
142
ACCEPTED MANUSCRIPT 9 also qualitatively compared newborn outcomes between potential eligibility and
150
ineligibility for the late preterm antenatal steroids, subdividing the ineligible group into
151
those who were ineligible due to prior steroids and those who were ineligible due to
152
chorioamnionitis or pre-pregnancy diabetes.
153 154
AC C
EP
TE D
M AN U
SC
155
RI PT
149
ACCEPTED MANUSCRIPT 10 156
Results A total of 62300 singleton deliveries from 12 centers were entered into the
158
database during the study period. Of these, 60190 deliveries (97%) met the inclusion
159
criteria. The prevalence of preterm delivery (23+0-36+6 weeks) was 6.6% (n=3976).
160
Seventy-four percent of preterm deliveries were in the late preterm period (n=2942),
161
representing 4.9% of all singleton deliveries.
RI PT
157
Three percent (n=1908) of all pregnancies received antenatal steroids at some
163
point. Of the 1908 pregnancies exposed to antenatal steroids, 42.3% (n=808) delivered in
164
the early preterm period, 25.5% (n=486) in the late preterm period, and 32.2% (n=614) at
165
term. Among the 1034 early preterm deliveries the opportunity for antenatal steroids was
166
missed in 21.9% (n=226).
M AN U
SC
162
Based on the 2016 guidance, 80.3% (n=2363) of late preterm deliveries would be
168
considered eligible for antenatal steroids if late preterm birth was anticipated with enough
169
time to administer steroids. Of the late preterm population, 16.3% (480/2942) were
170
ineligible due to having already received antenatal steroids and 3.4% (99/2942) were
171
ineligible due to pre-pregnancy diabetes or chorioamnionitis, Those potentially eligible
172
for antenatal steroids tended to be younger, were less likely to be African American Non
173
Hispanic, had a lower prevalence of pre-pregnancy hypertension, and had a lower
174
prevalence of pre-eclampsia/gestational hypertension, compared with those who were
175
ineligible (Table 1).
176
potentially eligible for antenatal steroids were more likely to have a vaginal delivery and
177
to deliver at 36 weeks of gestation.
AC C
EP
TE D
167
Additionally, compared with those who were ineligible, those
ACCEPTED MANUSCRIPT 11 Newborn respiratory complications were recorded in 10.6% (312/2942) of all late
179
preterm deliveries in the study population and 79.5% (248/312) of these occurred in
180
pregnancies potentially eligible for antenatal steroids. Examining gestational week at
181
delivery among newborns with respiratory complications by eligibility for antenatal
182
steroids revealed that more than two thirds (69.0%; 171/248) of newborn respiratory
183
complications in those potentially eligible for antenatal steroids occurred among those
184
delivering at 34-35 weeks (Figure 1).
SC
RI PT
178
Of pregnancies delivering at 34 and 35 weeks of gestation, those potentially
186
eligible for antenatal steroids had a higher rate of newborn respiratory complications than
187
those who were ineligible (p< 0.05) (Table 2). For deliveries at 36 weeks, there was no
188
statistically significant difference in newborn respiratory complications between those
189
potentially eligible for antenatal steroids and those who were ineligible. No differences in
190
resuscitation at delivery or glucose instability were detected by potential eligibility for
191
antenatal steroids. Compared to those ineligible for antenatal steroids, those who were
192
potentially eligible had a lower rate of NICU admission regardless of gestational week at
193
delivery (Table 2).
EP
TE D
M AN U
185
Subdividing those ineligible for steroids by reason for ineligibility (prior receipt
195
of steroids, pre-pregnancy diabetes or chorioamnionitis) and comparing them to those
196
who were potentially eligible, slightly different patterns emerged. At nearly all
197
gestational weeks, those ineligible for antenatal steroids due to pre-pregnancy diabetes or
198
chorioamnionitis had the highest prevalence of all included outcomes (Table 3). Among
199
newborns delivering at 34 and 35 weeks, compared with those who were ineligible due to
200
prior receipt of antenatal steroids, those who were potentially eligible had a higher
AC C
194
ACCEPTED MANUSCRIPT 12 201
prevalence of respiratory complications.
202
among those delivering at 36 weeks. A similar pattern was observed for resuscitation at
203
delivery. Consistent patterns were less clear for glucose instability and NICU admission.
AC C
EP
TE D
M AN U
SC
RI PT
However, this difference was not apparent
ACCEPTED MANUSCRIPT 13 204
Comment Our study showed that antenatal steroid use for early preterm delivery is still
206
suboptimal. In examining late preterm deliveries we found that those potentially
207
ineligible based on pre-pregnancy diabetes or chorioamnionitis were at the highest risk
208
for respiratory complications regardless of gestational week at delivery. Further, while
209
the majority of late preterm deliveries among those potentially eligible for antenatal
210
steroids occurred at 36 weeks, the majority of respiratory complications in the steroid
211
eligible group occurred among those delivering at 34-35 weeks of gestation.
SC
RI PT
205
The ALPS (Antenatal Late Preterm Steroids) trial demonstrated statistically
213
significant benefits of antenatal steroids for late preterm delivery, but there remain a
214
number of issues surrounding implementation of the findings, which our study
215
illustrates.1
M AN U
212
First is how accurately preterm delivery can be predicted in the general population
217
in order to minimize steroid exposure for those who ultimately deliver at term.13 Our
218
findings highlight how difficult this is to achieve for early preterm deliveries as nearly
219
60% of singleton pregnancies receiving antenatal steroids in our study population
220
delivered at >34 weeks, and one third delivered at term. Conversely the opportunity for
221
appropriate use of steroids was missed in approximately 1 in 5 early preterm deliveries.
222
Similarly in the ALPS trial only 11% of pregnancies screened for the trial were eligible
223
and even then 16% delivered at term.1 Of the potentially late preterm steroid eligible
224
group, we do not know how many may receive steroids prior to preterm delivery nor how
225
many more pregnancies may be exposed to late preterm steroids and ultimately deliver at
226
> 37 weeks. In addition to those that ultimately deliver at term, there is the possibility of
AC C
EP
TE D
216
ACCEPTED MANUSCRIPT 14 “treatment creep” to include other groups such as a multiple pregnancies, pre-pregnancy
228
diabetics, and scheduled cesarean deliveries before 39 weeks of gestation.14 This could
229
further increase steroid exposure to the pregnant population. Without adherence to strict
230
criteria for determining risk for preterm delivery, extending antenatal steroid use to the
231
late preterm population could lead to a substantially larger population of newborns being
232
unnecessarily exposed to antenatal steroids.
RI PT
227
The second issue relates to which pregnancies are most likely to benefit from late
234
preterm steroids. In our study, the highest rate of newborn complications in late preterm
235
deliveries was in the group that had pre-pregnancy diabetes or chorioamnionitis and had
236
not received earlier steroids. While urgent delivery for overt chorioamnionitis likely
237
precludes receipt of late preterm steroids, the situation with pre-pregnancy diabetes is
238
more complicated. Pre-pregnancy diabetes is not considered an absolute contraindication
239
to antenatal steroids but the use of antenatal steroids can pose a challenge for glucose
240
control and may increase risks for exacerbation of neonatal hypoglycemia.15 However,
241
antenatal steroids, if effective, may significantly improve newborn respiratory outcomes
242
in this high-risk group.
EP
TE D
M AN U
SC
233
The third issue relates to the upper limit for eligibility for antenatal steroids. Fetal
244
respiratory maturation continues throughout the late prenatal period. More than half of
245
the potentially steroid eligible late preterm deliveries in our study were at 36 weeks,
246
when the number needed to treat is likely higher than the 35 reported in the ALPS trial.
247
Additionally, by 36 weeks of gestation newborn respiratory outcomes in our study were
248
similar in those pregnancies that were potentially eligible for antenatal steroids and those
249
that were ineligible, raising the question as to whether both or neither of these groups
AC C
243
ACCEPTED MANUSCRIPT 15 should receive antenatal steroids. Our findings suggest that restricting antenatal steroids
251
for late preterm pregnancies to those expected to deliver at 34-35 weeks could reduce the
252
target population for late preterm antenatal steroids by half (1.6% versus 4.0% of total
253
deliveries) while still capturing the majority (69%) of newborn respiratory complications
254
in this group.
255
Institute for Health and Clinical Excellence (NICE), to consider antenatal steroids up to
256
35+6 weeks of gestation.16
RI PT
250
SC
This finding supports the recommendation from the UK’s National
The last issue relates to the unknown risks of late preterm steroid exposure,
258
particularly long-term neurodevelopment. There are very limited data relating to long-
259
term follow-up after exposure to antenatal steroids after 34 weeks of gestation and there
260
is a growing body of literature highlighting the effects of antenatal steroids, some of
261
which have the potential for harm.7 Dalziel et al followed up 192 adults (87 exposed to
262
betamethasone in utero and 105 exposed to placebo) from a randomized controlled trial
263
and reported no obvious adverse health or neurodevelopmental outcomes following
264
preterm antenatal steroids.17 However, two other long-term follow-up studies of children
265
raised the possibility of adverse neurodevelopmental effects associated with exposure to
266
antenatal steroids later in pregnancy.18,19 An additional potential for harm relates to the
267
60% increase in neonatal hypoglycemia reported in the ALPS trial. While this may
268
appear to be a fairly minor short-term trade off for improved respiratory outcomes, it is
269
unclear if this side effect is benign in the long-term. Two recent studies have produced
270
conflicting results: one reporting transient mild neonatal hypoglycemia to be associated
271
with no increase in adverse neurodevelopmental outcomes at 2 years of age and another
272
reporting lower achievement test scores at 10 years. 20,21 The majority of those eligible
AC C
EP
TE D
M AN U
257
ACCEPTED MANUSCRIPT 16 for late preterm antenatal steroids delivered at 36 weeks of gestation, when their risk of
274
glucose instability was lowest. This raises questions regarding the increased risk of
275
hypoglycemia and the added burden on the healthcare system as well as potential long-
276
term outcomes.
RI PT
273
Strengths of our study include the large study population, the depth of clinical
278
information, and the ability to evaluate outcomes in the late preterm delivery population
279
prior to the SMFM and ACOG statements. Limitations of our study include that data on
280
chorioamnionitis were not available throughout the entire study period. Additionally, the
281
study did not explore whether late preterm deliveries were identified as at risk for preterm
282
birth prenatally nor whether there would have been adequate time to administer steroids
283
before birth. We thereby likely overestimated the potentially antenatal steroid eligible
284
group. Conversely, we likely underestimated the number of pregnancies potentially
285
eligible for late preterm antenatal steroids since it is probable that a portion of term
286
deliveries were considered at high risk for late preterm delivery. The timing of earlier
287
antenatal steroids was also not known nor was the number of earlier courses. Therefore,
288
some women may have received multiple doses or additional doses after 34 weeks,
289
although anecdotally adherence to a 34 week cutoff was the typical practice in our area
290
prior to 2016. Mode of delivery is also a risk factor for respiratory complications and the
291
Royal College of Obstetricians and Gynecologists in the UK recommends antenatal
292
steroids for planned cesarean delivery up to 38+6 weeks.22
293
stratification by mode of delivery (planned cesarean, cesarean in labor, and vaginal birth)
294
in addition to gestational week at delivery resulted in such small numbers in some cells,
295
that meaningful assessment of the influence of mode of delivery was precluded. Our
AC C
EP
TE D
M AN U
SC
277
However, in our study,
ACCEPTED MANUSCRIPT 17 study also had less strictly defined outcome measures compared to the ALPS trial, so
297
“glucose instability” was based on clinical documentation of this issue rather than
298
specific blood glucose levels and “respiratory complications” were not subdivided into
299
specific diagnoses. Finally, we did not evaluate the proportion of singleton pregnancies
300
with pre-labor ruptured membranes in the late preterm period for which use of late
301
preterm antenatal steroids remains controversial. While the 2016 SMFM and ACOG
302
statements about late preterm antenatal steroids did not specifically address such use in
303
the setting of pre-labor rupture of the membranes, this group made up 20% of the
304
population in the ALPS trial on which these statements were based.
M AN U
SC
RI PT
296
In recognition of the benefits of late preterm antenatal steroids professional
306
organizations have issued statements supporting the use of late preterm steroids in a
307
select group of pregnant women considered at high risk of delivery before 37 weeks of
308
gestation. However, the implementation raises questions including how to accurately
309
predict late preterm birth, who should receive late preterm antenatal steroids based on
310
pregnancy complications and gestational age, what the economic and health implications
311
may be, and the effect of antenatal steroids (either directly or indirectly through
312
hypoglycemia) on long-term outcomes.
EP
Given the difficulty in accurately predicting preterm delivery and the lack of
AC C
313
TE D
305
314
information about the long-term neurodevelopmental outcomes associated with late
315
preterm steroids, this intervention should be targeted to those most likely to gain benefit.
316
One of the limitations of the ALPS trial is that it evaluated outcomes for those who
317
received steroids throughout the late preterm period. However, the rate of respiratory
318
complications is substantially lower at 36 compared to 34 weeks of gestation.
ACCEPTED MANUSCRIPT 18 Consequently the estimated number needed to treat will change across the late preterm
320
period and is likely considerably higher at 36 weeks than it is at 34 weeks. In addition,
321
although the ALPS trial was not statistically powered to evaluate the effect of
322
betamethasone by gestational week, a subgroup analysis of those recruited at 36 weeks,
323
reported little to no difference in the rate of the primary outcome or of severe respiratory
324
outcomes in the placebo and betamethasone groups (primary outcome: placebo 7.1% vs
325
control 7.1%; severe respiratory outcomes (placebo 3.8% vs control 2.9%).
SC
RI PT
319
Our findings suggest that restricting antenatal steroids to eligible pregnancies at
327
34-35 weeks may capture most of the respiratory complications among late preterm
328
deliveries while minimizing antenatal steroid exposure at the population level. Careful
329
consideration is needed to ensure that guidelines target those pregnancies that could
330
benefit most from late preterm steroids while minimizing unnecessary exposure. Future
331
monitoring of the effect of late preterm antenatal steroids is essential to evaluate short-
332
and long-term outcomes of this change in practice.
AC C
EP
TE D
M AN U
326
ACCEPTED MANUSCRIPT 19 References
334
1 Gyamfi-Bannerman C, Thom EA, Blackwell SC, Tita ATN, Reddy UM, Saade GR,
335
Rouse DJ, McKenna DS, Clark EAS, Thorp JM Jr., Chien EK, Peaceman AM, Gibbs RS,
336
Swamy GK, Norton ME, Casey BM, Caritis SN, Tolosa JE, Sorokin Y, VanDorsten JP,
337
and Jain L, for the NICHD Maternal–Fetal Medicine Units Network* Antenatal
338
Betamethasone for Women at Risk for Late Preterm Delivery N Engl J Med
339
2016;374:1311-20.
340
2 Antenatal corticosteroid therapy for fetal maturation. Committee Opinion No. 677.
341
American College of Obstetricians and Gynecologists. Obstet Gynecol 2016;128:e187–
342
94.
343
3 Implementation of the use of antenatal corticosteroids in the late preterm birth period
344
in women at risk for preterm delivery American Journal of Obstetrics & Gynecology
345
2016; Volume 215:B13 - B15.
346
4 Martin JA, Hamilton BE, Osterman MJ, Curtin SC, Matthews TJ. Births: final data for
347
2013. Natl Vital Stat Rep 2015;64:1-65.
TE D
M AN U
SC
RI PT
333
5 Antenatal Corticosteroid Clinical Practice Guidelines Panel. Antenatal
349
corticosteroids given to women prior to birth to improve fetal, infant, child and adult
350
health: Clinical Practice Guidelines. 2015. Liggins Institute, The University of Auckland,
351
New Zealand.
352
6 Goldenberg and McClure Use of Antenatal Corticosteroid Prophylaxis. Obstetrics &
353
Gynecology 2015;125: 285-287.
354
7 Kemp, MW, Newnham JP, Challis JG, Jobe AH, and Stock SJ. The clinical use of
355
corticosteroids in pregnancy. Human Reproduction Update, 2016;22,:240–259.
AC C
EP
348
ACCEPTED MANUSCRIPT 20 356
8 Vidaeff AC, Belfort MA, Steer PJ. Antenatal corticosteroids: a time for more careful
357
scrutiny of the indications? BJOG 2016;123:1067-1069.
358
9 Ridout A, Watson H, Shennan A. Antenatal corticosteroids in perspective: rationalising current practice. BJOG 2016:123:1070.
360
10 Antenatal Betamethasone for Women at Risk for Late Preterm Delivery.
361
Redlich A, Bottger R, Costa SD. New England Journal of Medicine. 375:486, 2016 Aug
362
04. [Comment. Letter]
363
11 Antenatal Betamethasone for Women at Risk for Late Preterm Delivery.
364
Smith GC. N Engl J Med 375:486, 2016 Aug 04.
M AN U
SC
RI PT
359
[Comment. Letter]
366
12 Kamath-Rayne BD, Rozance PJ, Goldenberg RL, Jobe AH. Antenatal
367
corticosteroids beyond 34 weeks gestation: What do we do now? American Journal of
368
Obstetrics & Gynecology 2016;215 :423 – 430.
TE D
365
13 Profit J, Goldstein BA, Tamaresis J, Kan P, Lee HC. Regional Variation in
370
Antenatal Corticosteroid Use: A Network-Level Quality Improvement Study. Pediatrics
371
2015;135:2014-2177.
372
14 Stutchfield P, Whitaker R, Russell I. Antenatal betamethasone and incidence of
373
neonatal respiratory distress after elective caesarean section: pragmatic randomised trial.
374
BMJ 2005;331:662.
AC C
375
EP
369
15 National Institute for Health and Clinical Excellence. NICE
Clinical
376
Guideline 63: Diabetes in pregnancy. Management of diabetes and its complications from
377
pre-conception to the postnatal period. London: NICE; 2008.
378
16 National Institute for Health and Clinical Excellence. NICE Clinical Guideline
ACCEPTED MANUSCRIPT 21 (NG25) : Preterm Labor and Birth. London: NICE; 2015.
380
17 Dalziel SR, Lim VK, Lambert A, McCarthy D, Parag V, Rodgers A, Harding JE.
381
Antenatal Exposure To Betamethasone: Psychological Functioning And Health Related
382
Quality Of Life 31 Years After Inclusion In Randomised Controlled Trial. British
383
Medical Journal 2005;331:665-668.
384
18 Stutchfield PR, Whitaker R, Gliddon AE, Hobson L, Kotecha S, Doull IJ.
385
Behavioural, educational and respiratory outcomes of antenatal betamethasone for term
386
caesarean section (ASTECS trial). Arch Dis Child Fetal Neonatal Ed 2013;98:F195-200.
387
19 Asztalos et al. Association between gestational age at birth, antenatal corticosteroids,
388
and outcomes at 5 years: multiple courses of antenatal corticosteroids for preterm birth
389
study at 5 years of age (MACS-5). BMC Pregnancy and Childbirth 2014; 14:272.
390
20 McKinlay, Christopher JD, Jane M. Alsweiler, Judith M. Ansell, Nicola S. Anstice, J.
391
Geoffrey Chase, Gregory D. Gamble, Deborah L. Harris et al. Neonatal glycemia and
392
neurodevelopmental outcomes at 2 years. N Engl J Med 2015;373:1507-18.
393
21 Kaiser, J R, Bai, S, Gibson N, Holland G, Lin, T M, Swearingen, C J, El Hassan, NO.
394
Association between transient newborn hypoglycemia and fourth-grade achievement test
395
proficiency: a population-based study. JAMA pediatrics 2015;169:913-921.
396
22 Royal College of Obstetricians and Gynaecologists. Antenatal Corticosteroids to
397
Reduce Neonatal Morbidity and Mortality. Green-top Guideline No. 7 October 2010 .
398
23 Supplement to: Gyamfi-Bannerman C, Thom EA, Blackwell SC, et al. Antenatal
399
betamethasone for women at risk for late preterm delivery. N Engl J Med 2016;374:1311-
400
20.
401
AC C
EP
TE D
M AN U
SC
RI PT
379
ACCEPTED MANUSCRIPT 22 402
AC C
EP
TE D
M AN U
SC
RI PT
Table 1. Characteristics of those with late preterm deliveries by potential eligibility for antenatal steroids Ineligible Eligible (N=579) (N=2,363) Maternal demographics n % n % Maternal age at admission <20 24 4.1 107 4.5 20-34 373 64.4 1621 68.6 ≥35 182 31.4 635 26.9 a Maternal race/ethnicity 223 50.7 1023 50.2 White, NH African American, NH 23 5.2 77 3.8 Hispanic 73 16.6 389 19.1 Asian/Pacific Islander 81 18.4 384 18.9 b 7 1.6 53 2.6 AIAN Other 33 7.5 111 5.4 Private insurance No 370 63.9 1449 61.3 Yes 209 36.1 914 38.7 c <25 258 48.2 1148 51.9 Pre-pregnancy BMI 25-29 127 23.7 536 24.2 ≥30 150 28.0 529 23.9 Pregnancy characteristics Pre-pregnancy Htd No 538 92.9 2271 96.1 Yes 41 7.1 92 3.9 Pre-pregnancy diabetes No 476 82.2 2363 100.0 Yes 103 17.8 0 0.0 Nulliparous No 351 60.6 1330 56.3 Yes 228 39.4 1033 43.7 e No 515 88.9 2056 87.0 GDM Yes 64 11.1 307 13.0 f Pre-eclampsia/gestational Ht No 451 77.9 1964 83.1 Yes 128 22.1 399 16.9 Substance abuse No 572 98.8 2323 98.3 Yes 7 1.2 40 1.7 Nicotine use No 562 97.1 2296 97.2 Yes 17 2.9 67 2.8 Delivery characteristics Vaginal delivery No 322 55.6 805 34.1 Yes 257 44.4 1558 65.9 Cesarean with labor No 517 89.3 2194 92.8 Yes 62 10.7 169 7.2 Cesarean without labor No 319 55.1 1727 73.1 Yes 260 44.9 636 26.9 Gestational week on delivery 34 207 35.8 315 13.3
ACCEPTED MANUSCRIPT 23
Low birth weight (<2,500 g)
175 197
30.2 34.0
637 1411
27.0 59.7
Male Female No Yes
335 243 298 281
58.0 42.0 51.5 48.5
1293 1056 1614 749
55.0 45.0 68.3 31.7
a
RI PT
Newborn characteristics Gender
35 36
NH Non-Hispanic AIAN American Indian/Alaska Native c BMI Body mass index (kg/m2) Pre-pregnancy Htd – pre-pregnancy hypertension e GDM Gestational diabetes mellitus f Pre-eclampsia/gestational Ht – pre-eclampsia or gestational hypertension
AC C
EP
TE D
M AN U
SC
b
ACCEPTED MANUSCRIPT
24
14.9 (95)
0.22
11.6 (74)
0.48 <0.001
5.2 (33) 45.1 (287)
EP AC C
8.6 (15)
0.03
5.5 (77)
8.1 (16)
0.13
8.6 (15)
0.25
7.3 (103)
10.7 (21)
0.10
8.6 (15) 62.3 (109)
0.09 <0.001
3.7 (52) 20.1 (283)
5.1 (10) 33.0 (65)
0.34 <0.001
M AN U
0.02
TE D
Newborn 24.1 (76) 15.9 (33) respiratory complications Resuscitation at 16.5 (52) 12.6 (26) delivery Glucose instability 7.9 (25) 6.3 (13) NICU admission 87.0 (274) 95.2 (197)
SC
RI PT
Table 2. Newborn complications by gestational week on delivery and eligibility for antenatal steroids among late preterm deliveries Gestational week 34 35 36 on delivery Eligible Ineligible p Eligible Ineligible p Eligible Ineligible p (N=315) (N=207) (N=637) (N=175) (N=1411) (N=197) % (n) % (n) % (n) % (n) % (n) % (n)
ACCEPTED MANUSCRIPT
25
EP
Inelig-prior steroids = ineligible for late preterm steroids due to prior receipt of antenatal steroids in this pregnancy. Inelig-chorio/diab = ineligible due to pre-pregnancy diabetes or chorioamnionitis
b
AC C
a
TE D
M AN U
SC
RI PT
Table 3. Newborn complications by gestational week on delivery by eligibility and reason for ineligibility for antenatal steroids among late preterm deliveries Gestational week 34 weeks 35 weeks 36 weeks on delivery a b Eligible Eligible Inelig-prior IneligEligible Inelig-prior IneligIneligIneligsteroids chorio/diab (N=1411) (N=315) steroids chorio/diab prior chorio/diab (N=637) (N=149) (N=26) (N=144) (N=53) steroids (N=20) (N=187) % % % % % % % % % Newborn 24.1 13.9 35.0 14.9 7.4 15.4 5.5 4.9 17.0 respiratory complications Resuscitation at 16.5 11.8 20.0 11.6 8.7 7.7 7.3 6.3 22.6 delivery Glucose instability 7.9 4.8 20.0 5.2 6.0 23.1 3.7 0.7 17.0 NICU admission 87.0 96.3 85.0 45.1 61.7 65.4 20.1 27.8 47.2
ACCEPTED MANUSCRIPT
26
RI PT
Figure Legend
AC C
EP
TE D
M AN U
SC
Figure 1. Distribution of weeks at delivery among late preterm deliveries with respiratory complications by eligibility for antenatal steroid
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT