.
Abstract
BACKGROUND
Induction of labor among low-risk, 39-week nulliparas increased significantly in the United States following publication of the outcomes of A Randomized Trial of Induction Versus Expectant Management trial.
However, the rates of labor induction and outcomes in non-nulliparous patients and the wider impacts on the labor unit have not been reported widely.
OBJECTIVE
This study aimed to compare the induction of labor rates and outcomes before and after the liberal implementation of 39-week elective induction at a single center.
STUDY DESIGN
This was a retrospective cohort study comparing the delivery characteristics of pregnancies 1 year before and 1 year after adoption of a new 39-week elective induction policy at a single, tertiary-care center. Notably, elective induction was not restricted to nulliparas.
We examined all live, singleton, in-born deliveries ≥36 weeks gestation, excluding those with fetal anomalies and prolonged antenatal admission.
Deliveries at ≥39 weeks gestation were further subcategorized as being high risk (diabetes mellitus, chronic hypertension, intrauterine growth restriction, history of fetal demise or cholestasis) or low risk, nulliparas vs multiparas, and with or without a previous cesarean delivery.
Elective deliveries were those without a maternal, fetal, or obstetrical indication. Primary outcomes included gestational age and indications for delivery, rates of labor induction and elective induction, and time from admission to delivery.
Secondary outcomes included the rate of cesarean deliveries, indications for cesarean deliveries, and maternal and newborn morbidities.
The outcomes were compared using Wilcoxon rank-sum tests or chi-square tests as appropriate. The odds of cesarean delivery were analyzed using multivariate logistic regression and controlling for relevant confounders.
RESULTS
A total of 2672 pre-implementation and 2526 post-implementation deliveries were studied. Among patients at ≥39 weeks gestation, elective delivery increased (pre-implementation, 344/1788 [19.2%] vs post-implementation, 684/1710 [40.0%]; P<.01) and admission for labor or ruptured membranes decreased (pre-implementation, 920/1788 [51.5%] vs post-implementation, 579/1710 [33.9%]; P<.01).
Labor induction in the 39th week of gestation increased among low-risk and high-risk nulliparas, multiparas, and those with a previous cesarean delivery (P<.05 for each pairwise comparison), and the rate of 39-week elective inductions increased in all low-risk subgroups.
Deliveries at 36 to 38 weeks gestation were similar in the proportion, timing, indications for delivery, and rate of labor induction.
The odds of cesarean delivery was unchanged overall (adjusted odds ratio, 0.97; 95% confidence interval, 0.83–1.14) and for low-risk, ≥39-week nulliparas (adjusted odds ratio, 0.90; 95% confidence interval, 0.66–1.23) and low-risk, ≥39-week multiparas (adjusted odds ratio, 1.18; 95% confidence interval, 0.71–1.98).
Among all deliveries, the median (interquartile range) time from admission to delivery increased significantly (pre-implementation, 12.8 [6.0–21.6] hours vs post-implementation, 15.6 [7.1–25.1] hours; P<.01) and the total cumulative patient care time from admission to delivery increased by 15% (pre-implementation, 41,578 hours vs post-implementation, 47,605 hours) when normalized by delivery volume.
Chorioamnionitis incidence increased, whereas other maternal and neonatal morbidities were unchanged.
CONCLUSION
Following adoption of a nonrestrictive, 39-week elective induction policy at a single, tertiary-care center, the rates of 39-week induction of labor and elective inductions increased among nulliparas, multiparas, and those with a previous cesarean delivery.
The rate of cesarean delivery was unchanged, and the median time from admission to delivery and the cumulative admission to delivery hours increased significantly.
Future studies are needed to further explore the full scope of the impacts on labor unit operations, costs, and patient experiences and outcomes.
AJOG Global Reports at a Glance.
Why was this study conducted?
This study was conducted to assess real-world outcomes in a single center after liberalization of a 39-week elective induction policy that was not exclusive offered to nulliparas.
Key findings
The 39-week elective induction policy was applied widely among nulliparas, multiparas, and patients with a previous cesarean delivery. The rate of cesarean delivery was unchanged, and the median time spent per patient from admission to delivery in the labor and delivery unit increased by several hours. Other morbidities were mostly unchanged.
What does this add to what is known?
Our study showed that when openly permitting 39-week elective inductions in a real-world setting, this service was not exclusively offered to nulliparous patients. We quantified the degree of change in labor induction, the time from admission to delivery, and morbidities across an entire labor and delivery unit.
Introduction
In the year following publication of the outcomes of the A Randomized Trial of Induction Versus Expectant Management (ARRIVE) trial,1 induction of labor (IOL) among low-risk nulliparas at ≥39 weeks gestation increased in the United States from 30.2% to 36.1%.2 Less reported, however, is the degree to which other subgroups of patients at ≥39 weeks gestation may be undergoing elective IOL at 39 weeks gestation and how labor unit operations and outcomes might be impacted by so many additional inductions.
In the early spring of 2020, our institution liberalized its labor induction policy to explicitly permit elective IOL at 39 weeks gestation, a policy that notably did not exclude multiparas. In the current report, we explored multiple possible consequences of this new policy, using broad inclusion criteria to study changes in labor induction, delivery timing and indications, early-term birth, cesarean delivery (CD) rates and reasons, time from admission to delivery, and maternal and neonatal morbidities. By including pregnancies at ≥36 weeks gestation, we sought to understand these effects within the larger context of the overall labor unit.
Materials and Methods
We performed a retrospective cohort study examining all deliveries at ≥36 weeks gestation that occurred 1 year before and 1 year after the adoption of a liberalized labor induction policy at the Santa Clara Valley Medical Center (SCVMC) in San Jose, California, a county government–owned and –operated academic, tertiary-care medical center. All deliveries were performed by an in-house hospital team made up of obstetrics and gynecology residents and attending physicians. No deliveries were performed by private physicians at our institution.
In March and April of 2020, the SCVMC altered its departmental policy to specifically permit elective IOL at 39 weeks gestation, which was previously neither specifically prohibited nor supported. Our policy contained no a priori exclusions (eg, multiparty, previous CD) because these were thought to unnecessarily exceed the American College of Obstetricians and Gynecologists (ACOG) guidelines. Attending physicians were encouraged to use professional judgment and the available literature to determine which candidates were suitable and who should be recommended to undergo an elective 39-week IOL. Although our policy originated out of concern for a pending surge in COVID-19 admissions that might overwhelm hospital resources in the spring of 2020, subsequent ebbs and surges in hospital COVID-19 census did not alter our induction policy across the study period. Delivery at 41 weeks gestation was recommended for all pregnancies, and deliveries <39 weeks gestation required an appropriate medical indication3 or approval by the maternal-fetal medicine division.
Up to 4 inductions could be scheduled 7 days a week. Patients waited at home and were called in when a labor room became available. Our labor unit has 14 rooms, 2 operating rooms, and 8 to 12 nurses per shift. Preferred cervical ripening methods included oral misoprostol (if the patient had no previous CD) and/or a Foley balloon inflated to 60 to 80 cc. We did not offer outpatient cervical ripening during the study period and no patients were referred to deliver in other hospitals because of a lack of capacity. It is not customary to admit patients the night before scheduled IOL for cervical ripening.
Study periods extended from March 1, 2019 to February 29, 2020 (pre-implementation period [PRE]), and May 1, 2020 to April 30, 2021 (post-implementation period [POST]). March and April of 2020 were excluded from the analysis because it served as a phase-in period. By mid-April 2020, COVID-19 testing occurred on or before arrival for all patients admitted to the labor and delivery unit.4 We excluded multiple gestations, cases with intrauterine fetal demise, life-limiting conditions of the fetus or newborn, delivery outside the hospital, admission before 36 weeks gestation, and prolonged antenatal admission unless IOL began within 72 hours. Permission for this study was granted by the SCVMC institutional review board.
Data were obtained from the electronic medical records (Epic Systems Corporation, Madison, WI) via standardized delivery reports. Gestational age in the current study refers to the time of admission rather than the time of delivery to avoid upward bias in patients who underwent IOL. Information on diabetes mellitus, hypertension, and other maternal and infant morbidities was obtained from patient discharge data submitted to the Maternal Data Center of the California Maternal Quality Care Collaborative of which the SCVMC is a member. Severe maternal morbidity and mortality were identified from the International Classification of Diseases, 10th Revision, procedure and diagnosis codes based on the Centers for Disease Control and Prevention–defined specifications. Case reviews of the submitted data were conducted by trained nurses and hospital quality and safety leads.
The primary indication for delivery was manually abstracted from the admission history and physical examination and was categorized as either labor or spontaneous rupture of membranes (SROM) (provider diagnosed labor, regardless of contraction frequency or cervical dilation or SROM), medical indication (hypertensive disorder, diabetes mellitus, bleeding, oligohydramnios, or other ACOG-approved indication)3, or elective (no ACOG-approved indication). The primary indication for CD, if performed, was also collected manually. CD was categorized as planned if the indication reasonably suggested (in the authors’ opinion) that the CD had been the intended mode of delivery before admission (eg, desired repeat, placenta previa, etc.).
Primary outcomes included gestational age and indication for delivery, rates of IOL and elective IOL, and admission-to-delivery time. To distinguish changes in the rates of IOL in more detail, deliveries were divided into 2 primary cohorts, namely deliveries during 36 to 38 weeks gestation and those at ≥39 weeks gestation. Deliveries at ≥39 weeks gestation were then further divided into the following subsets: those with and those without high-risk medical conditions; nulliparous vs multiparous; and those with and those without a previous CD. High-risk conditions were those presumed to have been present before 39 weeks gestation for which delivery was indicated at 39 weeks (pregestational or gestational diabetes mellitus, chronic hypertension, cholestasis, history of fetal demise, and/or intrauterine growth restriction). In otherwise low-risk patients, gestational hypertension and preeclampsia were assumed to have developed anew at or after 39 weeks gestation, because delivery would have been indicated on diagnosis if found earlier.3
Secondary outcomes included the rate of CD, indications for CD, and maternal and newborn morbidities. Severe maternal morbidity was defined as any of 20 possible acute maternal conditions (Appendix A). An unexpected newborn complication was defined as a moderate or severe neonatal complication, excluding congenital malformations, birthweight <2500 g, or exposure to maternal drug use (Appendix A). Postpartum length of stay was not reported because changes to visitation and discharge policies during the COVID-19 pandemic in the second year of this study were thought to be a significant source of bias related to that outcome.
To examine the evidence for delays in care, we compared the time from admission to administration of the initial cervical ripening agent in patients who underwent labor induction and the proportion of planned CDs that occurred during the daytime shift hours (from 7 am to 3 pm) but excluding those who presented with labor or SROM.
Baseline characteristics, delivery outcomes, and morbidities were compared using Wilcoxon rank-sum tests, chi-square tests, or Fisher’s exact tests, as appropriate, with P<.05 considered significant. Gestational age at admission was compared by time-to-event analysis using Cox proportional hazard regression and controlling for diabetes mellitus, chronic hypertension, placenta previa, and a history of transfundal uterine surgery. The rate of CD was compared using multivariate logistic regression and controlling for relevant confounders. In the group overall, these confounders were age, prepregnancy body mass index (BMI), race and ethnicity, diabetes mellitus, previous CD, malpresentation, and placenta previa or other absolute contraindication to vaginal delivery. For low-risk, ≥39-week nulliparas and low-risk, ≥39-week multiparas, the rate of CD was adjusted for age, prepregnancy BMI, and maternal race and ethnicity.5,6
Results
Of 5593 delivered patients, 395 were excluded (Figure 1), leaving 2672 PRE and 2526 POST deliveries for analysis. Delivered patients were primarily Hispanic and insured by Medi-Cal or other public insurance (Table 1).
Figure 1.
POST, post-implementation deliveries; PRE, pre-implementation deliveries.
Jelks. Embracing 39-week elective induction. Am J Obstet Gynecol Glob Rep 2023.
Table 1.
Maternal characteristics at baseline
Characteristics | PRE n=2672 | POST n=2526 | P value |
---|---|---|---|
Age (y) | 29 (24–34) | 29 (34–34) | .17 |
Gravity | 2 (1–4) | 2 (1–4) | .80 |
Parity | 1 (0–2) | 1 (0–2) | .80 |
Nulliparous | 942 (35.3) | 900 (35.6) | .84 |
Prepregnancy BMI (kg/m2) | 27 (23–32) | 27 (23–32) | 1.00 |
<25 | 896 (33.5) | 808 (32.0) | .06 |
25 to <30 | 814 (30.5) | 839 (33.2) | |
30 to <35 | 535 (20.0) | 480 (19.0) | |
35 to <40 | 248 (9.3) | 260 (10.3) | |
≥40 | 179 (6.7) | 139 (5.5) | |
Previous cesarean delivery | 523 (19.6) | 466 (18.5) | .32 |
Payer category | |||
Medi-Cal or other Government | 2409 (90.2) | 2307 (91.3) | .26 |
Private | 245 (9.2) | 200 (7.9) | |
Self-pay | 18 (0.7) | 19 (0.8) | |
Race and ethnicity | |||
Non-Hispanic White | 211 (7.9) | 180 (7.1) | .04 |
Non-Hispanic Black | 140 (5.2) | 111 (4.4) | |
Hispanic | 1886 (70.6) | 1878 (74.3) | |
Asian or Pacific Islander | 425 (15.9) | 345 (13.7) | |
Other or unknown | 10 (0.4) | 12 (0.5) | |
Diabetes mellitus | |||
Pregestational type 1 or 2 | 65 (2.4) | 52 (2.1) | .42 |
Gestational diabetes mellitus | 519 (19.4) | 580 (23.0) | .01 |
Chronic hypertension | 81 (3.0) | 83 (3.3) | .66 |
The data are shown as median (interquartile range) or number (percentage).
BMI, body mass index.
Jelks. Embracing 39-week elective induction. Am J Obstet Gynecol Glob Rep 2023.
Gestational age at admission was significantly different in the POST group (P<.01) (Figure 2) because of the increased deliveries in the 39th week of gestation. Deliveries at 36 to 38 weeks gestation were similar in proportion, timing, indications for delivery, and rate of labor induction (Table 2). Among the deliveries at ≥39 weeks gestation, admission after labor or SROM decreased and elective delivery increased.
Figure 2.
POST, post-implementation deliveries; PRE, pre-implementation deliveries.
Jelks. Embracing 39-week elective induction. Am J Obstet Gynecol Glob Rep 2023.
Table 2.
Indication for delivery and induction of labor beyond and before 39 weeks gestation
Outcome | PRE | POST | P value |
---|---|---|---|
36–38 wk of gestation deliveries | 885 (33.1) | 819 (32.4) | .60a |
Indication for delivery | .86 | ||
Labor or SROM | 542 (61.2) | 509 (62.1) | |
Medical Indication | 336 (38.0) | 305 (37.2) | |
Elective | 7 (0.8) | 5 (0.6) | |
Planned CDb | 150 (16.9) | 122 (14.9) | .28 |
Induction of labor | 254 (28.7) | 232 (28.3) | .91 |
≥39 wk of gestation deliveriesc | 1788 (66.9) | 1710 (67.6) | .60a |
Indication for delivery | <.01 | ||
Labor or SROM | 920 (51.5) | 579 (33.9) | |
Medical indication | 524 (29.3) | 447 (26.1) | |
Elective | 344 (19.2) | 684 (40.0) | |
Planned CD | 238 (13.3) | 220 (12.9) | .73 |
Induction of labor | 651 (36.4) | 926 (54.2) | <.01 |
Elective induction of labor | 194 (10.9) | 551 (32.2) | <.01 |
All patients, induction of labor | 904/2672 (33.8) | 1157/2526 (45.8) | <.01 |
The data are shown as number (percentage).
CD, cesarean delivery; POST, post-implementation deliveries; PRE, pre-implementation deliveries; SROM, spontaneous rupture of membranes.
a
Comparison between the 36 to 38 weeks gestation deliveries and the ≥39 weeks gestation deliveries
b
Deemed planned CD for any of the following indications: desired repeat CD, placenta previa, malpresentation, history of shoulder dystocia, macrosomia, or other nonemergent condition(s) that precluded attempt of labor or vaginal delivery, unless such condition occurred after initiation of labor induction
c
A total of 23 patients (4 in the PRE group and 19 in the POST group) who were admitted in the late evening hours at 38+6 weeks gestation for elective 39-week induction of labor were included in the ≥39-week group for analysis.
Jelks. Embracing 39-week elective induction. Am J Obstet Gynecol Glob Rep 2023.
Among the deliveries at ≥39 weeks gestation without planned CD, the rate of 39-week IOL increased significantly in all subgroups (P<.05 for each pairwise comparison); 39-week elective IOL increased in all low-risk subgroups (Table 3).
Table 3.
Induction of labor in the 39th week of gestation by parity and risk level
Outcome | IOL in the 39th week of gestation | Elective IOL in the 39th week of gestation | ||||
---|---|---|---|---|---|---|
Subgroup | PRE | POST | P value | PRE | POST | P value |
Low-risk nulliparas | 52/520 (10.0) | 159/509 (31.2) | <.01 | 29/520 (5.6) | 144/509 (28.3) | <.01 |
Low-risk multiparas | 68/600 (11.3) | 191/564 (33.9) | <.01 | 50/600 (8.3) | 182/564 (32.3) | <.01 |
Low-risk, previous CD | 6/109 (5.5) | 28/99 (28.3) | <.01 | 4/109 (3.7) | 26/99 (26.3) | <.01 |
High-risk nulliparas | 49/109 (45.0) | 64/96 (66.7) | <.01 | |||
High-risk multiparas | 107/175 (61.1) | 139/184 (75.5) | <.01 | |||
High-risk, previous CD | 12/37 (32.4) | 23/38 (60.5) | .03 |
The data are shown as number/total number (percentage).
High-risk refers to presence of any of the following conditions in a ≥39-week pregnancy: chronic hypertension, pregestational or gestational diabetes mellitus, cholestasis of pregnancy, history of fetal demise, and/or intrauterine growth restriction. Low-risk refers to the absence of any of those conditions.
CD, cesarean delivery; IOL, induction of labor; POST, post-implementation deliveries; PRE, pre-implementation deliveries.
Jelks. Embracing 39-week elective induction. Am J Obstet Gynecol Glob Rep 2023.
The overall median time from admission to delivery increased significantly (Table 4). When all deliveries were taken together, the total cumulative admission-to-delivery hours increased by 15% (PRE, 41,578 hours vs POST, 47,605 hours) after normalization for delivery volume, and this was driven by an increase in the labor time of patients opting for elective IOL at ≥39 weeks gestation (Figure 3).
Table 4.
Time from admission to delivery and labor induction management
Outcome | PRE | POST | P value |
---|---|---|---|
Admission to delivery time (h) | 12.8 (6.0–21.6) | 15.6 (7.1–25.1) | <.01 |
Planned CD, without labor/SROM | 4.5 (3.5–6.2) | 4.8 (3.8–6.5) | .25 |
Nullipara, labor/SROM | 15.1 (10.5–20.4) | 14.6 (9.9–21.1) | .78 |
Nullipara, IOL | 29.9 (21.4–39.5) | 31.8 (24.6–40.2) | .01 |
Multipara, labor/SROM | 7.0 (3.6–11.0) | 6.9 (3.2–11.6) | .92 |
Multipara, IOL | 19.7 (14.6–26.3) | 20.3 (15.6–25.9) | .27 |
Cervical ripening used during IOL | 762/905 (84.2%) | 1018/1159 (87.8%) | .02 |
Misoprostol | 595/905 (65.7%) | 854/1159 (73.7%) | <.01 |
Foley balloon | 567/905 (62.7%) | 716/1159 (61.8%) | .72 |
Dual agenta | 39/905 (4.3%) | 38/1159 (3.3%) | .27 |
Time from admission to initial administration of cervical ripening agent in IOL (h) | 3.1 (2.2–4.6) | 3.3 (2.3–4.6) | .08 |
The data are shown as median (interquartile range) or number/total number (percentage).
CD, cesarean delivery; IOL, induction of labor; POST, post-implementation deliveries; PRE, pre-implementation deliveries; SROM, spontaneous rupture of membranes.
aDual agent indicates misoprostol plus Foley balloon placed within 1 hour of each other as initial induction method.
Jelks. Embracing 39-week elective induction. Am J Obstet Gynecol Glob Rep 2023.
Figure 3.
CD, cesarean delivery; IOL, induction of labor; PRE, pre-implementation period; POST, post-implementation; SROM, spontaneous rupture of membranes.
Jelks. Embracing 39-week elective induction. Am J Obstet Gynecol Glob Rep 2023.
Secondary outcomes are reported in Table 5 for the cohorts overall and in Table 6 for low-risk, ≥39-week nulliparas and low-risk, ≥39-week multiparas. Indications for CD were similar, including for all types of labor arrest disorders. The rate of CD was unchanged overall and among ≥39-week, low-risk nulliparas and low-risk multiparas. Among all deliveries, chorioamnionitis was more common in the POST period, whereas other maternal and neonatal morbidities were similar.
Table 5.
Secondary outcomes in all study patients
Outcome | PRE n=2672 | POST n=2526 | P value or adjusted odds ratio (95% CI) |
---|---|---|---|
Maternal | |||
Epidural | 1728 (64.7) | 1698 (67.2) | .06 |
CD | 699 (26.2) | 626 (24.8) | 0.974 (0.83–1.14)a |
Operative vaginal delivery | 76 (2.8) | 67 (2.7) | .74 |
Vaginal birth after CD | 161/523 (30.8) | 140/466 (30.0) | .85 |
Indication for cesarean deliveryb | |||
Previa or previous transfundal uterine surgery | 24 (3.4) | 20 (3.2) | .85 |
Malpresentation | 96 (13.7) | 79 (12.6) | |
Desired repeat CD | 266 (38.1) | 242 (38.7) | |
Fetal condition | 15 (2.1) | 7 (1.1) | |
Maternal health | 11 (1.6) | 16 (2.6) | |
Arrest of labor: latent phase | 54 (7.7) | 52 (8.3) | |
Arrest of labor: active phase | 67 (9.6) | 59 (9.4) | |
Arrest of labor: second stage | 57 (8.2) | 49 (7.8) | |
NRFHT or other intrapartum emergency | 104 (14.9) | 94 (15.2) | |
Maternal request | 5 (0.7) | 7 (1.1) | |
3rd- or 4th-degree perineal laceration | 111 (4.2) | 107 (4.2) | .94 |
Preeclampsia or eclampsia | 281 (10.5) | 287 (11.4) | .35 |
Chorioamnionitis | 213 (8.0) | 242 (9.6) | <.05 |
QBL ≥1500 mL | 145 (5.4) | 115 (4.6) | .17 |
Transfusion | 53 (2.0) | 69 (2.7) | .09 |
Severe maternal morbidity excluding transfusionc | 26 (1.0) | 36 (1.4) | .17 |
ICU admission | 3 (0.1) | 1 (0.0) | .66 |
Neonatal | |||
Birthweight (g) | 3370 (3080–3360) | 3340 (3050–3637) | .03 |
APGAR score <7 at 5 min | 24 (0.9) | 26 (1.0) | .74 |
Unexpected newborn complicationde | 67 (2.5) | 64 (2.5) | .98 |
Unexpected newborn complication, severede | 28 (1.0) | 26 (1.0) | .95 |
Data are shown as number (percentage), number/total number (percentage), or median (interquartile range).
CD, cesarean delivery; ICU, intensive care unit; NRFHT, nonreassuring fetal heart tracing; POST, post-implementation deliveries; PRE, pre-implementation deliveries; QBL, quantitative blood loss.
a
Adjusted for age, prepregnancy body mass index, race and ethnicity, diabetes mellitus, previous CD, fetal presentation at delivery, and placenta previa or other absolute contraindication to vaginal delivery
b
Percentages refer to the subset of patients who underwent CD
c
Severe maternal morbidity is defined as any of 20 possible acute maternal conditions including, but not limited to, pulmonary edema, sepsis, shock, venous thromboembolism, heart failure, and others; a complete list is available in Appendix A
d
An unexpected newborn complication is defined as a moderate or severe neonatal complication, excluding congenital malformations, severe fetal conditions, birthweight <2500 g, or exposure to maternal drug use; a complete list is available in Appendix A
e
Data are not available for 1 infant in the POST period that was given up for adoption.
Jelks. Embracing 39-week elective induction. Am J Obstet Gynecol Glob Rep 2023.
Table 6.
Secondary outcomes in low-risk nulliparas and low-risk multiparas at ≥39 weeks gestation
Outcome | PRE | POST | P value or adjusted odds ratio (95% CI) |
---|---|---|---|
Low-risk mulliparas | n=520 | n=509 | |
Epidural | 458 (88.1) | 468 (91.8) | .06 |
Cesarean delivery | 105 (20.2) | 94 (18.4) | 0.90 (0.66–1.23)a |
Preeclampsia or eclampsia | 79 (15.2) | 76 (14.9) | .97 |
Chorioamnionitis | 94 (18.1) | 110 (21.6) | .18 |
QBL ≥1500 mL | 35 (6.7) | 34 (6.7) | .93 |
Transfusion | 19 (3.7) | 28 (5.5) | .21 |
ICU admission | 1 (0.2) | 0 (0) | — |
Severe maternal morbidity excluding transfusionb | 6 (1.2) | 11 (2.2) | .31 |
Unexpected newborn complication, severec | 11 (2.1) | 7 (1.4) | .50 |
APGAR score <7 at 5 min | 6 (1.2) | 4 (0.8) | .77 |
Low-risk multiparas | n=600 | n=564 | |
Epidural | 401 (66.8) | 374 (66.4) | .90 |
Cesarean delivery | 30 (5.0) | 33 (5.9) | 1.18 (0.71–1.98)a |
Preeclampsia or eclampsia | 29 (4.8) | 36 (6.4) | .31 |
Chorioamnionitis | 23 (3.8) | 21 (3.7) | .96 |
QBL ≥1500 mL | 23 (3.8) | 12 (2.1) | .13 |
Transfusion | 6 (1.0) | 5 (0.9) | .92 |
ICU admission | 0 (0) | 0 (0) | — |
Severe maternal morbidity excluding transfusionb | 3 (0.5) | 6 (1.1) | .45 |
Unexpected newborn complication, severec | 2 (0.3) | 2 (0.4) | .66 |
APGAR score <7 at 5 min | 1 (0.2) | 5 (0.9) | .19 |
Data is shown as number (percentage).
ICU, intensive care unit; POST, post-implementation deliveries; PRE, pre-implementation deliveries; QBL, quantitative blood loss.
a
Adjusted for maternal age, race and ethnicity, and prepregnancy body mass index
b
Severe maternal morbidity is defined as any of 20 possible acute maternal conditions including, but not limited to, pulmonary edema, sepsis, shock, venous thromboembolism, heart failure, and others; a complete list is available in Appendix A
c
An Unexpected newborn complication is defined as a moderate or severe neonatal complication, excluding congenital malformations, severe fetal conditions, birthweight <2500 g, or exposure to maternal drug use; a complete list is available in Appendix A.
Jelks. Embracing 39-week elective induction. Am J Obstet Gynecol Glob Rep 2023.
We did not find evidence of significant care delays in the POST period. Time to administration of the initial cervical ripening agent was similar. The proportion of planned CDs (excluding patients with labor or SROM) that occurred during the day shift hours from 7 am to 3 pm was similar (PRE, 245/299 [81.9%] vs POST, 227/275 [82.5%]; P=.85).
Comment
Principal findings
This study reports on the system-wide impacts of a 39-week elective IOL policy implementation in a post–ARRIVE trial era. In a detailed look at our single center, we found that adoption of a policy of permissive elective 39-week IOL led to application not only to cases with low-risk nulliparas. We observed increases in IOL and elective IOL among all subgroups of patients after 39 weeks gestation, thereby increasing our overall rate of IOL from 33.9% to 45.8%. The patient care time from admission to delivery in the labor and delivery unit increased by 15% as a consequence.
We did not observe a significant change in the rate of CDs either overall or in the subgroups of low-risk, ≥39-week nulliparas or low-risk, ≥39-week multiparas. Chorioamnionitis was notably increased in the POST cohort; all other maternal and neonatal outcomes were unchanged, including the rate of preeclampsia.
Because more full-term deliveries were shifted to the 39th week, we did not observe an increase in either elective or medically indicated deliveries between 36 and 38 weeks gestation.
Because our POST cohort was cared for in the era of COVID-19, we considered the possibility of treatment delays owing to testing or infection control measures. Among the limited outcomes available to analyze this variable, we found no difference in the timing of administration of the initial cervical ripening agent among induced patients nor in the proportion of planned CDs for patients not presenting for labor or SROM during daytime hours between 7 am and 3 pm.
Results in the context of what is known
Studies to date comparing elective IOL at term with expectant management, whether prospective or retrospective, have focused on selected groups of patients, generally nulliparous and low-risk patients who make up only a portion of the deliveries in any labor unit at a time.1,7, 8, 9 Having located no comparable studies, we elected to take a different approach to examine how liberal implementation of 39-week elective IOL may affect the overall delivering population across a single labor unit. We found that permitting elective 39-week IOL in an unrestricted manner effectively opened the door for many patients other than low-risk nulliparas to take advantage of this option.
Our finding of a longer time from admission to delivery among induced patients substantiates previous studies.1,7,8 We add additional information about the degree of change in the cumulative admission-to-delivery time across a whole population under liberal implementation of an elective IOL policy.
One of the key findings of the ARRIVE trial was a lower rate of CD in low-risk nulliparas randomized to IOL,1 which has been corroborated in a meta-analysis of randomized trials10 and a meta-analysis of cohort studies.9 In this study, we observed an increase in the rate of 39-week IOL among low-risk nulliparas from 10.0% to 31.2% and we did not observe a statistically significant decrease in the CD rate. Instead of questioning the generalizability of previous studies, we suggest a caveat based on our real-world experience, which is that across an entire year and an entire labor and delivery unit, the primary effect of permitting 39-week elective inductions was a visibly higher induction rate, whereas any improvement in the CD rate may only be barely noticeable.
Our finding of an increased rate of chorioamnionitis has not been found in other studies1,7,9 despite the association between IOL and longer labors. Our findings may warrant further investigation into local contributing factors, including the number of cervical examinations during labor.11 Our finding of a lack of change in the rate of preeclampsia is also in contrast with the finding of the ARRIVE trial.1 We attribute this to a smaller difference in the rate of IOL between the groups in our study and the groups in the ARRIVE trial. We did not find a change in adverse neonatal outcomes with liberal use of a 39-week IOL policy, which is consistent with previous studies.1,2,7,8,10
We were unable to find any studies describing the rates of late-preterm and early-term births after liberalization of elective IOL at 39 weeks gestation, which was our incentive to study this issue. Such a shift might have been possible if providers became more likely to deliver complicated pregnancies earlier in their ACOG-recommended delivery windows3 than previously because more low-risk pregnancies were delivered at 39 weeks gestation. We found no such difference.
Clinical implications
Our experience suggests that if more capacity for labor induction is made available, increased demand will likely fill it, leading to a significant increase in the overall admission-to-delivery hours in labor and delivery units. The ability to offer elective induction at 39 weeks gestation thus likely depends most heavily on the labor and delivery unit capacity, bed space, and staffing. With a drop in national delivery numbers over the last decade,12 an increasing number of hospitals may find that, similar to ours, they are more able to physically accommodate increased lengths of stay in labor and delivery units, whereas this change may be less feasible in units that are already operating close to capacity.
Labor units that offer elective 39-week IOL are also likely to be confronted with the dilemma of which patients should be included or excluded from accessing this service. Although the Society for Maternal-Fetal Medicine states that it is “reasonable to offer elective IOL to low risk nulliparas” based on the ARRIVE trial,13 elective 39-week IOL among multiparas is not overtly endorsed by national societies and has not been shown to conclusively improve outcomes.8,10 In our center, providers felt that restricting elective IOL to only nulliparas threatened to unnecessarily infringe on patient autonomy and professional provider judgment given that the ACOG does not otherwise bar or constrain this practice.14
IOL in patients with a previous CD remains even more controversial with a recent, large retrospective study showing a higher CD rate after labor induction when compared with expectant management.15 Our inclusion of a very small number of such patients suggests no difference in vaginal birth or morbidity after liberalization of 39-week IOL. Even if centers restrict elective IOL to patients without a previous CD, it will have little effect on the overall volume—only 1 in 13 patients (43/552) who underwent elective IOL in our population in the POST period had a previous CD.
Finally, our experience suggests that improvement in rates of CD and hypertensive disorders seen in randomized trials of elective IOL may not manifest in a real-world setting or it may not be perceptible at the local level, whereas other adverse outcomes, such as chorioamnionitis, may actually increase.
Research implications
Based on our experience, more studies on the rates, costs, and outcomes of elective IOL in the present era in real-world settings and across a range of obstetrical patients are required. Future studies comparing intended admissions for elective IOL with those that are actually accommodated are warranted to more fully evaluate the impacts of elective IOL on patient care, feasibility, staffing, and labor unit operations. Additional key areas of future study also include long-term maternal and infant outcomes with elective IOL, including long-term breastfeeding rates, future pregnancy outcomes, and possible effects on patients’ perceptions of trial of labor after a CD and the success rates. Finally, studies on fair access to 39-week elective induction among women of different races, socioeconomic classes, insurance statuses, and across urban and rural geographic regions are warranted.
Strengths and limitations
The specific characteristics of our hospital as an urban, academic, tertiary-care hospital without private delivering physicians may limit the generalizability of our findings. We were unable to determine the intention of patients and providers in terms of the delivery timing and only had information on the actual timing of admission. Retrospective categorization of delivering patients as “low risk” or “high risk” is also potentially imprecise. Our own lack of unbiased data on postpartum maternal and neonatal length of stay limits our ability to report on this important consideration in elective IOL. And although we observed little to no difference in tangible measures of patient care and morbidity between the study periods, we are unable to know if the concurrent COVID-19 pandemic may have affected patient care or clinical decision-making in other unmeasured ways in the POST period.
Strengths of our study include clear pre- and postadoption time periods and the use of documented medical admission diagnoses and CD indications to distinguish whether a condition developed before or after admission and thus whether a given delivery was elective or medically indicated.
Conclusion
Following adoption of a nonrestrictive, 39-week elective induction policy at a single, tertiary-care center, we found a significant increase in 39-week deliveries and in 39-week IOL and elective IOL among low-risk and high-risk nulliparas, multiparas, and patients with a previous CD. We observed no change in the CD rate but observed significantly longer median and cumulative patient care times from admission to delivery. Future studies are needed to see if the changes we described also occur in other centers and to determine the full scope of impacts on labor unit operations and patient experiences and outcomes with increasing use of elective IOL.
Footnotes
The authors report no conflict of interest.
Patient consent was not required because no personal information or details were included.
This study did not receive any financial support.
Cite this article as: Jelks AT, Yao AQ, Byrne JD. Impacts of embracing 39-week elective induction across an entire labor and delivery unit. Am J Obstet Gynecol Glob Rep 2023;XX:x.ex–x.ex.
References
- 1.Grobman WA, Rice MM, Reddy UM, et al. Labor Induction versus Expectant Management in Low-Risk Nulliparous Women. N Engl J Med. 2018;379:513–523. doi: 10.1056/NEJMoa1800566. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Gilroy LC, Al-Kouatly HB, Minkoff HL, Mclaren RA. Changes in obstetrical practices and pregnancy outcomes following the ARRIVE trial. Am J Obstet Gynecol. 2022;226:716. doi: 10.1016/j.ajog.2022.02.003. e1–12. [DOI] [PubMed] [Google Scholar]
- 3.Medically indicated late-preterm and early-term deliveries: ACOG committee Opinion, Number 818. Obstet Gynecol. 2021;137:e29–e33. doi: 10.1097/AOG.0000000000004245. [DOI] [PubMed] [Google Scholar]
- 4.Sophia Y, Rishi B, James B, Andrea J. Universal testing for severe acute respiratory syndrome coronavirus 2 upon admission to three labor and delivery units in Santa Clara County, CA. Clin J Obstet Gynecol. 2020;3:109–113. [Google Scholar]
- 5.Rossi RM, Requarth E, Warshak CR, Dufendach KR, Hall ES, DeFranco EA. Risk calculator to predict cesarean delivery among women undergoing induction of labor. Obstet Gynecol. 2020;135:559–568. doi: 10.1097/AOG.0000000000003696. [DOI] [PubMed] [Google Scholar]
- 6.Debbink MP, Ugwu LG, Grobman WA, et al. Racial and ethnic inequities in cesarean birth and maternal morbidity in a low-risk, nulliparous cohort. Obstet Gynecol. 2022;139:73–82. doi: 10.1097/AOG.0000000000004620. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Bailit JL, Grobman W, Zhao Y, et al. Nonmedically indicated induction vs expectant treatment in term nulliparous women. Am J Obstet Gynecol. 2015;212:103. doi: 10.1016/j.ajog.2014.06.054. e1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Souter V, Painter I, Sitcov K, Caughey AB. Maternal and newborn outcomes with elective induction of labor at term. Am J Obstet Gynecol. 2019;220:273. doi: 10.1016/j.ajog.2019.01.223. e1–11. [DOI] [PubMed] [Google Scholar]
- 9.Grobman WA, Caughey AB. Elective induction of labor at 39 weeks compared with expectant management: a meta-analysis of cohort studies. Am J Obstet Gynecol. 2019;221:304–310. doi: 10.1016/j.ajog.2019.02.046. [DOI] [PubMed] [Google Scholar]
- 10.Middleton P, Shepherd E, Morris J, Crowther CA, Gomersall JC. Induction of labour at or beyond 37 weeks gestation. Cochrane Database Syst Rev. 2020;7 doi: 10.1002/14651858.CD004945.pub5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gluck O, Mizrachi Y, Ganer Herman HG, Bar J, Kovo M, Weiner E. The correlation between the number of vaginal examinations during active labor and febrile morbidity, a retrospective cohort study. BMC Pregnancy Childbirth. 2020;20:246. doi: 10.1186/s12884-020-02925-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hamilton BE, Martin JA, Osterman MJK. National Center for Health Statistics; 2021. Births: provisional data for 2020. Vital statistics rapid release.www.cdc.gov/nchs/data/vsrr/vsrr012-508.pdf Available at: Accessed January 15, 2021. [Google Scholar]
- 13.Society of Maternal-Fetal (SMFM) Publications Committee Electronic address: pubs@smfm.org. SMFM statement on elective induction of labor in low-risk nulliparous women at term: the ARRIVE trial. Am J Obstet Gynecol. 2019;221:B2–B4. doi: 10.1016/j.ajog.2018.08.009. [DOI] [PubMed] [Google Scholar]
- 14.ACOG Practice Bulletin No. 205: vaginal birth after cesarean delivery. Obstet Gynecol. 2019;133:e110–e127. doi: 10.1097/AOG.0000000000003078. [DOI] [PubMed] [Google Scholar]
- 15.Park BY, Cryer A, Betoni J, et al. Outcomes of labor induction at 39 weeks in pregnancies with a prior cesarean delivery. J Matern Fetal Neonatal Med. 2022;35:2853–2858. doi: 10.1080/14767058.2020.1807505. [DOI] [PubMed] [Google Scholar]