Volume 201, Issue 5 , Pages 477.e1-477.e7, November 2009
The association between the length of first stage of labor, mode of delivery, and perinatal outcomes in women undergoing induction of labor
Article Outline
Objective
To estimate the association between the lengths of the first stage of labor, mode of delivery, and perinatal outcomes in women undergoing labor induction.
Study Design
Retrospective cohort study of singleton, term pregnancies with labor induction and delivered during the second stage. The length of the first stage was examined by 6-hour intervals as predictors of mode of delivery and perinatal morbidity using χ2 test and multivariable logistic regression analysis.
Results
There were 3620 women who met study criteria. Compared with women with a first stage between 0-12 hours, women with longer first stages had a higher risk of cesarean delivery during the second stage, up to an adjusted odds ratio of 7.44 in those with a first stage ≥24 hours (95% confidence interval [CI], 3.43–16.1). Women with a first stage ≥24 hours also had higher odds of postpartum hemorrhage (adjusted odds ratio [aOR], 3.16; 95% CI, 1.73–5.79), chorioamnionitis (aOR, 2.83; 95% CI, 1.19–6.69), and neonatal admission to the intensive care nursery (aOR, 2.03; 95% CI, 1.10–3.74).
Conclusion
In women who underwent induction of labor, even when a second stage of labor was reached, the risk for cesarean delivery and maternal morbidity remained increased when the length of the first stage was longer than 24 hours. However, in this clinical scenario, the frequency of cesarean delivery remains less than 50%. The decision for surgical intervention thus should not be based on the elapse of time alone.
Key words: induction of labor, length of labor, perinatal complications
The goal of induction of labor is to achieve vaginal delivery.1 In the United States, the frequency of labor induction has increased more than 2-fold from 9.5% in 1990 to 22.4% for term pregnancies in 2004.2 A similar trend of rise in labor induction also occurs worldwide.3, 4 Many studies have reported an association between induction of labor and cesarean delivery, both for medically indicated and elective inductions,5 as well as in low-risk populations.6, 7, 8
The Bishop pelvic scoring system was devised to predict the probability of vaginal delivery based on cervical status.9 Labor induction may involve cervical ripening methods, either with mechanical dilators or pharmaceutical agents, when the cervix is deemed unfavorable. In this setting, the induction process can be prolonged. Although studies on induction often focus on methods and efficacy of different induction agents, little data exist regarding the duration of labor induction and associated outcomes. One recent study examined factors associated with the length of the latent phase during labor induction and reported that only the modified Bishop score at admission is associated with the length of latent labor; however, the association between perinatal outcome and duration of induction remains largely unexplored.10
Given the current rise in the frequency of induction, it is not uncommon to encounter a woman undergoing induction of labor for multiple days. The current American College of Obstetricians and Gynecologists (ACOG) practice guidelines on induction of labor do not specify a time limitation from the initiation of induction to delivery.1 Thus, how long is too long, and what constitutes a “failed induction of labor”?
If, despite a long first stage of labor, women who subsequently reached the second stage have a high likelihood of achieving vaginal delivery, perhaps clinicians should exercise patience with close surveillance to maximize a woman's potential for vaginal birth. Alternatively, if a long first stage is indicative of a high risk of cesarean delivery and perinatal morbidity even when the second stage is achieved, then there may exist a time threshold after which the risk of potential complications outweigh continuing labor induction. Thus, we sought to estimate whether the duration of the first stage of labor in women undergoing induction is associated with mode of delivery and perinatal outcomes during the second stage. We hypothesized that even beyond 24 hours of an induction of labor, the majority of women would still achieve vaginal delivery.
Materials and Methods
We designed a retrospective cohort study of all term and postterm, cephalic, live singleton births to women who had labor induction and delivered in the second stage of labor at the University of California, San Francisco (UCSF) between 1985 and 2001. Institutional review board approval was obtained from the Committee on Human Research at UCSF. The exclusion criteria were multiple gestations, delivery before 37 weeks' gestation, noncephalic presentation, spontaneous onset of labor, cesarean delivery during the first stage of labor, placenta previa, intrauterine fetal demise, or known lethal congenital anomalies. All deliveries at this academic institution were performed either by resident physicians with supervision by the attending physicians or by clinical nurse midwives.
The maternal characteristics and labor information was prospectively collected by the managing physicians on labor and delivery, and the neonatal information was similarly collected by the pediatricians using a neonatal database. The maternal and neonatal databases were then linked using 2 unique maternal and neonatal identifiers and crosschecked to ensure accuracy. The combined perinatal database was maintained with daily chart review by trained abstractors for both the maternal and neonatal information to ensure accuracy and to minimize missing data. The database was also reviewed monthly by trained physicians for quality assurance.
The beginning of the first stage of labor was defined as onset of painful, regular contractions, occurring every 5 minutes or at least 3 contractions in a 10-minute period as reported by the parturient or staff as applicable, in the presence of cervical change documented by the clinicians staffing labor and delivery. The end of the first stage was defined by complete cervical dilation. Thus, the entire duration of first stage included both the latent and active phases of labor; of note, we did not include the time interval from initiation of labor induction to onset of labor as part of first stage. All inductions were performed as inpatient, and oxytocin was administered by the nursing staff according to the institutional protocol, which remained unchanged during the study period. The second stage of labor was defined as time duration from complete cervical dilation to delivery of the neonate. The duration of first stage of labor was categorized into 4 intervals: 0-12 hours, 12-18 hours, 18-24 hours, and ≥24 hours. The 12-, 18-, and 24-hour thresholds represent the 75th, 90th, and 95th percentile of first stage duration of the study cohort. Maternal characteristics such as age, parity, race/ethnicity, prepregnancy body mass index (BMI), gestational age at delivery, epidural anesthesia use during labor, and the length of second stage were examined as factors associated with mode of delivery.
The primary outcome of interest was mode of delivery: spontaneous vaginal delivery (SVD), operative vaginal delivery (including both vacuum-assisted vaginal delivery and forceps delivery), and cesarean delivery. Maternal outcomes examined included incidence of third- or fourth-degree perineal laceration, postpartum hemorrhage (PPH, defined as estimated blood loss >500 mL after vaginal delivery or >1000 mL after cesarean delivery), chorioamnionitis, and endomyometritis. Neonatal outcomes examined included 5-minute Apgar score <7, umbilical artery (UA) pH <7.0, umbilical artery base deficit ≥-12 mmol/L, shoulder dystocia, birth trauma (a composite variable for brachial plexus injury, facial nerve palsy, clavicular fracture, skull fracture, and head lacerations as diagnosed by the attending pediatrician caring for the neonates) and rates of admission to the neonatal intensive care nursery (NICU) as well as a composite variable, “neonatal complication,” which included 5-minute Apgar score <7, UA pH <7.0, UA base deficit ≥-12 mmol/L, shoulder dystocia, NICU admissions, and birth trauma.
Statistical analysis was performed by using STATA v9.0 (StataCorp, College Station, TX). Length of labor was examined using survival analysis. The length of first stage of labor was compared among women who had induction of labor with women who had spontaneous onset of labor (n = 27,328) and statistical comparison was performed using log-rank test. Maternal and neonatal outcomes were analyzed using the χ2 test for univariable analysis. Of note, although there were fluctuations in cesarean delivery (3-9%) and operative vaginal delivery (3-7%), there were no identifiable trends (P > .05) relative to the year of delivery. Potential confounding variables were controlled for using multivariable logistic regression analysis, with the length of first stage between 0-12 hours designated as the referent. The covariates included in the multivariable regression models included maternal age, parity, gestational age, race/ethnicity, BMI, gestational age at delivery, mode of delivery (for postpartum hemorrhage, endomyometritis, and shoulder dystocia), the length of second stage of labor, epidural anesthesia, premature rupture of the membranes, year of delivery, and delivering faculty. In addition, episiotomy was adjusted for the analyses of perineal lacerations. We also performed subgroup analyses, stratifying the cohort by parity and by use of cervical ripening agents. Of note, we did not have cervical ripening information on 206 women (5.7% of the cohort). Statistical significance was indicated by P value less than .05 and 95% confidence intervals (CIs).
Results
Of the 3620 women meeting study criteria who had undergone induction of labor and delivered in the second stage of labor, the median length of the first stage of labor was 7.58 hours (intraquartile range, 4.27–12.50 hours). Of these, 2602 (71.9%) had a length of first stage between 0-12 hours, 593 (16.4%) had a first stage between 12-18 hours, 244 (6.7%) had a first stage between 18-24 hours, and 181 (5.0%) had a first stage lasting longer than 24 hours. The majority of the study population had indicated inductions. These included postterm inductions (37%), preeclampsia/gestational hypertension (12%), gestational/pregestational diabetes mellitus (5%), premature rupture of the membranes (4%), fetal indications (18%), maternal medical conditions (2%), and others/unknown (22%). Maternal characteristics associated with a longer first stage of labor included nulliparity and gestational age at delivery ≥41 completed weeks (P < .001 for both, Table 1). The utilization of epidural anesthesia during labor was also associated with a longer duration of first stage (P < .001, Table 1), although we did not have information regarding the timing of epidural placement relative to the first stage. We also observed a positive association between the duration of first stage and likelihood of having a prolonged second stage >3 hours and birthweight >4000 g (P < .001 for both, Table 1).
TABLE 1. Factors associated with the length of first stage of labor
| Factors | Length of first stage of labor | |||
|---|---|---|---|---|
| 0-12 h (n = 2602) | 12-18 h (n = 593) | 18-24 h (n = 244) | ≥24 h (n = 181) | |
| Women | ||||
 Nulliparaa (n = 2129) | 64.2% | 20.0% | 9.3% | 6.5% |
| Multiparaa (n = 1491) | 82.6% | 11.3% | 3.2% | 2.9% |
| Gestational age <41 wksa (n = 2179) | 74.2% | 15.7% | 6.0% | 4.1% |
| Gestational age ≥41 wksa (n = 1439) | 68.4% | 17.4% | 7.9% | 6.3% |
| Maternal age <35 y (n = 2028) | 71.4% | 16.7% | 7.0% | 4.9% |
| Maternal age ≥35 y (n = 691) | 74.0% | 14.9% | 5.6% | 5.5% |
| Ethnicity | ||||
| White (n = 1529) | 71.8% | 16.9% | 6.6% | 4.7% |
| African American (n = 546) | 70.9% | 19.2% | 5.3% | 4.6% |
| Latina/Hispanics (n = 457) | 72.0% | 14.7% | 7.9% | 5.5% |
| Asian (n = 830) | 70.6% | 15.2% | 8.1% | 6.1% |
| Others (n = 258) | 78.3% | 14.3% | 4.3% | 3.1% |
| BMI <30 kg/m2 (n = 2079) | 73.0% | 16.8% | 6.3% | 3.9% |
| BMI ≥30 kg/m2 (n = 351) | 69.8% | 17.6% | 8.6% | 4.0% |
| Epidural usea (n = 2628) | 66.0% | 19.8% | 8.2% | 6.0% |
| No epidurala (n = 984) | 87.2% | 7.5% | 3.0% | 2.3% |
| 2nd stage <3 ha (n = 2675) | 85.6% | 75.0% | 67.2% | 75.6% |
| 2nd stage ≥3 ha (n = 561) | 14.4% | 25.0% | 32.8% | 24.4% |
| Birthweight ≤4000 ga (n = 3070) | 86.4% | 81.3% | 82.4% | 76.8% |
| Birthweight >4000 ga (n = 550) | 13.6% | 18.7% | 17.6% | 23.2% |
aP < .001. |
The length of first stage of labor in women who had induction of labor was compared to those who had spontaneous labor using survival analysis. We observed no difference in the length of first stage between the 2 groups. In women who had induction of labor, the median length of first stage was 7.66 hours; it was 7.75 hours for those who had spontaneous labor (P = .29 by log-rank test, Figure 1).
FIGURE.
Kaplan-Meier survival curves
Length of first stage of labor (h) in women with induction of labor and in women with spontaneous labor (P = .29 by log-rank test).
Cheng. Length of first stage and outcomes of induction of labor. Am J Obstet Gynecol 2009.
To remove the potential confounding effect of parity on perinatal outcomes and mode of delivery, we stratified our cohort into nulliparous and multiparous women. In nulliparous women, we observed a positive association between duration of first stage and operative delivery (Table 2). The proportion of cesarean delivery performed during the second stage increased with a longer first stage; this was particularly was true when cesarean was indicated for labor dystocia rather than for fetal intolerance of labor (Table 2). We also observed increased maternal morbidity, birth trauma, and composite neonatal complications with a longer first stage (P < .001 for all, Table 2).
TABLE 2. Subgroup analysis by parity: perinatal outcomes associated with the first stage of labor in nulliparous and multiparous women who had undergone induction of labor and delivered during the second stage
| Nulliparous women | 0-12 h (n = 1369) | 12-18 h (n = 425) | 18-24 h (n = 197) | ≥24 h (n = 138) |
|---|---|---|---|---|
| Mode of deliverya | ||||
| SVD (n = 1200) | 63.3% | 43.8% | 43.6% | 43.5% |
| Op VD (n = 484) | 24.3% | 23.5% | 19.3% | 10.9% |
| Cesarean (n = 445) | 12.4% | 32.7% | 37.1% | 45.6% |
| Indications for cesareana | ||||
| Labor dystocia | 62.2% | 65.9% | 61.0% | 72.0% |
| Fetal intolerance | 21.6% | 15.5% | 25.4% | 20.0% |
| Other | 16.2% | 18.6% | 13.6% | 8.0% |
| 3rd or 4th degree lac (n = 278) | 16.4% | 16.1% | 19.4% | 16.0% |
| PPHa (n = 682) | 29.3% | 34.2% | 43.5% | 46.6% |
| Chorioamnionitisa (n = 317) | 9.9% | 20.7% | 26.9% | 29.0% |
| Endomyometritisa (n = 84) | 2.0% | 8.5% | 3.0% | 10.9% |
| 5-min Apgar <7 (n = 96) | 4.0% | 4.9% | 5.6% | 6.5% |
| Birth traumaab (n = 29) | 0.8% | 2.8% | 3.1% | 0% |
| ICN admission (n = 134) | 5.9% | 6.1% | 6.6% | 10.1% |
| Neonatal compsac (n = 109) | 4.5% | 6.4% | 7.6% | 3.6% |
| Multiparous women | (n = 1233) | (n = 168) | (n = 47) | (n = 43) |
|---|---|---|---|---|
| Mode of deliverya | ||||
| SVD (n = 1198) | 83.5% | 72.0% | 51.1% | 53.5% |
| Op VD (n = 147) | 10.1% | 8.9% | 6.4% | 11.6% |
| Cesarean (n = 146) | 6.4% | 19.1% | 42.5% | 34.9% |
| Indications for cesareana | ||||
| Labor dystocia | 47.7% | 68.2% | 88.9% | 66.7% |
| Fetal intolerance | 38.5% | 27.3% | 5.6% | 20.0% |
| Other indications | 13.8% | 4.5% | 5.5% | 13.3% |
| 3rd or 4th degree lac (n = 50) | 3.5% | 6.6% | 3.7% | 0% |
| PPHa (n = 298) | 18.7% | 20.9% | 36.2% | 42.9% |
| Chorioamnionitisa (n = 75) | 3.3% | 7.1% | 27.7% | 20.9% |
| Endomyometritisa (n = 29) | 1.3% | 3.6% | 10.6% | 4.7% |
| 5-min Apgar <7 (n = 52) | 3.4% | 4.8% | 2.1% | 2.3% |
| Birth traumaa, b (n = 15) | 1.1% | 0% | 0% | 2.3% |
| NICU admissiond (n = 120) | 7.5% | 7.7% | 17.0% | 16.3% |
| Neonatal compsc (n = 50) | 3.4% | 2.4% | 4.3% | 4.6% |
aP < .001; |
bBirth trauma (includes brachial plexus injury, facial nerve palsy, clavicular fracture, skull fracture, head laceration as defined and diagnosed by the attending ediatrician); |
cComposite variable for 5-min Apgar <7, UA pH <7.0, UA base deficit ≥-12 mmol/L, shoulder dystocia, and birth trauma; |
dP < .05. |
Similar association between length of first stage and cesarean delivery was seen in multiparous women who had induction of labor. In our cohort of multiparous women whose first stage lasted longer than 24 hours, the likelihood of SVD was 53%, whereas 12% had operative vaginal deliveries and 35% delivered by cesarean (P < .001, Table 2). They also experienced higher frequency of maternal morbidity (Table 2).
Women whose induction required the use of cervical ripening agents (either prostaglandins or mechanical cervical dilators) may be different from those with a more favorable cervix not requiring cervical ripening agents, particular in nulliparous parturient. In our cohort of nulliparous women who received cervical ripening agents as part of their labor induction process, a longer first stage was associated with a higher likelihood of cesarean delivery during the second stage (Table 3). Maternal complications and birth trauma were higher with increasing length of first stage (Table 3).
TABLE 3. Subgroup analysis by use of cervical ripening agents: perinatal outcomes associated with the first stage of labor in nulliparous women who received cervical ripening agents for labor induction and subsequently delivered during the second stage
| Use of cervical ripening agents for nulliparous women | 0-12 h (n = 740) | 12-18 h (n = 239) | 18-24 h (n = 106) | ≥24 h (n = 85) |
|---|---|---|---|---|
| Mode of delivery | ||||
| SVD (n = 630) | 57.3% | 24.8% | 33.5% | 36.1% |
| Op VD (n = 281) | 29.9% | 38.4% | 30.9% | 19.2% |
| Cesareana (n = 259) | 12.8% | 36.8% | 35.9% | 44.7% |
| 3rd or 4th degree lac (n = 162) | 18.3% | 15.2% | 19.1% | 17.0% |
| PPHb (n = 389) | 30.4% | 34.5% | 43.1% | 53.8% |
| Chorioamnionitisb (n = 195) | 11.6% | 21.8% | 26.4% | 34.1% |
| Endomyometritisa (n = 42) | 2.3% | 7.5% | 1.9% | 5.9% |
| 5-min Apgar <7 (n = 50) | 5.3% | 5.3% | 5.9% | 8.5% |
| Birth traumaa (n = 22) | 1.1% | 4.2% | 3.8% | 0% |
| ICN admission (n = 84) | 6.9% | 7.5% | 6.6% | 9.4% |
| Neonatal compsc, d (n = 75) | 5.7% | 8.4% | 9.4% | 3.5% |
aP < .05; |
bP < .001; |
cBirth trauma (includes brachial plexus injury, facial nerve palsy, clavicular fracture, skull fracture, head laceration as defined and diagnosed by the attending pediatrician); |
dComposite variable for 5-min Apgar <7, UA pH <7.0, UA base deficit ≥-12 mmol/L, shoulder dystocia, and birth trauma. |
Because we did not observe large differences in perinatal outcomes when our cohort was stratified by parity or by cervical ripening agents, we performed multivariable logistic regression analyses for the entire cohort. Designating first stage duration of 0-12 hours as the referent, women whose first stage lasted between 12-18 and 18-24 hours had twice the odds of having cesarean delivery during the second stage, whereas these odds were nearly 8 times higher in women whose first stage lasted longer than 24 hours (Table 4). Women with a first stage of labor lasting ≥24 hours also had increased risk of cesarean delivery as well as 3 times the odds of postpartum hemorrhage (adjusted odds ratio [aOR], 3.16; 95% CI, 1.73–5.81) and increased likelihood of chorioamnionitis (Table 4). Although we did not observe statistically significant differences in 5-minute Apgar scores or composite neonatal complications among the comparison groups, neonates born to women with a first stage between 12-18 hours and 18-24 hours had higher odds of birth trauma and other adverse outcomes (Table 4). Because elapse of time is continuous, we also performed multivariable linear regression analysis using the same covariates as the logistic regression model and observed that progressive length of first stage in women who had induction of labor is associated with higher risk of cesarean delivery, chorioamnionitis, and NICU admissions.
TABLE 4. aOR of perinatal outcomes examined by the length of first stage of labor in women who had induction of labor and delivered during the second stage
| Length of second stagea | 12-18 h | 18-24 h | ≥24 h | |||
|---|---|---|---|---|---|---|
| aOR | 95% CI | aOR | 95% CI | aOR | 95% CI | |
| Mode of deliveryb | ||||||
| Op VDc | 0.97 | 0.69–1.37 | 0.73 | 0.41–1.30 | 0.91 | 0.38–2.15 |
| Cesarean deliveryc | 2.06f | 1.33–3.20f | 2.13f | 1.10–4.09f | 7.44f | 3.43–16.1f |
| Maternal complicationsb | ||||||
| 3rd or 4th degree perineal lac | 0.89 | 0.58–1.35 | 1.08 | 0.59–1.99 | 0.66 | 0.21–2.07 |
| PPH | 0.95 | 0.70–1.28 | 1.19 | 0.75–1.89 | 3.16f | 1.73–5.79f |
| Chorioamnionitis | 1.99f | 1.29–3.07f | 3.75f | 2.12–6.62f | 2.83f | 1.19–6.69f |
| Endomyometritis | 2.21f | 1.06–4.61f | 0.29 | 0.04–2.33 | 1.98 | 0.55–7.11 |
| Neonatal complicationsb | ||||||
| 5-min Apgar <7 | 1.32 | 0.80–2.20 | 1.69 | 0.84–3.38 | 1.35 | 0.57–3.22 |
| Birth traumad | 2.30f | 1.05–5.03f | 3.05f | 1.10–8.43f | 0.88 | 0.12–6.70 |
| NICU admission | 1.08 | 0.66–1.62 | 1.76 | 1.00–3.09 | 2.03f | 1.10–3.74f |
| Neonatal complicationse | 1.16 | 0.70–1.91 | 1.55 | 0.79–3.04 | 0.97 | 0.38–2.49 |
aReference comparison group: white, women with first stage of labor between 0-12 h duration; |
bAdjusting for maternal age, parity, race/ethnicity, BMI, length of the second stage of labor, gestational age at delivery, use of cervical ripening agent, epidural anesthesia, premature rupture of membranes, and year of delivery. In addition, perineal laceration adjusted for episiotomy; PPH, endomyometritis, and birth trauma were adjusted for mode of delivery; perineal laceration was examined only in women with VDs; NICU adjusted for chorioamnionitis and mode of delivery (number of observations in the multivariable regression models: n = 2048 for models that estimated maternal outcomes, n = 1883 for models that estimated perineal lacerations and op VDs (in women with VD only), and n = 3021 for models estimating neonatal outcomes); |
cCompared with spontaneous vaginal delivery; |
dComposite variable for brachial plexus injury, facial nerve palsy, clavicular fracture, skull fracture, and head laceration, defined and diagnosed by the attending pediatrician; |
eComposite variable for 5-min Apgar <7, UA pH <7.0, UA base deficit ≥-12 mmol/L, shoulder dystocia, and birth trauma; |
fStatistical significance. |
Comment
We report that in women who had induction of labor, a longer length of the first stage of labor is associated with a higher risk of cesarean delivery during the second stage. This association was seen for both nulliparous and multiparous women and regardless whether cervical ripening agents were used as part of the induction process. However, despite the increased risk, more than half of these women with a first stage longer than 24 hours still achieved vaginal delivery. Thus, we do not advocate surgical intervention simply based on the elapse of time during labor.
Many studies have identified characteristics such as increasing maternal age, gestational age, BMI, and birthweight as risk factors associated with cesarean delivery in either spontaneous labor or induction of labor.6, 7, 10, 11 Cervical status assessed by the Bishop pelvic scoring system is particularly important in labor induction.6, 10 In our study, we observed that in nulliparous women who did receive cervical ripening agents as part of the induction process, the duration of first stage was positively associated with cesarean delivery. Likely, factors such as fetal occiput position, uterine contractile strength, or chorioamnionitis contribute to the length of first stage and influence mode of delivery during the second stage of labor.
It is not surprising that women with a longer first stage of labor are at higher risk of having postpartum hemorrhage and chorioamnionitis. We also observed that first stage durations of 12-18 hours and 18-24 hours are associated with higher odds of birth trauma compared to 0-12 hours, even after controlling for mode of delivery and duration of second stage. In women who eventually had cesarean deliveries, the majority were performed for labor dystocia. Although the risk of intensive care nursery admissions was higher for those with a first stage greater than 18 hours, it may be related to sepsis workups in the setting of chorioamnionitis. The association between the length of first stage and neonatal outcomes should be taken into consideration when managing women undergoing induction of labor. This includes reassessment of indications of induction as well as the risks and benefits of continuing induction.
The challenge of labor induction extends beyond “ripening” an unfavorable cervix and stimulating uterine contractions to achieve active labor. We observe that even when complete cervical dilation is achieved, the risk of cesarean delivery and perinatal complications remains increased in women whose first stage is longer. Although current practice guidelines established by the ACOG on induction of labor do not specify a time limit for induction of labor nor is the definition of a “failed induction” clearly stated,1 we also do not recommend terminating induction of labor when a first stage progresses beyond 24 hours.
Our study has limitations. Although we were able to examine use of cervical ripening agents as a marker for cervical status, the initial cervical examination at time of labor induction was not available. But when we stratified our cohort by cervical ripening agents, we did observe similar associations for both subgroups. In addition, we did not have information regarding the exact time in which women enter the active phase of labor. Thus, we did not examine the length of the first stage by latent vs active phase separately. Because women having indicated inductions are likely different from those choosing elective induction, our study results may not apply to women undergoing elective inductions. Also, some of the neonatal outcomes of interest, such as birth trauma or umbilical arterial pH <7.0, were rare events, thus the absence of association may be a result of insufficient statistical power. Future studies would benefit from multicenter, prospective design with adequate power to examine the more rare outcomes in this setting. Finally, as a retrospective study, our effect estimates may be prone to missing data bias or confounding bias. Although we attempted to control for potential confounders by using statistical techniques, there may be residual confounding that we did not identify or could not observe and control for.
This study examined mode of delivery and perinatal outcomes associated with the duration of first stage in women who had an induction of labor. In our cohort, women with a first stage lasting beyond 12 hours have higher odds of cesarean delivery, though the majority still deliver vaginally even beyond 24 hours. Thus, by being patient in the management of first stage of labor, the number needed to allow progression of labor to achieve an additional vaginal delivery is less than 2. Although we do not recommend a specific time limit for labor induction, the question of when to intervene should involve a thorough evaluation of the ongoing risks of further expectant management vs the risks of surgical intervention as well as the incorporation of patient preferences into the decision-making process.
References
- . Clinical management guidelines for obstetricians-gynecologists. Induction of labor. ACOG practice bulletin no. 19. Washington, DC: The College; 1999;
- . Births: final data for 2004. Natl Vital Stat Report. 2006;55:1–101
- . Trends in births at and beyond term: evidence of a change?. BJOG. 1999;106:937–942
- . Outcome after elective labor induction in nulliparous women: a matched cohort study. Am J Obstet Gynecol. 2002;186:240–244
- . Induction of labor and the relationship to cesarean delivery: a review of 7001 consecutive inductions. Am J Obstet Gynecol. 1999;180:628–633
- . Bishop score and risk of cesraen delivery after induction of labor in nulliparous women. Am J Obstet Gynecol. 2005;105:690–697
- . Risks of induction of labour in uncomplicated pregnancies. Paediatr Perinat Epidemiol. 2001;15:131–139
- . Labor progression and risk of cesarean delivery in electively induced nulliparas. Obstet Gynecol. 2005;105:698–704
- . Pelvic scoring for elective induction. Obstet Gynecol. 1964;24:266–268
- . Factors associated with the length of the latent phase during labor induction. Eur J Obstet Gynecol Reprod Biol. 2006;132:163–166
- . The association of maternal weight with cesarean risk, labor induction and cervical dilation rate during labor induction. Obstet Gynecol. 2004;103:452–456
Cite this article as: Cheng YW, Delany SS, Hopkins LM, et al. The association between the length of first stage of labor, mode of delivery, and perinatal outcomes in women undergoing induction of labor. Am J Obstet Gynecol 2009;201:477.e1-7.
Reprints not available from the authors.
Support provided by the Robert Wood Johnson Foundation for A.B.C. as a Physician Faculty Scholar.
PII: S0002-9378(09)00532-8
doi:10.1016/j.ajog.2009.05.024
© 2009 Mosby, Inc. All rights reserved.
Volume 201, Issue 5 , Pages 477.e1-477.e7, November 2009
