Recently, trials of labor after cesarean have become less frequent, as an increasing number of studies that have focused on adverse outcomes such as uterine rupture and perinatal morbidity have been published.1 For example, Smith et al2 estimated that the relative risk of perinatal death among women having a trial of labor was 11 times that of women having a planned repeat cesarean delivery. Similarly, investigators have sought to elucidate factors that particularly increase risk during a trial of labor. One factor that has been associated with increased risk of uterine rupture and increased maternal and neonatal morbidity has been labor induction, a practice that has become increasingly common.3–9 Yet, although the relative risk for neonatal morbidity associated with a trial of labor may be increased in relation to an elective cesarean, the attributable risk remains small, and many patients and physicians believe that an attempt at a vaginal delivery, particularly if the chances of a vaginal delivery are substantial, is a worthwhile endeavor.
Little information is available, however, that specifies factors predictive of vaginal delivery in the context of an induction in a woman with a prior uterine scar. These factors could aid the clinician in the selection and counseling of candidates for trial of labor. Although several studies have attempted to characterize obstetric and demographic variables that influence the success of a trial-of-labor outcome, these studies did not analyze the factors that are relevant in the setting of induction, specifically.4–6,10–13 Consequently, the present investigation was undertaken in an effort to identify and characterize the associations between various obstetric factors and trial-of-labor outcome in women undergoing induction of labor with a history of cesarean delivery.
MATERIALS AND METHODS
All patients who presented to labor and delivery at Northwestern Memorial Hospital between January 1996 and August 2001 for induction of labor with a history of cesarean delivery were identified from the perinatal database. The medical records of these women were reviewed by a single physician (J.G.) and were included in the final analysis if they met inclusion criteria. Inclusion criteria were: singleton pregnancy, vertex presentation, history of 1 cesarean delivery, presentation for a scheduled induction of labor, intact membranes, absence of labor (uterine contractions less than every 10 minutes), and gestational age of more than 34 weeks.
Once the study group was selected, demographic and intrapartum data were abstracted from the medical records. These data included maternal age, race, height, weight, a detailed obstetric history including indication for the first cesarean, the indication for the current induction of labor, cervical dilation, effacement, and station at the time of admission, and the planned method of labor induction. In addition, the outcome of each trial of labor and corresponding maternal and neonatal morbidity data were recorded.
The SPSS statistical package (SPSS, Chicago, IL) was used for statistical analysis. The association of variables under study with trial of labor outcome (vaginal or cesarean delivery) was first assessed with univariable analysis; the χ2 test was used to analyze categorical variables, and the t test was used for continuous variables. All tests were 2-tailed. Predictors significant at P < .05 in univariable analysis were then evaluated by using binary logistic regression modeling, and those which continued to be significant at P < .05 in the multivariable analysis were kept in the final model. Descriptive statistics for the maternal and neonatal morbidity associated with an induced trial of labor are also presented.
Approval for this study was obtained from the Northwestern University Institutional Review Board.
A total of 620 patients presented during the study period for scheduled induction of labor with a history of cesarean. Of these patients, 191 did not meet inclusion criteria because of a prior cesarean, multiple gestation, estimated gestation age less than 34 weeks of gestation, or rupture of membranes before the induction began, and thus were excluded. Of the 429 women in the study population, 334 (77.9%) had a successful vaginal birth after cesarean and 95 (22.1%) underwent a repeat cesarean. Of the patients who delivered vaginally, 47 (14.1%) had an instrumental delivery. Cervical ripening was required in 96 instances; 83 women received an extraamniotic saline infusion, whereas 12 women received osmotic dilators, and 1 woman received prostaglandin. Inductions were initiated at the discretion of the attending physician; in 139 cases (32%), the inductions were medically indicated. In the remaining 290 cases (68%), elective inductions were performed. The elective inductions were typically in women with a ripe cervix, although in 31 cases women who underwent elective inductions also needed cervical ripening. Further characteristics of the population, stratified by cases and controls, and the variables significantly associated with eventual vaginal delivery, are shown in Table 1 and Table 2.
Logistic regression analysis was performed to evaluate the effect of each independent variable on the trial-of-labor outcome while controlling for confounding factors, and the results are summarized in Table 3. Women who achieved vaginal birth after an induced trial of labor were more likely to have had a prior vaginal delivery (odds ratio [OR] 3.75; 95% confidence interval [CI] 1.96, 7.18). Conversely, a labor dystocia as the reason for the primary cesarean (OR 0.46; 95% CI 0.27, 0.79), admission for induction on or past their estimated date of delivery (OR 0.46; 95% CI 0.27, 0.78), need for cervical ripening (OR 0.35; 95% CI 0.20, 0.61), or the presence of gestational or preexisting diabetes (OR 0.16; 95% CI 0.06, 0.40), all increased the chance that the trial of labor would result in a repeat cesarean.
There were no maternal or perinatal deaths in the study population. Twelve neonates (2.8%) had a 5-minute Apgar score lower than 7 at 5 minutes, and no neonate had a 5-minute Apgar of 3 or lower. Five neonates (1.3%, n = 387) had an arterial cord gas lower than 7.0. Neonatal admission to the intensive care unit occurred in 23 cases (5.4%) and was more common after a failed trial of labor than after vaginal delivery (11 [11.6%] versus 12 [3.6%] cases, P < .01). There were 33 cases (7.7%) of chorioamnionitis and 8 cases (1.9%) of postpartum endometritis.
There were 5 instances of asymptomatic uterine dehiscence (1.2%), defined as a separation of the uterine scar not associated with a nonreassuring fetal heart rate pattern or maternal hemodynamic changes. One case was noted during routine exploration of the uterus after spontaneous vaginal delivery, and the other 4 cases were noted at the time of cesarean performed for arrest of dilation. There were also 4 (0.9%) cases of symptomatic uterine rupture among the study population. Two of these cases were diagnosed after operative vaginal delivery, and 2 diagnoses were made at the time of cesarean delivery for a nonreassuring fetal heart tracing. Among the 4 cases of uterine rupture, there was 1 hysterectomy performed and 3 admissions to the intensive care unit (2 maternal and 1 neonatal). One rupture was associated with prostaglandin use for cervical ripening.
Since 1996, the national cesarean delivery rate has been steadily rising. Centers for Disease Control data from 2000 report an overall rate of 22.9%.1 Factors contributing to this trend are not completely understood and there remains wide variation among practitioners, hospitals, and geographic regions of the United States. Nevertheless, one of the results is an increasing number of women presenting for obstetric care with a history of cesarean delivery. Although the relative safety and efficacy of vaginal birth after cesarean has been repeatedly documented, there has been a growing concern among patients and physicians of the inherent risks associated with trial of labor, particularly in the setting of an induction.9
Induction of labor has been implicated by several groups as a risk factor for failed trial of labor.4–6 In addition, it may also be a risk factor for maternal and neonatal morbidity. Zelop et al3 suggested that among women with a history of 1 cesarean delivery, induction of labor with oxytocin is associated with a 4.6-fold increased risk of uterine rupture when compared with spontaneous labor. As the rate of labor induction in the United States has more than doubled over the last decade, obstetricians are increasingly faced with the clinical question of whether to proceed with an elective repeat cesarean or induce labor in the setting of a scarred uterus. Thus, in women who need to undergo induction of labor, the ability to predict which women are more likely to have a successful trial of labor would be particularly useful.
To date, several groups have attempted to identify characteristics associated with successful trial-of-labor outcome.4–6,10–13 However, none of these studies has focused exclusively on patients undergoing labor induction. Induced labor is known to have different characteristics than spontaneous labor and results in a higher rate of cesarean delivery.14 Therefore, it is not clear that the predictive variables identified in prior studies apply to trial of labor in patients undergoing labor induction. Using logistic regression, we found several significant predictors could be identified that were associated with subsequent vaginal delivery. Specifically, history of vaginal delivery, nonrecurring indication for the prior cesarean delivery, induction before the estimated due date, a ripe cervix on admission, and no maternal history of diabetes all increased the chance of a vaginal delivery after an induced trial of labor.
Morbidity in our study population was relatively low, with no maternal or perinatal deaths, although our power to detect small increases in these possibilities is limited. The rate of symptomatic uterine rupture was 0.9%, more consistent with the rate of rupture among spontaneously laboring patients after 1 prior cesarean delivery than the rate of 2.3% noted by Zelop et al3 among patients with induced or augmented labor and 1 prior cesarean delivery. However, Zelop et al3 excluded women with a prior vaginal delivery, and this exclusion may have resulted in a group at relatively high risk of uterine rupture; these investigators have also demonstrated patients with a prior vaginal delivery are less likely to suffer a uterine rupture.15 In our study, 40% of patients had a prior vaginal delivery, and 92% of the women in this group had a successful trial-of-labor outcome. No patient with a prior vaginal delivery experienced uterine dehiscence or symptomatic uterine rupture.
Based on our logistic-regression analysis, physicians can gain insight into the factors that increase the chance of a successful induction of labor and can impart this insight to their patients during counseling. In this way, patients could be better able to make decisions based on their individual characteristics. Ultimately, as with a nonscarred uterus, spontaneous labor is most likely to result in a vaginal delivery.14 Nevertheless, if an induction of labor is indicated in a patient with 1 prior cesarean, the available literature, including data from our study, support a reasonable likelihood of successful vaginal delivery, particularly when certain characteristics are taken into account. Induction of labor in this setting may not be contraindicated provided that close patient monitoring and the ability to perform emergency cesarean delivery are available.
1. Menacker F, Curtin S. Trends in cesarean birth and vaginal birth after previous cesarean, 1991–99. Nat Vital Stat Rep 2001;49:1–15.
2. Smith F, Pell J, Cameron A, Dobbie R. Risk of perinatal death associated with labor after previous cesarean delivery in uncomplicated term pregnancies. JAMA 2002;287:2684–9.
3. Zelop C, Shipp T, Repke J, Cohen A, Caughey A, Lieberman E. Uterine rupture during induced or augmented labor in gravid women with one prior cesarean delivery. Am J Obstet Gynecol 1999;181:882–6.
4. Pickhardt MG, Martin JNJr, Meydrech EF, Blake PG, Martin RW, Perry KGJr, et al. Vaginal birth after cesarean delivery: are there useful and valid predictors or success or failure? Am J Obstet Gynecol 1992;166:1811–9.
5. Troyer LR, Parisi VM. Obstetric parameters affecting success in a trial of labor: designation of a scoring system. Am J Obstet Gynecol 1992;167:1099–104.
6. Learman LA. Predictors of repeat cesarean delivery after trial of labor: do any exist? J Am Coll Surg 1996;182:257–62.
7. Martin J, Hamilton B, Ventura S, Menacker F, Park M. Births: final data for 2000. Nat Vital Stat Rep 2002;50:13–4.
8. Rosen MG, Dickinson JC. Vaginal birth after cesarean: a meta-analysis of indicators for success. Obstet Gynecol 1990;76:865–9.
9. Lydon-Rochelle M, Hold VL, Easterling TR, Martin DP. Risk of uterine rupture during labor among women with a prior cesarean delivery. N Engl J Med 2001;345:3–8.
10. Silver RK, Gibbs RS. Predictors of vaginal delivery in patients with a previous cesarean section who require oxytocin. Am J Obstet Gynecol 1987;156:57–60.
11. Whiteside DC, Mahan CS, Cook JC. Factors associated with successful vaginal delivery after cesarean section. J Reprod Med 1983;28:785–8.
12. Weinstein D, Benshushan A, Tanos V, Zilberstein R, Rojansky N. Predictive score for vaginal birth after cesarean section. Am J Obstet Gynecol 1996;174:192–8.
13. Flamm BL, Geiger AM. Vaginal birth after cesarean delivery: an admission scoring system. Obstet Gynecol 1997;90:907–10.
14. Seyb ST, Berka RJ, Socol ML, Dooley SL. Risk of cesarean delivery with elective induction of labor at term in nulliparous women. Obstet Gynecol 1999;94:600–7.
© 2004 The American College of Obstetricians and Gynecologists
15. Zelop CM, Shipp TD, Repke JT, Cohen A, Lieberman E. Effect of previous vaginal delivery on the risk of uterine rupture during a subsequent trial of labor. Am J Obstet Gynecol 2000;183:1184–6.