Materials and Methods
Between July 1984 and June 1996, we reviewed medical records of gravidas with histories of cesarean delivery who intended a trial of labor at or after 24 weeks' gestation at the Brigham & Women's Hospital. Specific details regarding the identification of the population and review of the charts were reported.3
For the current analysis, the study population was limited to women with term pregnancies who had a single previous cesarean and no other deliveries. We included patients with low transverse (Kerr), low vertical (Kronig), and unknown hysterotomies. We determined the rates of symptomatic uterine rupture and cesarean birth for term deliveries before or at 280 days' gestation (40 weeks) compared with those after 40 weeks' gestation. In addition to examining the overall association, we did separate analyses for spontaneous and induced labor because induction was associated with higher rates of uterine rupture and might be associated with increased risk of cesarean delivery. Induction was defined as the initiation of regular uterine contractions through use of oxytocin, prostaglandin E2 gel, or both.
Outcomes of interest were symptomatic uterine rupture and rate of cesarean delivery. Uterine rupture was defined as complete disruption of the layers of the uterus in association with intraperitoneal or vaginal hemorrhage, need for hysterectomy, bladder injury caused by uterine scar disruption, extrusion of any portion of the fetal–placental unit, cesarean for nonreassuring fetal heart tracing, or suspected uterine rupture.
Statistical significance for comparisons of categorical data was evaluated using χ2 or Fisher exact tests. P < .05 was considered statistically significant.
Multiple logistic regression was used to examine the association of delivery after term with the rate of uterine rupture and control for confounding factors. A similar model was constructed to analyze the rate of cesarean. Potential confounders incorporated into the model were birth weight greater than 4000 g and indication for prior cesarean delivery. Categories for prior cesarean indication were breech, failure to progress, nonreassuring fetal status, and “other.” In the logistic regression, prior indication was modeled as three indicator variables with prior cesarean for breech as the referent group. The association was examined overall and separate analyses were done for spontaneous and induced labors. Adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were calculated.
Approval was obtained from the Institutional Review Board at Brigham and Women's Hospital to abstract data from charts of all women attempting a trial of labor after previous cesarean for the analysis of labor outcomes.
Our population included 2775 women at term who had one prior scar and no other deliveries. There were 1504 women who delivered between 37 and 40 weeks' gestation and 1271 women who delivered after 40 weeks. Of the 1504 gravidas with deliveries before or at 40 weeks, 1214 (81%) labored spontaneously and 290 (19%) were induced. Of the 1271 women beyond 40 weeks, 1001 (79%) had spontaneous labor and 270 (21%) were induced. Characteristics of the study population showed no differences in proportion of women receiving public aid, proportion of white women, and proportion of women laboring with previous Kerr hysterotomies among those delivered at or before 40 weeks compared with those delivered after 40 weeks (Table 1). However, mean age of those undergoing a trial of labor after 40 weeks was younger.
Symptomatic uterine rupture occurred in 29 women. Overall, uterine ruptures occurred in 0.8% of women delivering between 37 and 40 weeks' gestation and 1.3% of women delivering after 40 weeks (P = .2, Table 2). For spontaneous labor, uterine ruptures occurred in 0.5% of gravidas delivering at or before 40 weeks compared with 1.0% delivering after 40 weeks (P = .2). For induced labor, rates of uterine rupture were 2.1% for gravidas at or before 40 weeks and 2.6% for those after 40 weeks (P = .7).
Logistic regression analysis controlling for birth weight and indication for prior cesarean delivery showed that gestational age beyond 40 weeks was not an independent significant predictor of symptomatic uterine rupture for women with spontaneous (adjusted OR 2.1, CI 0.7, 5.7) or induced (adjusted OR 1.1, CI 0.4, 3.4) labor.
Cesarean delivery rates were calculated. Overall, rate of cesarean delivery was higher for women after 40 weeks' gestation (35.4% compared with 26.7%, P < .001, Table 2). This higher rate of cesarean after 40 weeks was present for women with spontaneous labor (25.0% compared with 33.5%, P = .001) and induced labor (33.8% compared with 43.0%, P = .03).
Logistic regression analysis controlling for indication for previous cesarean and birth weight indicated that gestational age after 40 weeks was an independent predictor of cesarean during a subsequent trial of labor for spontaneous (adjusted OR 1.5, CI 1.2, 1.8) and induced (adjusted OR 1.5, CI 1.1, 2.2) labor. In that model, birth weight over 4000 g was a significant predictor of rate of cesarean for spontaneous labor (adjusted OR 1.5, CI 1.1, 2.0) but not for women with induced labor (adjusted OR 1.4, CI 0.8, 2.2). Lack of statistical significance for women with induced labor likely reflected the smaller numbers of women in that group.
As term approaches, clinicians must determine the consequences of waiting for spontaneous labor compared with induction before 40 weeks. To determine the best treatments, we directly compared the rates of cesarean and uterine rupture for women induced at or before 40 weeks with the outcome in women with spontaneous or induced labor after 40 weeks. In our population, induction before 40 weeks associated with a cesarean rate (33.8%) lower than the cesarean rate after 40 weeks (43.0%, P = .03) but similar to the cesarean rate for spontaneous labor after 40 weeks (33.5%, P = .9). In contrast, induction of labor before 40 weeks associated with a rate of uterine rupture (2.1%) similar to that with induction after 40 weeks (2.6%, P = .7) but higher than the rupture rate associated with spontaneous labor after 40 weeks (1.0%, P = .2). The nonsignificant difference might have been caused by lack of power resulting from the rarity of rupture.
Counseling with respect to patient selection for VBAC evolves during the course of the pregnancy. As Callahan et al2 suggested, even the most committed patient–physician team may waver in their decision to attempt VBAC as the woman passes her due date.
Our study has several clinically relevant implications for that decision in women with one prior cesarean and no other deliveries. As term approaches, a decision must be made whether to induce or allow the pregnancy to progress beyond 40 weeks. Our data suggest that waiting for onset of spontaneous labor might be the more prudent option in the absence of other risk factors mandating delivery because the risk of symptomatic uterine rupture does not increase substantially for women with spontaneous onset of labor after 40 weeks. In contrast, induction of labor was associated with increased rate of uterine rupture.4
In addition, although the success of a trial of labor was lower for women with spontaneous labor after 40 weeks compared with spontaneous labor before or at 40 weeks, awaiting the spontaneous onset of labor rather than inducing does not decrease the chances of successful vaginal delivery. Women induced before or at 40 weeks had a 33.8% cesarean rate, similar to the 33.5% rate for women with spontaneous labor after 40 weeks.
Induced labor before 40 weeks was associated with 2.1% risk of uterine rupture, whereas spontaneous labor after 40 weeks was associated with only 1% risk (P = .2). Although not statistically significant, this difference might be due to type II error. Despite our robust sample size, we had only 25% power of detecting a difference of this magnitude. A sample size of 6200 women would be required to answer this question.
If the gravida does not go into spontaneous labor after 40 weeks, the choices for the woman and her physician become more difficult. In our study, induction beyond 40 weeks was associated with higher cesarean delivery rate (43%) and higher risk of uterine rupture (2.6%). Physicians and their patients must consider the risks and benefits of a trial of labor compared with elective repeat cesarean.
Our study has several limitations. Patients were managed by many attending physicians who employed a variety of management styles. Obstetric practices might have changed over the 12 years of the study, affecting our results. For example, patients who attempted labor post-term had a different (higher or lower) baseline risk of rupture or cesarean delivery. Although we controlled for confounding factors in our analyses, it is not possible to rule out the presence of residual confounding. A prospective study that collected extensive information on the baseline risk would be useful to sort out these issues. Such a study is currently being done by the National Institute of Child Health and Human Development, but the results are years away. In the meantime, our study can provide the basis for informed discussions regarding the safety and efficacy of a trial of labor after 40 weeks.
1. American College of Obstetricians and Gynecologists. Vaginal birth after previous cesarean section. ACOG practice bulletin no. 5. Washington, DC: American College of Obstetricians and Gynecologists, 1998.
2. Callahan C, Chescheir N, Steiner BD. Safety and efficacy of attempted vaginal birth after cesarean beyond the estimated date of delivery. Reprod Med 1999;44:606–10.
3. Shipp TD, Zelop CM, Repke JT, Cohen A, Caughey AB, Lieberman E. Intrapartum uterine rupture and dehiscence in patients with prior lower uterine segment vertical and transverse incisions. Obstet Gynecol 1999;94:735–40.
4. Zelop CM, Shipp TD, Repke JT, Cohen A, Caughey AB, 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.