Objective
We sought to describe labor patterns in women with a trial of labor after cesarean (TOLAC) with normal neonatal outcomes.
Study Design
In a retrospective observational study at 12 US centers (2002 through 2008), we examined time interval for each centimeter of cervical dilation and compared labor progression stratified by spontaneous or induced labor in 2892 multiparous women with TOLAC (second delivery) and 56,301 nulliparous women at 37 0/7 to 41 6/7 weeks of gestation. Analyses were performed including women with intrapartum cesarean delivery, and also limiting only to women who delivered vaginally.
Results
Labor was induced in 23.4% of TOLAC and 44.1% of nulliparous women ( P < .001). Cesarean delivery rates were 57.7% in TOLAC vs 19.0% in nulliparous women ( P < .001). Oxytocin was used in 52.4% of TOLAC vs 64.3% of nulliparous women with spontaneous labor ( P < .001) and 89.8% of TOLAC vs 91.6% of nulliparous women with induced labor ( P = .099); however, TOLAC had lower maximum doses of oxytocin compared to nulliparous women: median (90th percentile): 6 (18) mU/min vs 12 (28) mU/min, respectively ( P < .001). Median (95th percentile) labor duration for TOLAC vs nulliparous women with spontaneous labor from 4-10 cm was 0.9 (2.2) hours longer ( P = .007). For women who entered labor spontaneously and achieved vaginal delivery, labor patterns for TOLAC were similar to nulliparous women. For induced labor, labor duration for TOLAC vs nulliparous women from 4-10 cm was 1.5 (4.6) hours longer ( P < .001). For women who achieved vaginal delivery, labor patterns were slower for induced TOLAC compared to nulliparous women.
Conclusion
Labor duration for TOLAC was slower compared to nulliparous labor, particularly for induced labor. By improved understanding of the rates of progress at different points in labor, this new information on labor curves in women undergoing TOLAC, particularly for induction, should help physicians when managing labor.
Cesarean delivery accounted for 32.8% of deliveries in 2012. In a prior Consortium on Safe Labor (CSL) study, the most common reason for cesarean was elective repeat due to a previous uterine scar, accounting for 30.9% of all cesarean deliveries. There has been a national interest in increasing the vaginal birth after cesarean (VBAC) rate in women with a prior low transverse cesarean delivery to decrease the overall cesarean rate. The rate of VBAC started to decline in 1996, prompting a 2010 consensus conference by the National Institutes of Health. After review of the available data, the conference concluded that VBAC was a safe option for many women with a prior low transverse cesarean delivery. A recent American Congress of Obstetricians and Gynecologists (ACOG) practice bulletin has revisited VBAC and emphasized the importance of discussing VBAC with all patients who meet criteria for it. Prior cesarean delivery may be a marker of dysfunctional labor, so it is important to understand labor patterns in subsequent deliveries. Exploring labor patterns among these women may be clinically useful for counseling as well as to guide clinical management during the course of labor among women attempting a VBAC.
Data on labor patterns for trial of labor after cesarean (TOLAC) are limited to single institutions with small numbers, with the older studies conducted prior to the use of modern statistical methods. In addition, labor patterns in women undergoing induction of labor with prior uterine scar have not been studied. The objectives of this study were to compare spontaneous and induced labor characteristics for women with normal neonatal outcomes undergoing TOLAC who had 1 prior cesarean and no vaginal deliveries to nulliparous women in labor, and to compare the course of labor for women who achieved vaginal delivery (eg, having a successful VBAC).
Materials and Methods
The CSL was a Eunice Kennedy Shriver National Institute of Child Health and Human Development multicenter collaborative study designed to characterize labor and delivery in a contemporary US obstetrical clinical practice. The CSL included 12 clinical centers (19 hospitals) spanning 9 ACOG districts from 2002 through 2008. Detailed information was obtained from electronic medical records on maternal demographics, medical history, reproductive and prenatal history, labor and delivery summary, and postpartum and newborn information. Newborn records were linked to information from the neonatal intensive care unit (NICU). Labor progression data including date and time of repeated cervical exams were extracted from the electronic labor database. Oxytocin data included date and start of medication, and starting and maximum doses. Data transferred from the clinical centers were mapped to predefined common categories for each variable at a data-coordinating center. Data cleaning, inquiries, recoding, and logic checking were performed. Validation of data was performed for 4 important outcome diagnoses: cesarean delivery for nonreassuring fetal heart rate tracing, neonatal asphyxia, NICU admission for respiratory conditions, and shoulder dystocia. Data electronically transferred from the medical records were highly concordant with data that were hand abstracted from the records (>95% for all except for 1, 91.1% for clinical diagnosis of shoulder dystocia). Institutional review board approval was obtained at all participating institutions and the data-coordinating center as listed in the Acknowledgments section. Since this study represented a retrospective review of electronic medical records, it was classified as exempt by the Office of Human Subjects Research at the National Institutes of Health.
There were 228,438 deliveries in the CSL. For this analysis, we limited it to woman’s first pregnancy in the dataset (n = 208,695), singleton gestations (n = 203,999), delivering between 37 0/7 and 41 6/7 weeks of gestation (n = 178,582) with vertex presentation (154,894), and with either spontaneous or induced labor (n = 141,919). Fetal anomalies (n = 7616) and antepartum stillbirths (n = 160) were excluded (remaining n = 134,143). We also excluded labor that resulted in uterine rupture (n = 60) to describe labor patterns without this complication. We further limited the study sample to exclude neonates with a 5-minute Apgar score <7, who sustained a birth injury, or who were admitted to the NICU (n = 125,096) as was previously done in the primary CSL labor patterns study. There were 2892 multiparous women (parity = 1) undergoing TOLAC with 1 prior cesarean delivery and no prior vaginal deliveries and 56,301 nulliparous women who comprised the final study sample.
Statistical analysis
Demographics were compared between women undergoing TOLAC and nulliparous women using χ 2 test for categorical variables or Wilcoxon rank sum test for continuous variables. A subanalysis of 6 hospitals where specific oxytocin dosing information was available was also performed to compare starting and maximum doses of oxytocin as well as cervical dilation at oxytocin start. Two analyses were conducted to compare labor progression. First, we examined the pattern of labor by investigating the relationship between duration of labor and cervical dilation only for women with a vaginal delivery. We limited the analysis to women with a vaginal delivery to first evaluate the labor patterns in women who achieved a successful VBAC and to replicate the labor analysis performed in the original CSL study. A repeated measures regression with a polynomial function was used to model the curve of cervical dilation. Second, we performed an analysis comprising all women attempting TOLAC, which included women with an intrapartum cesarean delivery, and examined the interval-censored time interval of cervical dilatation from 1 cm to the next by calculating median (95th percentile) traverse times (hour) for women undergoing TOLAC vs nulliparous women as previously described. P values were obtained from a censored regression adjusting for maternal age, race, body mass index (BMI) at delivery, insurance, epidural use, and oxytocin. Multiple imputation was performed for missing admission BMI (n = 10,174) and maternal age (n = 71) using prepregnancy BMI, parity, race, insurance, smoking, diabetes, hypertension, and site with the MICE approach in R, Version 3.0.2 (R Foundation for Statistical Computing, Vienna, Austria). The duration of second stage labor was compared by using a Cox regression model with the same covariates listed above for the interval-censored models. The rest of the statistical analyses were performed using SAS, version 9.3 (SAS Institute Inc, Cary, NC).
Results
Compared to nulliparous women, women undergoing TOLAC were older, less likely to be white, and had higher prepregnancy BMI ( P < .001 for these variables) ( Table 1 ). Gestational age at delivery was slightly earlier for women undergoing TOLAC (39.1 vs 39.4 weeks, P <. 001), but there were no clinically meaningful differences in median cervical dilation (3 cm), effacement (80%), or fetal station (–2) upon admission. Women undergoing TOLAC when compared to nulliparous women were less likely to be induced (23.4% vs 44.1%, P < .001), have an epidural (47.9% vs 58.6%, P < .001), and have oxytocin augmentation in spontaneous (52.4% vs 64.3%, P < .001) and induced (89.8% vs 91.6%, P < .001) labor.
| Characteristic | TOLAC a (n = 2892) | Nulliparous (n = 56,301) | P value |
|---|---|---|---|
| Age, y, mean ± SD | 28.3 ± 5.7 | 24.9 ± 5.9 | < .001 |
| Race, % | < .001 | ||
| White | 43.0 | 49.7 | |
| Black | 23.7 | 21.0 | |
| Hispanic | 20.3 | 16.7 | |
| Asian/Pacific Islander | 5.3 | 5.2 | |
| Other/unknown | 7.6 | 7.4 | |
| Prepregnancy BMI, kg/m 2 , mean ± SD | 26.5 ± 6.7 | 24.2 ± 5.6 | < .001 |
| Admission BMI, kg/m 2 , mean ± SD | 31.9 ± 6.5 | 30.1 ± 5.9 | < .001 |
| Gestational age at delivery, wk, mean ± SD | 39.1 ± 1.1 | 39.4 ± 1.3 | < .001 |
| Preexisting diabetes, % | 1.6 | 0.7 | < .001 |
| Gestational diabetes, % | 4.9 | 3.6 | < .001 |
| Chronic hypertension, % | 2.1 | 1.1 | < .001 |
| Gestational hypertension, % | 1.8 | 3.7 | < .001 |
| Preeclampsia/HELLP, % | 3.4 | 5.0 | < .001 |
| Eclampsia, % | 0.1 | 0.1 | .439 |
| Cervical dilation at admission, cm, median (10th, 90th percentile) | 3 (0.5, 6) | 3 (1, 6) | .004 |
| Cervical effacement at admission, %, median (10th, 90th percentiles) | 80 (30, 100) | 80 (50, 100) | < .001 |
| Station at admission, %, median (10th, 90th percentiles) | –2 (–3, 0) | –2 (–3, 0) | < .001 |
| Epidural, % | 47.9 | 58.6 | < .001 |
| Induction, % | 23.4 | 44.1 | < .001 |
| Oxytocin for spontaneous labor, % | 52.4 | 64.3 | < .001 |
| Oxytocin for induced labor, % | 89.8 | 91.6 | .099 |
| Cesarean delivery, % | 57.7 | 19.0 | < .001 |
| Estimated blood loss, mL, mean ± SD | 574 ± 273 | 426 ± 304 | < .001 |
a Women (parity = 1) with 1 prior cesarean delivery and no prior vaginal deliveries attempting vaginal birth after cesarean.
For women presenting in spontaneous labor who had vaginal delivery, labor patterns were similar for VBAC and nulliparous women ( Figure 1 ). In the full cohort including women who had an intrapartum cesarean delivery, duration of labor from 4-10 cm for women undergoing TOLAC overall was a median of 0.9 hours (54 minutes) longer, with 95th percentile difference of 2.2 hours longer, and significant differences from 4-5 cm and 6-7 cm ( Table 2 ). Median second stage of labor was slightly shorter (0.1 hour, or 6 minutes) for women undergoing TOLAC, regardless of epidural status.
| Interval, cm | TOLAC, a h Median (95th percentile) | Nulliparous women, h Median (95th percentile) | P value |
|---|---|---|---|
| 4–5 | 1.8 (12.7) | 1.5 (8.9) | < .001 |
| 5–6 | 1.0 (5.1) | 0.9 (4.5) | .085 |
| 6–7 | 0.8 (4.3) | 0.7 (2.9) | < .001 |
| 7–8 | 0.6 (2.2) | 0.5 (2.0) | .573 |
| 8–9 | 0.5 (1.5) | 0.5 (1.7) | .622 |
| 9–10 | 0.5 (1.9) | 0.5 (2.0) | .376 |
| 4–10 | 7.4 (28.0) | 6.5 (25.8) | .007 |
| Second stage with epidural anesthesia | 1.0 (3.8) | 1.1 (3.8) | < .001 |
| Second stage without epidural anesthesia | 0.6 (2.8) | 0.7 (3.1) | .006 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
