Objective
The objective of the study was to examine whether the stage of labor dystocia causing a primary cesarean delivery (CD) affects a trial of labor after cesarean (TOLAC) success.
Study Design
This was a retrospective cohort study of women who had primary CD of singleton pregnancies for first- or second-stage labor dystocia and attempted TOLAC at a single hospital between 2002 and 2014. We compared TOLAC success rates between women whose primary CD was for first- vs second-stage labor dystocia and investigated whether the effect of prior dystocia stage on TOLAC success was modified by previous vaginal delivery (VD).
Results
A total of 238 women were included; nearly half (49%) achieved vaginal birth after cesarean (VBAC). Women with a history of second-stage labor dystocia were more likely to have VBAC compared with those with first-stage dystocia, although this trend was not statistically significant among the general population (55% vs 45%, adjusted odds ratio, 1.4, 95% confidence interval, 0.8–2.5]). However, among women without a prior VD, those with a history of second-stage dystocia did have statistically higher odds of achieving VBAC than those with prior first-stage dystocia (54% vs 38%, adjusted odds ratio, 1.8 [95% confidence interval, 1.0–3.3], P for interaction = .043).
Conclusion
Nearly half of women with a history of primary CD for labor dystocia will achieve VBAC. Women with a history of second-stage labor dystocia have a slightly higher VBAC rate, seen to a statistically significant degree in those without a history of prior VD. TOLAC should be offered to all eligible women and should not be discouraged in women with a prior second-stage arrest.
Because of the low risk of maternal and neonatal morbidity, a trial of labor after cesarean section (TOLAC) is considered a safe and reasonable option for women with a prior cesarean birth, with successful vaginal birth after cesarean (VBAC) rates among all those who attempt TOLAC reported between 60% and 74%. However, women who attempt TOLAC and end up delivering via unplanned repeat cesarean delivery (CD) are known to suffer higher rates of blood transfusion or hysterectomy compared with women who have elective, planned repeat CD.
To improve maternal morbidity by decreasing the rates of unplanned repeat CD, multiple studies have attempted to characterize predictors of TOLAC success. Data consistently suggest women have higher likelihood of VBAC if they have a history of vaginal delivery (VD) before or after a CD, had spontaneous labor in the TOLAC, or had a prior CD for a nonrecurrent indications such as breech presentation or nonreassuring fetal heart rate tracing. Conversely, women have lower likelihood of VBAC if the indication of CD was for labor dystocia; among these women, TOLAC success rates range from 13% to 80%.
Data remain limited, and conflicting, regarding the impact that stage of labor dystocia at the time of primary cesarean delivery may have on subsequent TOLAC success and whether the effect of labor dystocia is modified by other factors such as a previous VD or spontaneous labor in TOLAC. It also remains unclear whether women who have a primary CD for first- or second-stage labor dystocia are at risk for unplanned repeat CD for a recurrent indication.
This study aims to examine whether the stage of labor dystocia resulting in primary CD affects TOLAC success and whether this effect is modifiable by maternal or fetal factors.
Materials and Methods
This retrospective cohort study was conducted at the University of California, San Francisco (UCSF), a tertiary care, academic medical center. Between January 2002 and July 2014, the annual rate of cesarean delivery at UCSF ranged from 22.7% to 25.8% (average, 24.3%). There were no significant changes to labor management practices or TOLAC counseling during this time period, and prior cesarean alone is not considered an indication for induction.
Clinical information about all deliveries at UCSF is entered into a clinical research database immediately after birth by the delivering clinician. Data are validated by research coordinators shortly thereafter. Additional information that is not collected in this research database was obtained by detailed chart review. The UCSF Committee for Human Subjects Research approved this study.
Our study population consisted of women who had a primary CD of a singleton pregnancy for first- or second-stage labor dystocia at UCSF between January 2002 and July 2014 and attempted TOLAC with a subsequent singleton pregnancy at UCSF during the same time period. Our primary predictor was the stage of labor dystocia at time of prior primary CD, and the primary outcome was successful VBAC. First-stage labor dystocia was defined as a CD for a primary indication of failed induction of labor at any cervical dilation or active-phase arrest (cervical dilation at the time of CD of more than 4 cm but less than 10 cm); actual cervical dilation at the time of primary CD was not utilized as a variable for this study.
Second-stage labor dystocia was defined as a CD for arrest of descent (after full cervical dilation) or failed operative vaginal delivery. We excluded women whose main indication for primary CD was for nonreassuring fetal heart rate tracing or other nondystocia indication, even if labor dystocia was also present, to eliminate potential confounders from our analysis on the effect that stage of labor dystocia may have on TOLAC outcomes.
Demographic and obstetric characteristics of the TOLAC delivery were examined by descriptive statistics as well as χ 2 and t tests. TOLAC success rates between women whose primary CD was for first- vs second-stage labor dystocia were investigated with χ 2 and univariate and multivariate logistic regression. Covariates included maternal age, previous VD, infant weight, race or ethnicity, maternal diabetes mellitus (gestational or pregestational), and induction of labor for TOLAC.
We also tested the possibility of effect modification on the stage of labor dystocia by previous VD or induction of labor for TOLAC by adding the interaction term between each of these modifier variables and the stage of labor dystocia of primary CD into the separate multivariable models. Among women who failed TOLAC, logistic regression analysis was used to evaluate whether labor dystocia was recurrent.
We also conducted a literature search using the terms labor dystocia and VBAC to identify previously published data on this topic, and further references were identified via the bibliographies of those studies. The results of all applicable studies were stratified by stage of labor dystocia to create a patient-level metaanalysis of the relationship between a history of labor dystocia resulting in CD and TOLAC outcomes.
Results
A total of 405 women were identified as having a primary CD for labor dystocia and a subsequent delivery at UCSF between January 2002 and July 2014. Of these, 238 women (58.8%) attempted TOLAC, and TOLAC rates were similar among those with a history of first- or second-stage dystocia (58.1% vs 59.6%, P = .78). Demographic and obstetric characteristics at the time of the TOLAC attempt are reported in Table 1 .
| Characteristics | Total (n = 238) | History of primary CD for first-stage dystocia (n = 132) | History of primary CD for second-stage dystocia (n = 106) | P value |
|---|---|---|---|---|
| Maternal factors | ||||
| Maternal Age, y | 33.9 (±5.1) | 34.1 (±5.7) | 33.6 (±4.4) | .49 |
| Race or ethnicity | ||||
| White | 114 (47.9%) | 58 (43.9%) | 56 (52.8%) | .39 |
| Black | 29 (12.2%) | 20 (15.2%) | 9 (8.5%) | |
| Latina | 28 (11.8%) | 18 (13.6%) | 10 (9.4%) | |
| Asian or Pacific Islander | 46 (19.3%) | 25 (18.9%) | 21 (19.8%) | |
| Other or unknown | 21 (8.8%) | 21 (8.3%) | 10 (9.4%) | |
| Current diabetes mellitus b | 28 (11.3%) | 19 (14.4%) | 9 (8.5%) | .16 |
| History of vaginal delivery | 40 (16.8%) | 20 (15.2%) | 20 (18.9%) | .45 |
| Neonatal factors | ||||
| Gestational age at delivery, wks | 38.8 (±1.9) | 38.7 (±2.2) | 38.9 (±1.5) | .43 |
| Infant birthweight, g | 3499.4 (±594.0) | 3487.7 (±615.9) | 3514.1 (±567.9) | .73 |
| Intrapartum factors | ||||
| Labor induced | 49 (20.6%) | 33 (25.0%) | 16 (15.1%) | .06 |
| Labor augmented with oxytocin | 123 (51.7%) | 81 (61.4%) | 42 (39.6%) | < .001 |
| Epidural used | 161 (67.9%) | 94 (71.8%) | 67 (63.2%) | .16 |
a All data are presented as n (percentage) or mean (±SD)
The overall mean gestational age at delivery was slightly less than 39 weeks; most women in each group had spontaneous labor. Among women with prior first-stage dystocia and those with prior second-stage dystocia, characteristics during TOLAC attempt were similar, except for intrapartum oxytocin augmentation (61.4% vs 39.6%, P < .001), and induction of labor, with a marginal significance level (25.0% vs 15.1%, respectively; P = .06) ( Table 1 ).
Nearly half of those attempting TOLAC (49.2%) achieved VBAC. Although a higher TOLAC success rate was observed among women with a prior second-stage dystocia compared with those with first-stage dystocia, the difference was not statistically significant in the entire population (54.7% vs 44.7% , respectively; P = .12, adjusted odds ratio [aOR] 1.43 [95% confidence interval (CI), 0.82–2.47)) ( Table 2 ). However, when we investigated the possibility of effect modification by history of prior vaginal delivery, we found the interaction term of prior vaginal delivery to be statistically significant ( P for interaction term = .04).
| History of first-stage dystocia, n, % | History of second-stage dystocia, n, % | P value | aOR (95% Cl) for prior second- vs prior first-stage dystocia a | P value | P value for interaction | |
|---|---|---|---|---|---|---|
| All (n = 238) | 59/132 (45%) | 58/106 (55%) | .12 | 1.43 (0.82-2.47) | .20 | N/A |
| Prior VD | .043 | |||||
| No prior VD (n = 198) | 43/112 (38%) | 46/86 (54%) | .03 | 1.82 (1.00-3.32) b | .049 | |
| Prior VD (n = 40) | 16/20 (80%) | 12/20 (60%) | .17 | 0.35 (0.08-1.53) b | .16 | |
| Labor induction for TOLAC | .146 | |||||
| Spontaneous labor (n = 189) | 47/99 (47%) | 47/90 (52%) | .51 | 1.17 (0.62-2.15) b | .61 | |
| Labor induction (n = 49) | 12/33 (36%) | 11/16 (69%) | .03 | 3.70 (0.91-15.14) b | .07 |
a aOR for history of second-stage dystocia on VBAC, adjusting for maternal age, race-ethnicity, birthweight, prior vaginal delivery, induction of labor for TOLAC, and DM
b Multivariable model additionally included an interaction term between the stage of labor dystocia from the prior primary CD and the effect modifiers being investigated, as indicated.
Among women without prior VD, those with a history of second-stage dystocia had statistically significantly higher odds of achieving VBAC than those with previous first-stage dystocia (54% vs 38%, P = .03; aOR for history of second-stage dystocia, 1.8 [95% CI, 1.0–3.3]). Such an effect was not observed if the woman had a prior VD (aOR, 0.35 [95% CI, 0.08–1.53]) ( Table 2 ).
VBAC rates were similar among the women who had spontaneous labor for the TOLAC (52% vs 47.5%, respectively, for history of second-stage vs first-stage dystocia, P = .51). However, among women who had an induction of labor for their TOLAC, those with a history of second-stage dystocia (n = 11) had a higher VBAC rate compared with those with prior first-stage dystocia (n = 12) (68.7% vs 36.4%, P = .03) ( Table 2 ), although no statistically significant interaction effect was observed ( P = .17).
Of the 121 women who had a failed TOLAC, those with a history of second-stage labor dystocia were more likely to reach the second stage before having their unplanned repeat cesarean delivery (52% vs 19%, P < .001, aOR, 4.61 [95% CI, 1.86–11.43]).
When data among all previously published studies analyzing labor dystocia and subsequent VBAC rates were analyzed on a patient level, the overall VBAC rate of CD after first-stage dystocia was 69% (range, 45–80%) and after second-stage dystocia was 52% (range, 13–76%) ( Table 3 ). Of note, the overall VBAC rate with a history of second-stage dystocia improved to 66% if the outlying study with a 13% success rate was excluded ( Table 3 ).
