Background: Trial of labor after cesarean delivery (TOLAC) has long been accepted as a safe option for women with previous cesarean delivery. Previous efforts have been exerted in trials to predict the success rates of TOLAC according to specific parameters related to previous cesarean section and before TOLAC. We aimed to investigate the different indications of previous cesarean delivery as independent predictors for successful vaginal birth. Method s: A retrospective study was conducted in Armed Forces Hospitals of the Souther n Region between December 15, 2019, and July 1, 2020. The included 566 patients with previous cesarean section who were willing to undergo a trial of labor were divided into two groups according to the success of vaginal birth (VBAC). Results: The nonrecurring indications for previous cesarean delivery were higher in the successful group (fetal distress 54.7% vs 41.1%, malpresentation 26% vs 21.4%, multifetal pregnancy 3.8% vs 2.7%). Additionally, the successful VBAC group had a significantly higher percentage of previous successful VBAC (47.7% vs 21.9%) and prior vaginal deliveries (58.5% vs 44.2%) and less coincidence of medical disorders and meconium-stained liquor (18.1% vs 26.3% and 3.2% vs 8.2%, respectively) than the unsuccessful group. Conclusion: During counseling regarding trial of labor after cesarean section, indications for previous cesarean section not related to arrest of labor can predict higher success of VBAC. Moreover, previous successful vaginal delivery or VBAC improves the success rates.
Cesarean section is the most common surgical procedure conducted in daily practice worldwide. Over the years, it showed a steadily increasing trend. Despite being lifesaving in specific situations during pregnancy and labor, it remains a major surgery that may be complicated by morbid maternal and perinatal sequelae [
In an attempt to reduce cesarean delivery rates, TOLAC has long been accepted as a safe option for women with previous cesarean delivery with a lower risk of morbidities than repeat cesarean delivery and a minimal risk (0.2% - 0.8%) of uterine rupture. The ACOG and RCOG encouraged counseling women with a single cesarean scar for TOLAC with a high-predicted success rate of up to 75% [
Previous efforts have been exerted in trials to predict the success rates of TOLAC according to specific parameters related to the previous section and before TOLAC. These parameters included maternal obesity, cervical dilatation at admission, association with specific medical disorders, previous vaginal deliveries, previous successful vaginal birth after cesarean (VBAC) and indication for previous cesarean delivery [
The indications for cesarean delivery could be divided into two groups according to the tendency of persistence or recurrence. Indications related to arrest of labor progress and failure of labor induction usually have a higher tendency of recurrence. Fetal distress, malpresentation, multiple pregnancy and antepartum hemorrhage have a lower tendency to recur [
In this study, we aimed to investigate the different indications of previous cesarean delivery as independent predictors for successful VBAC.
This retrospective study was conducted in the Obstetrics and Gynecology department (140 beds with 7000 - 8000 deliveries per year) of the Armed Forces Hospitals of Southern Region (AFHSR), KSA, during the period between December 15, 2019, and July 1, 2020, after approval of the AFHSR Research Ethics Committee (approval code: AFHSRMREC/2019/OB/GYNAE/421). The study included all eligible patients who accepted TOLAC delivery during the period between January 2017 and the end of December 2019.
The primary outcome was testing the type of indication of previous cesarean section on the prediction of successful vaginal delivery for women willing to undergo trial of previous scar. Secondary outcomes are assessing the other previous and current covariates that could affect the chance of successful vaginal delivery as secondary predictors.
1) Aged 18 years or more.
2) Previous single uncomplicated cesarean delivery.
3) Singleton pregnancy.
1) Current multifetal deliveries.
2) More than a single previous uterine scar.
3) Malpresentation.
Electronic files of the legible cases were reviewed. Demographic data, including age, parity, gravidity, weight, height, BMI, indication for the previous cesarean delivery, and associated medical and surgical disorders, were collected. Labor was spontaneous or induced, mode of delivery, indication for repeat cesarean delivery, and stage of labor when cesarean was indicated. Obstetric complications included uterine rupture, postpartum hemorrhage, and extensions.
We used the software package: SPSSÓ version 25 [IBMÓ Corp., Armonk, NY]. The Shapiro-Wilk test was used to examine the numerical data for normality of distribution. Skewed data are presented as the median and interquartile range (IQR). Normally distributed data are presented as the mean ± standard deviation (SD). Categorical data are presented as numbers and percentages (%). Independent t tests were performed to compare continuous variables between the successful VBAC and nonsuccessful VBAC groups. The χ2 test and Fisher’s exact test were used to compare categorical variables between both groups as convenient. A P value below 0.05 was considered statistically significant.
We used binary logistic regression to test study parameters (including history of prior successful VBAC, number of prior successful VBAC procedures, interval in years from previous cesarean, prior vaginal delivery, associated medical disorders, different indications for previous cesarean delivery, presence of meconium-stained liquor at time of delivery, patient BMI, and type of onset of labor). All the previous parameters were tested in the model using enter mode. A P value < 0.05 was considered statistically significant. All results are shown in tables.
Between the beginning of 2017 and the end of 2019, 566 patients with a previous uncomplicated cesarean delivery attempted TOLAC. The mean age of the participants was 32 years, and 75% of them were overweight to category II obesity. The median parity was para 2, with a history of previous vaginal delivery in 52% of them. The mean duration from the last cesarean section was 4.7 years. Approximately 37% of the cases had at least one successful TOLAC on average. The most frequent indication of previous cesarean delivery was fetal distress (49%), followed by malpresentation (24%) and failure of labor progress (16%). Only 9% of cases needed mechanical induction of labor, while the remaining 91% entered spontaneous labor pain. Twenty-one percent had premature rupture of membranes before the start of labor. Sixty percent of the study population had successful vaginal delivery, while the remaining 40% needed cesarean delivery (
| Variable | Result (n =566) | ||
|---|---|---|---|
| Age (years) | 32.1 ± 5.2 | ||
| BMI (kg/m2) | 34.1 ± 6.8 | ||
| Average | 41 (7.2%) | ||
| Overweight | 115 (20.3%) | ||
| Category I obesity | 171 (30.2%) | ||
| Category II obesity | 145 (25.6%) | ||
| Category III obesity | 94 (16.6%) | ||
| Parity | 2 (1 - 4) | ||
| Prior VBAC | 210 (37.1%) | ||
| Number of prior VBAC | 0 (0 - 1) | ||
| Prior vaginal delivery | 299 (52.8%) | ||
| Indication of previous cesarean | |||
| Fetal distress | 279 (49.3%) | ||
| Failed progress | 91 (16.1%) | ||
| Failed induction of labour | 17 (3%) | ||
| Malpresentation | 137 (24.2%) | ||
| Multifetal pregnancy | 19 (3.4%) | ||
| Antepartum hemorrhage | 23 (4.1%) | ||
| Interval from last cesarean (years) | 4.7 ± 3.5 | ||
| Antenatal care visits | 4 (2 - 7) | ||
| Associated medical disorders | 121 (21.4%) | ||
| Onset of labour | |||
| Spontaneous | 517 (91.3%) | ||
| Induced | 49 (8.7%) | ||
| Premature rupture of membranes | 119 (21%) | ||
| Meconium stained liquor | 31 (5.5%) | ||
| Mode of delivery | |||
| Vaginal | 342 (60.4%) | ||
| Cesarean | 224 (39.6%) | ||
| Delivery gestational age (weeks) | 38.6 ± 1.9 | ||
| Neonatal birth weight (grams) | 2901 ± 512 | ||
Data are presented as mean ± standard deviation, number (percentage) and median (interquartile range) as appropriate.
The study population was divided for comparison into an unsuccessful VBAC group and a successful VBAC group. The group that succeeded in accomplishing vaginal delivery was characterized by significantly higher parity (2 vs 1, P value 0.006), a higher percentage of previous successful VBAC (47.7% vs 21.9%, P value 0.000), prior vaginal deliveries before cesarean section (58.5% vs 44.2%), a longer time interval from the last cesarean section (5.1 vs 4 years), and less coincidence of medical disorders and meconium-stained liquor (18.1% vs 26.3% and 3.2% vs 8.2%, respectively) than the unsuccessful group. Regarding the primary outcome, the nonrecurring indications for previous cesarean delivery were higher in the successful group (fetal distress 54.7% vs 41.1%, Malpresentation 26% vs 21.4%, multifetal pregnancy 3.8% vs 2.7%), while the recurring indications were less frequent compared to the unsuccessful group (11.1% vs 23.7% failed progress, 2.3% vs 4% failed labor induction). There was no significant difference regarding the type of labor onset or incidence of premature rupture of membranes (
To test the previously mentioned parameters as independent predictors of successful vaginal birth after cesarean section, binary logistic regression analysis with enter mode showed multiple significant predictors. The indication for the previous cesarean section (which is the primary outcome for the current study) came on top of these positive predictors. Fetal distress (P value: 0.000, odds ratio: 6.762, 95% CI: 2.434 - 18.781), malpresentation (P value: 0.006, odds ratio: 4.346, 95% CI: 1.512 - 12.491) and multifetal pregnancy (P value: 0.027, odds ratio: 5.155, 95% CI: 1.207 - 22.018) favored successful delivery. However, the recurring indications (failure of progress and failed induction) failed to be predictors of success.
In addition, prior successful VBAC appeared to be a highly significant positive predictor of successful VBAC (P value: 0.000, odds ratio: 3.908, 95% CI: 2.051 - 7.450) and, to a lesser extent, previous vaginal delivery before cesarean section (P value: 0.006, odds ratio: 1.791, 95% CI: 1.186 - 2.706). On the other hand, both coincidence of medical disorders (P value: 0.020, odds ratio: 0.577, 95% CI: 0.364 - 0.916) and meconium-stained liquor (P value: 0.008, odds ratio: 0.355, 95% CI: 0.149 - 0.754) were significant negative predictors for successful vaginal delivery. The number of previous VBAC procedures, interval from the last cesarean section and BMI were not significant predictors for the current model (
During the counseling of women with previous cesarean scar regarding TOLAC, the physician should discuss the probabilities of success of VBAC. Usually, we discuss the general stated success rates (60% - 80%). This approach neglects an important item in counseling, which is individual variations between women in characteristic parameters, which in turn could modify the chances of success. Multiple attempts have been made to establish structured models to allow more individualized counseling according to patients’ own characteristics [
| Variable | Unsuccessful VBAC (n = 224) | Successful VBAC (n = 342) | P-value | ||
|---|---|---|---|---|---|
| Age (years) | 32.0 ± 5.4 | 32.1 ± 5.1 | 0.974 | ||
| BMI (kg/m2) | 34.7 ± 7.3 | 33.5 ± 6.5 | 0.046* | ||
| Average (18.5 to <25) | 17 (7.6%) | 24 (7%) | 0.004* | ||
| Overweight (25.0 to <30) | 47 (21%) | 68 (19.9%) | |||
| Category I obesity (30 to <35) | 48 (21.4%) | 123 (36%) | |||
| Category II obesity (35 to <40) | 63 (28.1%) | 82 (24%) | |||
| Category III obesity (40 or higher) | 49 (21.9%) | 45 (13.2%) | |||
| Parity | 1 (1 - 4) | 2 (1 - 4) | 0.006* | ||
| Prior VBAC | 47 (21%) | 163 (47.7%) | 0.000* | ||
| Number of prior VBAC | 0 (0-0) | 0 (0-1) | 0.000* | ||
| Prior vaginal delivery | 99 (44.2%) | 200 (58.5%) | 0.001* | ||
| Indication of previous cesarean | 0.000* | ||||
| Fetal distress | 92 (41.1%) | 187 (54.7%) | |||
| Failed progress | 53 (23.7%) | 38 (11.1%) | |||
| Failed induction of labour | 9 (4%) | 8 (2.3%) | |||
| Malpresentation | 48 (21.4%) | 89 (26%) | |||
| Multifetal pregnancy | 6 (2.7%) | 13 (.8%) | |||
| Antepartum hemorrhage | 16 (7.1%) | 7 (2%) | |||
| Interval from last cesarean (years) | 4 ± 3.1 | 5.1 ± 3.6 | 0.000* | ||
| Antenatal care visits | 5 (2 - 7) | 5 (2 - 7) | 0.652 | ||
| Associated medical disorders | 59 (26.3%) | 62 (18.1%) | 0.021* | ||
| Onset of labour | 0.542 | ||||
| Spontaneous | 207 (92.4%) | 310 (90.6%) | |||
| Induced | 17 (7.6%) | 32 (9.4%) | |||
| Premature rupture of membranes | 45 (20.1%) | 74 (21.6%) | 0.675 | ||
| Meconium stained liquor | 20 (8.9%) | 11 (3.2%) | 0.004* | ||
| Delivery gestational age (weeks) | 38.2 ± 2.3 | 38.8 ± 1.6 | 0.842 | ||
| Neonatal birth weight (grams) | 2805 ± 558 | 2963 ± 470 | 0.001* | ||
Data are presented as mean ± standard deviation, number (percentage) and median (interquartile range) as appropriate. * Significant result.
| Covariate | B | SE | P-value | OR | 95% CI | |
|---|---|---|---|---|---|---|
| Prior successful VBAC | 1.363 | 0.329 | 0.000* | 3.908 | 2.051 - 7.450 | |
| Number of previous VBAC | −0.035 | 0.168 | 0.836 | 0.966 | 0.696 - 1.341 | |
| Interval from last cesarean | −0.019 | 0.044 | 0.676 | 0.982 | 0.900 - 1.071 | |
| Prior vaginal delivery | 0.583 | 0.211 | 0.006* | 1.791 | 1.186 - 2.706 | |
| Associated medical disorders | −0.549 | 0.236 | 0.020* | 0.577 | 0.364 - 0.916 | |
| Indication of previous cesarean | ||||||
| Fetal distress | 1.911 | 0.521 | 0.000* | 6.762 | 2.434 - 18.781 | |
| Failure of labour progress | 0.710 | 0.549 | 0.196 | 2.035 | 0.693 - 5.972 | |
| Failed induction of labour | 0.784 | 0.738 | 0.288 | 2.190 | 0.515 - 9.308 | |
| Malpresentation | 1.469 | 0.539 | 0.006* | 4.346 | 1.512 - 12.491 | |
| Multifetal pregnancy | 1.640 | 0.741 | 0.027* | 5.155 | 1.207 - 22.018 | |
| Meconium stained liquor | −1.093 | 0.414 | 0.008* | 0.355 | 0.149 - 0.754 | |
| BMI | 0.003 | 0.045 | 0.955 | 1.003 | 0.917 - 1.096 | |
B; regression coefficient, SE; standard error, OR; odds ratio, 95% CI; 95% confidence interval. Hosmer and Lemeshow goodness of fit test was insignificant for all. χ2 (19, N = 566) = 110.95, P < 0.001. 18% to 24.1% of the variance in the dependent variable can be explained by independent variables in the model. Percentage of accuracy (PAC) of the model is 72.6%. * Significant predictor for successful VBAC.
In this study, we investigated the indications for previous cesarean delivery as significant independent predictors for success during counseling for TOLAC. The indications with a lower chance for recurrence were significantly reassuring parameters for successful delivery. This included, in order of significance, fetal distress (odds ratio: 6.76), pregnancy (odds ratio: 5.16) and malpresentation (odds ratio: 4.35). However, indications related to the arrest of labor with a higher tendency to be recurrent, such as failure of labor progress and failed labor induction, were nonreassuring. This finding seems logical and in concordance with the findings of similar previous studies [
In addition, more detailed parameters were studied and proved to also be significant independent modifiers of the success rates of VBAC during the trial of labor. On top of these variables comes the prior successful VBAC. According to the regression model, prior successful VBAC carries fourfold chances for repeat successful VBAC compared to women who did not have this chance. The number of previous VBACs did not appear to be a significant modifier of risk. This is in concordance with the findings of several previous studies [
Similarly, a history of precesarean vaginal delivery appeared to be reassuring for successful VBAC. According to our model, it increases the chances of success 1.8-fold compared to women who did not deliver vaginally before. RCOG stated that previous vaginal delivery, in particular previous VBAC, is considered the best predictor of successful TOLAC, with a success rate of 85% - 90% [
The presence of medical disorders with pregnancy and meconium-stained liquor during labor were also significant negative predictors of VBAC success. They were associated with a reduction in chances to 0.57 and 0.36 compared to women without such risk factors. Some other parameters failed to be significant predictors in our model but were proven to be significant in other studies. These included body mass index (BMI), interval from last cesarean delivery and neonatal birth weight. Some of these findings might be due to relatively comparable parameters between both groups.
The presence of medical disorders with pregnancy and meconium-stained liquor during labor were also significant negative predictors of VBAC success. They were associated with a reduction in chances to 0.57 and 0.36 compared to women without such risk factors. Some other parameters failed to be significant predictors in our model but were proven to be significant in other studies. These included body mass index (BMI), interval from last cesarean delivery and neonatal birth weight. Some of these findings might be due to relatively comparable parameters between both groups.
This study succeeded in providing solid parameters in the obstetric history and the current condition of women with previous cesarean scar that could help for better prediction and counseling regarding their chances for successful vaginal delivery. We have extracted our findings depending on a suitable number of participants in both arms that empowers the results. All data were available through detailed reporting on an electronic hospital information system, which reduces recall bias.
The retrospective nature is a limitation that exposes the findings despite all efforts to selection and recall bias. Additionally, this work could not guide counseling for women willing to undergo scar trial with more than one previous cesarean delivery. Prospective well powered studies regarding this concern are recommended.
During counseling regarding TOLAC, indications for the previous cesarean delivery not related to arrest of labor can predict higher success of VBAC. Moreover, previous vaginal delivery or VBAC improves the success rates.
HA: concept generation, design of the study, analysis, and interpretation of the data, drafting of the manuscript, and critical revision. AK: selecting participants, gathering data, and writing the manuscript. NM: concept generation, study design, writing of the manuscript, and critical revision.
The authors like to thank all the study participants. We owe a debt of gratitude to the reputable medical professionals and nursing personnel at King Faisal Military City Hospital in the Armed Forces Southern Region of Saudi Arabia for their unwavering support and collaboration.
Armed Forces Hospitals Southern Region, KSA.
The requirement for informed consent was waived. Due to the retrospective nature of the study, the data collected did not breach patient confidentiality or welfare, and the study was non practicably applicable without waiving. The ethics committee approved that waiver for the same reasons.
Each approach was carried out in accordance with the rules and regulations outlined in the Declaration of Helsinki.
Authors can afford raw dataset on reasonable request.
All authors have reviewed and approved the final version of the manuscript for publication.
The authors declare no conflicts of interest regarding the publication of this paper.
Atia, H., Khider, A. and Metwally, N.M. (2023) Prediction of Success Rates of Vaginal Birth after Cesarean Delivery According to the Previous Indication for Cesarean Delivery. Open Journal of Obstetrics and Gynecology, 13, 37-46. https://doi.org/10.4236/ojog.2023.131004