Objective: Our purpose was to determine the maternal risks associated with failed attempt at vaginal birth after cesarean compared with elective repeat cesarean delivery or successful vaginal birth after cesarean.
Study design: From 1989 to 1998 all patients attempting vaginal birth after cesarean and all patients undergoing repeat cesarean deliveries were reviewed. Data were extracted from a computerized obstetric database and from medical charts. The following three groups were defined: women who had successful vaginal birth after cesarean, women who had failed vaginal birth after cesarean, and women who underwent elective repeat cesarean. Criteria for the elective repeat cesarean group included no more than two previous low transverse or vertical incisions, fetus in cephalic or breech presentation, no previous uterine surgery, no active herpes, and adequate pelvis. Predictor variables included age, parity, type and number of previous incisions, reasons for repeat cesarean delivery, gestational age, and infant weight. Outcome variables included uterine rupture or dehiscence, hemorrhage >1000 mL, hemorrhage >2000 mL, need for transfusion, chorioamnionitis, endometritis, and length of hospital stay. The Student t test and the chi(2) test were used to compare categoric variables and means; maternal complications and factors associated with successful vaginal birth after cesarean were analyzed with multivariate logistic regression, allowing odds ratios, adjusted odds ratios, 95% confidence intervals, and P values to be calculated.
Results: A total of 29,255 patients were delivered during the study period, with 2450 having previously had cesarean delivery. Repeat cesarean deliveries were performed in 1461 women (5.0%), and 989 successful vaginal births after cesarean delivery occurred (3.4%). Charts were reviewed for 97.6% of all women who underwent repeat cesarean delivery and for 93% of all women who had vaginal birth after cesarean. Vaginal birth after cesarean was attempted by 1344 patients or 75% of all appropriate candidates. Vaginal birth after cesarean was successful in 921 women (69%) and unsuccessful in 424 women. Four hundred fifty-one patients undergoing cesarean delivery were deemed appropriate for vaginal birth after cesarean. Multiple gestations were excluded from analysis. Final groups included 431 repeat cesarean deliveries and 1324 attempted vaginal births after cesarean; in the latter group 908 were successful and 416 failed. The overall rate of uterine disruption was 1.1% of all women attempting labor; the rate of true rupture was 0.8%; and the rate of hysterectomy was 0.5%. Blood loss was lower (odds ratio, 0.5%; 95% confidence interval, 0.3-0.9) and chorioamnionitis was higher (odds ratio, 3.8%; 95% confidence interval, 2.3-6.4) in women who attempted vaginal births after cesarean. Compared with women who had successful vaginal births after cesarean, women who experienced failed vaginal births after cesarean had a rate of uterine rupture that was 8.9% (95% confidence interval, 1.9-42) higher, a rate of transfusion that was 3.9% (95% confidence interval, 1.1-13.3) higher, a rate of chorioamnionitis that was 1.5% (95% confidence interval, 1.1-2.1) higher, and a rate of endometritis that was 6.4% (95% confidence interval, 4.1-9.8) higher.
Conclusion: Patients who experience failed vaginal birth after cesarean have higher risks of uterine disruption and infectious morbidity compared with patients who have successful vaginal birth after cesarean or elective repeat cesarean delivery. Because actual numbers of morbid events are small, caution should be exercised in interpreting results and counseling patients. More accurate prediction for safe, successful vaginal birth after cesarean delivery is needed.