Anne H Mardy1, Cande V Ananth2, William A Grobman3, Cynthia Gyamfi-Bannerman4. 1. Department of Obstetrics and Gynecology, Columbia University Medical Center, New York, NY. Electronic address: aah2150@columbia.edu. 2. Department of Obstetrics and Gynecology, Columbia University Medical Center, New York, NY; Department of Epidemiology, Joseph L. Mailman School of Public Health, Columbia University, New York, NY. 3. Division of Maternal-Fetal Medicine, Northwestern University Feinberg School of Medicine, Chicago, IL. 4. Department of Obstetrics and Gynecology, Columbia University Medical Center, New York, NY; Division of Maternal-Fetal Medicine, Columbia University Medical Center, New York, NY.
Abstract
BACKGROUND: A validated model exists that predicts the probability of vaginal birth after cesarean delivery in patients at term who are undergoing a trial of labor after cesarean delivery. However, a model that predicts the success of a vaginal birth after cesarean delivery in the preterm period has not been developed. OBJECTIVE: We sought to develop and validate a predictive model for vaginal birth after cesarean delivery for women undergoing a trial of labor after cesarean delivery during the preterm period. STUDY DESIGN: We performed a secondary analysis of a prospective cohort study designed to evaluate perinatal outcomes in women with a prior cesarean scar. We included women with 1 prior low transverse cesarean delivery undergoing a trial of labor after cesarean delivery with a vertex singleton pregnancy in the preterm period (26-36 weeks). Using multivariable logistic regression modeling, we constructed a predictive model for vaginal birth after cesarean delivery with information known at admission for preterm delivery. Using a 70% to 30% random split of the data, the model was developed in the training data and subsequently confirmed in the validation data. Predictions and area under the curve were based on a 10-fold cross-validated jackknife estimation and based on 1000 bootstrap resampling methods. The adequacy of all models was evaluated based on the Hosmer-Lemeshow goodness-of-fit test. RESULTS: One thousand two hundred ninety-five women met our criteria for analysis. The significant predictors of vaginal birth after cesarean delivery success were chronic hypertension, hypertensive disease of pregnancy (gestational hypertension or preeclampsia), prior vaginal delivery, dilation on cervical examination at admission, prior vaginal birth after cesarean delivery, a recurring indication in a prior cesarean delivery, and induction of labor as well as a 2-way interactions between dilation and hypertensive disease of pregnancy, dilation and diabetes mellitus (pregestational or gestational), and induction of labor and hypertensive disease of pregnancy. The area under the curve from the prediction model was 0.80 (95% confidence interval, 0.77-0.83) and the model fit the data well (Hosmer-Lemeshow P = .367). The bootstrap and 10-fold cross-validated jackknife estimates of the corrected area under the curve of the model were 0.78 (95% confidence interval, 0.74-0.82) and 0.77 (95% confidence interval, 0.73-0.82), respectively, following incorporation of regression shrinkage. CONCLUSION: A cross-validated predictive model was created for patients undergoing a trial of labor after cesarean delivery in the preterm period using 8 variables known on admission. These factors were notably different from factors used in the model for term patients. This new model can be used to counsel patients in the preterm period who want to undergo a trial of labor after cesarean delivery on their predicted vaginal birth after cesarean delivery success.
BACKGROUND: A validated model exists that predicts the probability of vaginal birth after cesarean delivery in patients at term who are undergoing a trial of labor after cesarean delivery. However, a model that predicts the success of a vaginal birth after cesarean delivery in the preterm period has not been developed. OBJECTIVE: We sought to develop and validate a predictive model for vaginal birth after cesarean delivery for women undergoing a trial of labor after cesarean delivery during the preterm period. STUDY DESIGN: We performed a secondary analysis of a prospective cohort study designed to evaluate perinatal outcomes in women with a prior cesarean scar. We included women with 1 prior low transverse cesarean delivery undergoing a trial of labor after cesarean delivery with a vertex singleton pregnancy in the preterm period (26-36 weeks). Using multivariable logistic regression modeling, we constructed a predictive model for vaginal birth after cesarean delivery with information known at admission for preterm delivery. Using a 70% to 30% random split of the data, the model was developed in the training data and subsequently confirmed in the validation data. Predictions and area under the curve were based on a 10-fold cross-validated jackknife estimation and based on 1000 bootstrap resampling methods. The adequacy of all models was evaluated based on the Hosmer-Lemeshow goodness-of-fit test. RESULTS: One thousand two hundred ninety-five women met our criteria for analysis. The significant predictors of vaginal birth after cesarean delivery success were chronic hypertension, hypertensive disease of pregnancy (gestational hypertension or preeclampsia), prior vaginal delivery, dilation on cervical examination at admission, prior vaginal birth after cesarean delivery, a recurring indication in a prior cesarean delivery, and induction of labor as well as a 2-way interactions between dilation and hypertensive disease of pregnancy, dilation and diabetes mellitus (pregestational or gestational), and induction of labor and hypertensive disease of pregnancy. The area under the curve from the prediction model was 0.80 (95% confidence interval, 0.77-0.83) and the model fit the data well (Hosmer-Lemeshow P = .367). The bootstrap and 10-fold cross-validated jackknife estimates of the corrected area under the curve of the model were 0.78 (95% confidence interval, 0.74-0.82) and 0.77 (95% confidence interval, 0.73-0.82), respectively, following incorporation of regression shrinkage. CONCLUSION: A cross-validated predictive model was created for patients undergoing a trial of labor after cesarean delivery in the preterm period using 8 variables known on admission. These factors were notably different from factors used in the model for term patients. This new model can be used to counsel patients in the preterm period who want to undergo a trial of labor after cesarean delivery on their predicted vaginal birth after cesarean delivery success.
Authors: William A Grobman; Grecio Sandoval; Madeline Murguia Rice; Jennifer L Bailit; Suneet P Chauhan; Maged M Costantine; Cynthia Gyamfi-Bannerman; Torri D Metz; Samuel Parry; Dwight J Rouse; George R Saade; Hyagriv N Simhan; John M Thorp; Alan T N Tita; Monica Longo; Mark B Landon Journal: Am J Obstet Gynecol Date: 2021-05-24 Impact factor: 8.661
Authors: Jeong Ha Wie; Se Jin Lee; Sae Kyung Choi; Yun Sung Jo; Han Sung Hwang; Mi Hye Park; Yeon Hee Kim; Jae Eun Shin; Ki Cheol Kil; Su Mi Kim; Bong Suk Choi; Hanul Hong; Hyun-Joo Seol; Hye-Sung Won; Hyun Sun Ko; Sunghun Na Journal: Life (Basel) Date: 2022-04-18
Authors: Robert M Silver; Madeline Murguia Rice; William A Grobman; Uma M Reddy; Alan T N Tita; Gail Mallett; Kim Hill; Elizabeth A Thom; Yasser Y El-Sayed; Ronald J Wapner; Dwight J Rouse; George R Saade; John M Thorp; Suneet P Chauhan; Edward K Chien; Brian M Casey; Ronald S Gibbs; Sindhu K Srinivas; Geeta K Swamy; Hyagriv N Simhan; George A Macones Journal: Obstet Gynecol Date: 2020-10 Impact factor: 7.623
Authors: Aireen Wingert; Lisa Hartling; Meghan Sebastianski; Cydney Johnson; Robin Featherstone; Ben Vandermeer; R Douglas Wilson Journal: BMC Pregnancy Childbirth Date: 2019-12-30 Impact factor: 3.007