S L Kjos1, K Berkowitz, A Xiang. 1. Department of Obstetrics and Gynecology, University of Southern California, Keck School of Medicine, Los Angeles, California 90033, USA.
Abstract
OBJECTIVES: To identify independent risk factors for Cesarean delivery in women with pregnancy complicated by diabetes. METHODS: Retrospective analysis of pregnancies from 5735 diabetic women delivering liveborn infants. Maternal demographic, medical, obstetric historical factors and index pregnancy obstetric, glycemic and neonatal outcome parameters were evaluated for association with Cesarean delivery after a trial of labor. Individual risk factors were analyzed for association by chi2 and ANOVA. Independent predictors of Cesarean delivery and adjusted relative risk (RR) were identified by stepwise logistic regression. RESULTS: Trial of labor was permitted in 90.8% and 59.4% of women without (n = 4643) and with prior Cesarean delivery (n = 1092) and was successful in 85.2% and 56.9%, respectively. Eleven independent predictors were found. Five were related to obstetric history and maternal age: prior Cesarean delivery (RR 5.34, 95% CI 3.94-7.25), no prior live birth (RR 3.17, 95% CI 1.98-5.07), no prior vaginal delivery (RR 2.28, 95% CI 1.50-3.44), prior stillbirth (RR 1.71, 95% CI 1.09-2.68%) and maternal age > or = 35 years (RR 1.53, 95% CI 1.20-1.93). Two were related to the severity of diabetes at entry to diabetes care: requiring insulin (RR 1.53, 95% CI 1.20-1.93) and highest fasting plasma glucose level (RR 1.04, 95% CI 1.01-1.07). Two were related to obstetric factors: pre-eclampsia/hypertension (RR 2.56, 95% CI 2.00-3.27) and labor induction (RR 3.32, 95% CI 2.70-4.10). The remaining two were birth weight (per 250 g, RR 1.12, 95% CI 1.09-1.17) and pre-delivery body mass index (RR, 1.03, 95% CI 1.00-1.05). CONCLUSION: The majority of predictors were not modifiable, relating to obstetric history, maternal age and diabetes severity. Possible modifiable interventions to avoid/improve labor induction, and decrease birth weight and maternal weight gain might decrease risk of Cesarean delivery. Future studies must address these multiple predictors.
OBJECTIVES: To identify independent risk factors for Cesarean delivery in women with pregnancy complicated by diabetes. METHODS: Retrospective analysis of pregnancies from 5735 diabeticwomen delivering liveborn infants. Maternal demographic, medical, obstetric historical factors and index pregnancy obstetric, glycemic and neonatal outcome parameters were evaluated for association with Cesarean delivery after a trial of labor. Individual risk factors were analyzed for association by chi2 and ANOVA. Independent predictors of Cesarean delivery and adjusted relative risk (RR) were identified by stepwise logistic regression. RESULTS: Trial of labor was permitted in 90.8% and 59.4% of women without (n = 4643) and with prior Cesarean delivery (n = 1092) and was successful in 85.2% and 56.9%, respectively. Eleven independent predictors were found. Five were related to obstetric history and maternal age: prior Cesarean delivery (RR 5.34, 95% CI 3.94-7.25), no prior live birth (RR 3.17, 95% CI 1.98-5.07), no prior vaginal delivery (RR 2.28, 95% CI 1.50-3.44), prior stillbirth (RR 1.71, 95% CI 1.09-2.68%) and maternal age > or = 35 years (RR 1.53, 95% CI 1.20-1.93). Two were related to the severity of diabetes at entry to diabetes care: requiring insulin (RR 1.53, 95% CI 1.20-1.93) and highest fasting plasma glucose level (RR 1.04, 95% CI 1.01-1.07). Two were related to obstetric factors: pre-eclampsia/hypertension (RR 2.56, 95% CI 2.00-3.27) and labor induction (RR 3.32, 95% CI 2.70-4.10). The remaining two were birth weight (per 250 g, RR 1.12, 95% CI 1.09-1.17) and pre-delivery body mass index (RR, 1.03, 95% CI 1.00-1.05). CONCLUSION: The majority of predictors were not modifiable, relating to obstetric history, maternal age and diabetes severity. Possible modifiable interventions to avoid/improve labor induction, and decrease birth weight and maternal weight gain might decrease risk of Cesarean delivery. Future studies must address these multiple predictors.