Delivery outcomes in women with body mass index of 40 kg/m2 or greater stratified by planned and actual mode of delivery: a systematic review and meta-analysis.

INTRODUCTION Pregnant women with a body mass index (BMI) ≥40kg/m2 are at an increased risk of requiring planned- and unplanned cesarean deliveries (CD). The aim of this systematic review is to compare delivery outcomes in women with BMI ≥40kg/m2 based on planned and actual mode of delivery. MATERIAL AND METHODS Five databases were searched for English and French-language publications until February 2019, and all studies reporting on delivery outcomes in women with BMI ≥40kg/m2 , stratified by planned and actual mode of delivery, were included. Risk-of-bias was assessed using the Newcastle-Ottawa Scale. Relative risks (RR) and 95% confidence intervals were calculated using random-effects meta-analysis. RESULTS Ten observational studies were included. Anticipated vaginal birth vs. planned CD (5 studies, n=2216) was associated with higher risk for postpartum hemorrhage (13.0% vs. 4.1%, p<0.001, numbers needed to harm (NNH=11), I2 =0%) but lower risk for wound complications (7.6% vs. 14.5%, p<0.001, numbers needed to treat (NNT=15), I2 =58.3%). Planned trial of labour vs. repeat CD (3 studies, n=4144) was associated with higher risk for uterine dehiscence (0.94% vs. 0.42%, p=0.04, NNH=200, I2 =0%), endometritis (5.1% vs. 2.2%, p<0.001, NNH=35, I2 =0%), prolonged hospitalization (one study, 30.3% vs. 26.0%, p=0.003, NNH=23), low five-minute Apgar scores (4.9% vs. 1.7%, RR 2.95 (2.03, 4.28), NNH=30, I2 =0%) and birth trauma (1.1% vs. 0.2%, p<0.001, NNH=111, I2 =0%). Successful vaginal birth vs. intrapartum CD (n=3625) was associated with lower risk of postpartum hemorrhage (15.1% vs 70%, p<0.001, NNT=2, I2 =0%), wound complications (one study, 0% vs. 4.4%, p=0.007, NNT=23), prolonged hospitalization (one study, 1.9% vs. 6.7%, 0.04, NNT=21) and low five-minute Apgar scores (one study, 1.0% vs. 5.6%, p=0.03, NNT=22), but more birth trauma (5.9% vs. 0.6%, p=0.005, NNH=19, I2 =0%). Compared groups had dissimilar demographic characteristics. Although studies scored 6-7/9 on risk-of-bias assessment, they were at high-risk for confounding by indication. CONCLUSIONS Evidence from observational studies suggests clinical equipoise regarding the optimal mode of delivery in women with BMI ≥40kg/m2 and no prior CD, which is best addressed by a randomized trial. Based on an unplanned subgroup analysis, for women with BMI ≥40kg/m2 and prior CD, repeat CD may be associated with better clinical outcomes.