LAUSR

OBJECTIVE To assess the association of maternal BMI with a composite of severe maternal outcomes. STUDY DESIGN Secondary analysis of a cohort of deliveries on randomly selected days at 25 hospitals from 2008-2011. Data on comorbid conditions, intrapartum events, and postpartum course were collected. The reference group (REF, BMI 18.5-29.9), obese (OB, BMI 30-39.9), morbidly obese (MO, BMI 40-49.9) and super morbidly obese (SMO, BMI  50) women were compared. The composite of severe maternal outcomes was defined as death, ICU admission, ventilator use, DVT/PE, sepsis, hemorrhage, DIC, unplanned operative procedure, or stroke. Patients in the REF group were matched 1:1 with those in all other obesity groups based on propensity score using the baseline characteristics of age, race-ethnicity, previous cesarean, pre-existing diabetes, chronic hypertension, parity, cigarette use, and insurance status. Multivariable Poisson regression was used to estimate adjusted relative risks (aRR) and 95% confidence intervals (CI) for the association between BMI and the composite outcome. Because cesarean delivery may be in the causal pathway between obesity and adverse maternal outcomes, models were then adjusted for mode of delivery to evaluate potential mediation. RESULTS A total of 52,162 pregnant patients are included in the analysis. Risk of composite maternal outcomes was increased for SMO compared with REF, but not for OB and MO [OB aRR 1.06 95%CI (0.99-1.14); MO aRR 1.10 (95%CI 0.97-1.25); SMO aRR 1.32 95%CI (1.02-1.70)]. However, in the mediation analysis, cesarean appears to mediate 46% (95%CI 31% - 50%) of the risk of severe morbidity for SMO compared with REF. CONCLUSION Super morbid obesity is significantly associated with increased serious maternal morbidity and mortality; however, cesarean appears to mediate this association. Obesity and morbid obesity are not associated with maternal morbidity and mortality.