Association between World Health Organization categories of body mass index and relative risks for weight-related pregnancy outcomes: a retrospective cohort study

Abstract

To analyse the dose-dependent effect of body mass index (BMI) categories for common pregnancy outcomes. A retrospective cohort study of all deliveries that occurred between 1 January 2005 and 31 December 2009 in a tertiary maternity centre, in Sydney Australia. Common pregnancy outcomes were analysed against World Health Organization (WHO) BMI categories using multiple logistic regression analysis. From a total of 18,304 pregnancies, 9087 singleton pregnancies with complete data-sets were identified. Of these pregnancies, 4000 (44%) had a normal BMI, 470 (5.2%) were underweight, 2293 (25.2%) were overweight, 1316 (14.5%) were obese class I, 630 (6.9%) were obese class II and 378 (4.2%) were obese class III. Using the normal BMI category as the reference, there was a clear dose effect of BMI categories for hypertension (P < 0.001), pre-eclampsia (P < 0.001), caesarean section (P < 0.001), macrosomia (P < 0.001), large for gestational age (P < 0.001), small for gestational age (P < 0.001) and neonatal respiratory distress (P = 0.039). In contrast, despite a significant association with BMI (P < 0.001), a dose-dependent effect was not found for gestational diabetes. The results of our study have important clinical significance as the data, using WHO BMI categories, more accurately help stratify risk assessment in a clinically relevant dose-dependent relationship.