Exploring comparative assessment of adiposity...

Exploring comparative assessment of adiposity measures during pregnancy and postpartum

Abstract

BACKGROUND & AIMS: Excessive adiposity in pregnancy is associated with an altered cardiometabolic profile and adverse maternal and offspring outcomes. Pre-pregnancy body mass index (pBMI) is a proxy measure for adiposity that is most often used in clinical settings; however, it may not identify at-risk pregnancies caused by adiposity-related cardiometabolic dysfunction. The challenge is that validated direct adiposity measures are limited due to the dynamic nature of pregnancy. This exploratory analysis aimed to, 1) evaluate longitudinal changes in % body fat (BF) and the agreement between skinfold thickness (SFT) and bioelectrical impedance analysis (BIA) across pregnancy and in postpartum; 2) compare the discrimination power of SFT, BIA, and pBMI regarding adiposity status; and 3) assess agreement between SFT and BIA with dual-energy X-ray absorptiometry (DXA) in the postpartum. METHODS: Participants enrolled in the Be Healthy in Pregnancy (BHIP) RCT (NCT01693510) had demographic data and pBMI collected at enrollment and adiposity measured at 12-17, 26-28, and 36-38 weeks gestation by BIA (%BF) and SFT (sum and %BF), and also by DXA at 6 months postpartum. Agreement between methods was assessed by Bland Altman plots and McNemar's test and C-statistic for discriminative power. RESULTS: In 181 women with mean pBMI of 25.1 kg/m² (min: 17.4 kg/m², max: 39.6 kg/m²) and age 31.6 (SD: 4.0 yr), maternal adiposity increased significantly (p < 0.001) across pregnancy when measured by the sum of SFT or %BF by BIA, but not %BF by SFT. In early pregnancy, BF by BIA and SFT showed good agreement, with BIA values 1.8% greater than SFT, but low agreement in late pregnancy, with BIA values 7.1% greater than SFT. However, in the postpartum, agreement was similar to early pregnancy, and both BIA and SFT demonstrated good agreement with DXA. By pBMI, 45.5% of participants were categorized as overweight/obese, compared to 66.5% by BIA (p < 0.0001) and 54.5% by SFT (p < 0.0001). CONCLUSIONS: In comparison to SFT and BIA, the results suggest that pBMI is less sensitive in identifying participants with excessive adiposity, limiting its use as a screening tool for adiposity-related adverse outcomes in pregnancy. It would be preferable to use a direct measure of adiposity to screen for at-risk pregnancies. Both %BF by BIA and sum of SFT can quantify the change in adiposity across pregnancy and in the postpartum and thus could be adopted as clinical practice tools. Future research efforts should further refine and validate adiposity techniques for use, particularly in mid and late pregnancy. CLINICAL TRIAL: The BHIP clinical trial (NCT01693510). REGISTRATION SITE: https://clinicaltrials.gov/ct2/show/NCT01693510.