Cross-sectional evaluation of anthropometric measures vs MRI-measured visceral and hepatic fat in a Canadian cohort

Abstract Context Accurate assessment of excess body fat and its cardiometabolic risk is essential in clinical and epidemiologic research. Body-mass index (BMI), although widely used, does not capture visceral or ectopic fat. Objective To evaluate how anthropometric measures—BMI, percent body fat, waist circumference (WC), and waist-to-hip ratio (WHR)—relate to MRI-measured visceral adipose tissue (VAT) and hepatic fat fraction (HFF), overall, and by sex. Design Cross-sectional analysis within the Canadian Alliance for Healthy Heart and Minds cohort. Setting Community-based, pan-Canadian prospective study. Patients or Other Participants 6683 apparently healthy adults (mean age 57 ± 9 years; 3665 females) with baseline anthropometric and MRI measures; analyses adjusted for study center. Intervention(s) Not applicable. Main Outcome Measure(s) MRI-derived VAT and HFF. Associations were estimated using linear regression and mixed models, stratified by sex, ethnicity, age, and BMI category. Results Mean BMI was 26.7 kg/m2, WC 88.3 cm, VAT 71 mL, and HFF 5.7%. Females had lower VAT and HFF than males. Correlations with VAT ranged from 0.35 to 0.77 and with HFF from 0.26 to 0.47. Each 10 cm higher WC was associated with 21.0 mL higher VAT and 2.1% higher HFF; each 5 kg/m2 higher BMI predicted 24.7 mL and 2.8% higher values. When modeled jointly, WC remained strongly predictive, while BMI contributed modestly. Associations were smaller in females. VAT and HFF increased across WC tertiles within BMI categories. Conclusion WC is a robust surrogate for visceral and hepatic fat across BMI categories, supporting its use when MRI is not feasible.