Association of Body Shape Index (ABSI) with cardio-metabolic risk factors: A cross-sectional study of 6081 Caucasian adults

A Body Shape Index (ABSI) was specifically developed as a transformation of waist circumference (WC), statistically independent of BMI to better evaluate the relative contribution of WC to central obesity and clinical outcomes. Previous studies have found ABSI is associated with total mortality and cardiovascular events. However, no study has specifically evaluated the joint contribution of ABSI and BMI to cardio-metabolic outcomes (high triglycerides, low HDL, high fasting glucose and high blood pressure). With this aim, we performed a retrospective study on 6081 Caucasian adults. Subjects underwent a medical interview, anthropometric measurements, blood sampling, measurement of blood pressure, and measurement of visceral abdominal fat thickness (VAT) by ultrasound. Generalized linear models (GLM) were used to evaluate the sex and age adjusted association of ABSI with binary and continuous cardio-metabolic risk factors. Four pre-specified GLM were evaluated for each outcome: M1 = ABSI, BMI and ABSI*BMI interaction, M2 = ABSI and BMI, M3 = ABSI alone and M4 = BMI alone. Bayesian Information Criterion (BIC) was calculated and used to identify the best predictive model. ABSI and BMI contributed independently to all outcomes. Compared to BMI alone, the joint use of BMI and ABSI yielded significantly improved associations for having high triglycerides (BIC = 5261 vs. 5286), low HDL (BIC = 5371 vs. 5381), high fasting glucose (BIC = 6328 vs. 6337) but not high blood pressure (BIC = 6580 vs. 6580). The joint use of BMI and ABSI was also more strongly associated with VAT than BMI alone (BIC = 22930 vs. 23479). In conclusion, ABSI is a useful index for evaluating the independent contribution of WC, in addition to that of BMI, as a surrogate for central obesity on cardio-metabolic risk.