Newly Proposed Body Adiposity Index (BAI) by Bergman et al. Is Not Strongly Related to Cardiovascular Health Risk

Abstract

TO THE EDITOR: With great interest, we read the paper of Bergman et al., in which a new indicator of body adiposity was proposed (1). The body adiposity index (BAI) was developed from hip circumference and height (BAI = (hip circumference/height1.5) — 18), based on their correlations with each other and with adiposity measured by dual-energy X-ray absorptiometry (gold standard for adiposity). The authors showed that the new BAI predicted % body fat very well in both African-Americans and Mexican Americans. Bergman et al. concluded in their paper that although the BAI performed well as a predictor of total body fat percentage, it remained to be seen whether it is equally useful in predicting health outcomes compared to other indexes of adiposity. We were able to investigate this using data from the SUNSET study including three ethnic groups in the Netherlands: African-origin Surinamese, South Asian-origin Surinamese, and white Dutch (2,3). The study was based on a sample of 35–60-year old, noninstitutionalized people living in Amsterdam, the Netherlands. In all participants, weight, height, waist and hip circumferences, and blood pressure were measured. A fasting blood sample was taken for the measurement of lipids and glucose. We calculated BMI, waist-to-hip ratio, waist-to-height ratio, and the newly proposed BAI and studied the correlation of the anthropometric measures with the following metabolic risk factors: high-density lipoprotein, low-density lipoprotein, triglycerides, glucose, systolic blood pressure, and diastolic blood pressure. For the present analyses, 1,440 subjects (58.8% female) were included with a mean age of 45.4 ± 6.6 years, mean BMI of 27.3 ± 5.2 kg/m2, and mean BAI of 29.0 ± 6.2. So, on average our population was slightly older, whereas BMI was somewhat lower, than the population described by Bergman et al., but broad ranges were included in both studies. The results in Table 1 show that BMI, waist circumference, waist-to-hip ratio, and waist-to-height ratio were more strongly correlated with all studied metabolic risk factors than BAI across ethnic groups. So, although Bergman et al. found that BAI is a good predictor of adiposity and argue that it is a more practical and simple measure than BMI (because no weight measurement is required), the results of our study suggest that the BAI may be less useful than BMI to assess metabolic health risk. This should be taken in consideration when choosing between these two adiposity measures. As Bergman et al. themselves also state, previous work has shown that body fat distribution (rather than total adiposity) has been shown a stronger predictor of cardiovascular risk, due to metabolic differences in abdominal (particularly visceral) fat vs. gluteal fat (4,5,6). This is in accordance with our current results, where waist circumference, waist-to-hip ratio, or waist-to-height ratio were generally stronger correlates of metabolic risk factors than BMI. Since no height but only tape measurements are required, we suggest that waist circumference and waist-to-hip ratio may be even better candidates than BMI or BAI, when a simple and practical indicator of cardiovascular health risk is needed. The authors declared no conflict of interest.