BMI-C como preditor de Diabetes Mellitus tipo 2 - Validação em duas coortes independentes do NHANES (2015-16 e 2017-18)
Year: 2026
Type: preprint
Abstract: Objective: To compare the performance of BMI-C (Body Morphometric Index - Corrected), a novel proprietary morphometric index, with VAI (Visceral Adiposity Index) in predicting visceral adiposity and insulin resistance, using a representative US population dataset. Methods: Cross-sectional study with 1,630 participants from NHANES 2015-16 (aged >= 20 years). BMI-C was calculated via a proprietary algorithm available at bmi-c.com. VAI was calculated according to Amato et al., 2010. The gold standard for visceral fat was trunk fat by DEXA. Insulin resistance was estimated by fasting insulin. Pearson correlations, ROC curves and area under the curve (AUC) were calculated, along with concordance between indices at the 1.0 threshold. Results: BMI-C showed correlation with DEXA of 0.939 (vs. 0.324 for VAI; p<0.001) and with insulin of 0.459 (vs. 0.348; p<0.001). AUC for predicting high risk (DEXA > P75) was 0.971 for BMI-C >= 1.0 but VAI < 1.0. Among participants with normal BMI (<25 kg/m²), 524 (32.1% of the total sample) had BMI-C >= 1.0, with insulin 23.5% higher and trunk fat 103.8% higher than the healthy group (both indices < 1.0), and with visceral fat levels comparable to overweight individuals. Conclusion: BMI-C is superior to VAI in the early detection of visceral adiposity and insulin resistance, identifying a substantial subgroup of normal-BMI individuals who already have silent metabolic risk. As a low-cost index that does not require laboratory tests, BMI-C may be a useful tool for metabolic screening in primary care, especially in resource-limited populations.
Author: Brochieri, André Luís Silva
Language: English
Subfield: Endocrinology, Diabetes and Metabolism
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7208836625