Association between abdominal fat distribution, coronary artery calcium score, and all-cause mortality: data from a large cohort of patients submitted to cardiac CT scan

Abstract Background Obesity is a well-established risk factor for coronary artery disease (CAD) and all-cause mortality (ACM), yet traditional measures like body mass index (BMI) often fail to capture obesity-related cardiovascular risk. Emerging evidence suggests that abdominal fat distribution (AFD), particularly the balance between visceral (VAT) and subcutaneous adipose tissue (SAT), may better predict CAD than BMI alone. Aim This study evaluated the association between AFD ratios and coronary artery calcium (CAC) score, measured by coronary computed tomography angiography (CCTA), and their impact on long-term ACM. Methods Retrospective analysis of 1709 patients who underwent CCTA for CAD assessment between 2006 and 2023. A single-slice abdominal CT scan at the L4/L5-S1 level was also performed. CAC scores were extracted, and abdominal fat volumes (visceral and subcutaneous) were quantified using voxel filtering with attenuation values between -150 and -50 Hounsfield units. Patients were stratified into two groups: CAC = 0 (no CAD) and CAC 0 (evidence of CAD). The VAT/SAT ratio, VAT to total abdominal area ratio (VAT/TotAbd), and SAT to total abdominal area ratio (SAT/TotAbd) were analyzed. Logistic regression assessed the association between AFD ratios and CAC score, while Cox proportional hazards models evaluated the impact of AFD ratios on ACM. Results The study cohort had a mean age of 58±10 years, with 57% being male. Of these, 41% had a CAC score of 0. Logistic regression identified VAT/SAT ratio as the strongest predictor of CAC0 (OR= 1.29, p0.001), with a 30% increase in risk per unit rise in VAT/SAT. In contrast, a higher SAT/TotAbd ratio demonstrated a protective effect (OR=0.8, p=0.001), while VAT/TotAbd ratio was not significantly associated with CAC score (OR=1.1, p=0.2). Among clinical predictors, male sex was the strongest determinant of CAC0 (OR= 3.53–4.29, p 0.001), with men being 4 times more likely to have CAC0. As expected, age (OR=1.10, p0.001), diabetes (OR=2.05–2.19, p0.001), and dyslipidemia (OR=1.63–1.72, p0.001), were also significant predictors of the presence of CAD. Cox proportional hazards models revealed that CAC0 was a strong predictor of ACM, doubling the risk (HR=1.96–2.04, p0.01). A higher SAT/TotAbd ratio was associated with lower mortality risk (HR=0.80, p=0.004), while VAT/SAT ratio (HR=1.09, p=0.4) and VAT/TotAbd ratio (HR=0.88, p=0.2) showed no significant association. Conclusion This study highlights the distinct roles of VAT and SAT in the risk of CAD, demonstrating that VAT/SAT ratio was the strongest predictor of CAC0, suggesting that VAT is more metabolically active and contributes to CAD. On the other hand, SAT/TotAbd ratio appears to have a protective effect on long-term ACM. These findings support that the integration of AFD analysis into routine cardiovascular assessments may enhance early risk identification and inform personalized prevention strategies.