Quantification of epicardial fat volume using low-dose cardiac scan in patients with low calcium score

We investigated the accuracy of quantifying epicardial adipose tissue volume (EATV) using low-dose cardiac scan (EATV_{cardiac scan}) and evaluated its clinical utility in predicting coronary heart disease in patients with low or mild calcification.

In total, 204 patients with clinical symptoms of coronary heart disease and coronary artery calcium score (CACS) of <100 AU were enrolled in this retrospective study. After obtaining EATV_{cardiac scan} and EATV measured using computed tomography angiography (EATV_CTA), the agreement between the two measurements was evaluated using Pearson correlation coefficient and Bland-Altman analysis. Multivariate logistic regression was used to analyze the utility of EATV in predicting plaque and vulnerable plaque. Receiver operating characteristic curves were constructed.

The mean EATV_{cardiac scan} (101.51±41.57 cm³) and EATV_CTA (104.57±41.34 cm³) of all patients were similar, and the two measurements were strongly correlated (r=0.9596, P<0.001). The difference between EATV_{cardiac scan} and EATV_CTA was -3.0549, with only 4.9% (10/204) of patients having values outside the 95% confidence interval (CI) range (-26.15 to 20.04; P for agreement =0.0003). Further, a significant agreement was observed between EATV_{cardiac scan} and EATV_CTA in 126 patients with plaques, with an estimated difference of -3.354, and 6.35% (8/126) of patients had values outside the 95% CI range (-31.37 to 24.66; P for agreement =0.0095). After adjustment for age and sex, EATV_{cardiac scan} and EATV_CTA were significantly associated with plaque (all P values <0.001), and the areas under the curve (AUCs) were 0.662 and 0.670 (P=0.4331), respectively. In contrast, EATV_{cardiac scan} and EATV_CTA were not associated with vulnerable plaque (P>0.05), with AUCs of 0.550 and 0.530, respectively (P=0.2157).

The study results indicate that EATV_{cardiac scan} and EATV_CTA are equivalent. In addition, both methods provide comparable values for predicting coronary arteriosclerosis in patients with low-to-mild calcification (CACS of <100 AU).