Association between epicardial adipose tissue attenuation and long-term clinical outcomes after transcatheter aortic valve replacement
Y Manabe, H Kiriyama, T Furusawa, T Kamon, S Minatsuki, H Kadowaki, A Saito, S Kodera, M Ando, H Yamauchi, K Nakanishi, J Ishida, M Ono, N TakedaAbstract
Background
Epicardial adipose tissue (EAT) is a fat depot between the epicardial surface of the myocardium and the visceral pericardium and it has been proposed to play a crucial role in cardiac inflammation. Accumulating evidence links EAT to the development of various cardiovascular diseases, including coronary artery disease and atrial fibrillation. Aortic stenosis is one of the cardiovascular diseases that has gained increased attention in recent years due to the widespread adoption of transcatheter aortic valve replacement (TAVR); however, long-term mortality after TAVR remains substantial, highlighting the need for additional imaging biomarkers for risk stratification. Prior studies have demonstrated an association between EAT volume and prognosis after TAVR, whereas CT-derived EAT attenuation may provide complementary information by reflecting tissue composition and inflammatory activity. Nevertheless, evidence regarding the association between EAT attenuation and clinical outcomes after TAVR is limited, and its relationship with long-term prognosis remains unclear.
Purpose
We aimed to clarify the association between EAT attenuation and long-term clinical outcome in patients with severe aortic stenosis who underwent TAVR.
Methods
We retrospectively analyzed 496 patients who underwent TAVR at our hospital between October 2015 and June 2025. We excluded patients with unavailable CT data and those whose CT images were of insufficient quality for analysis. Finally, 415 cases were included. For analysis of EAT, we used REVORAS. Pre-TAVR CT images were reconstructed using REVORAS and EAT attenuation values were extracted. EAT was segmented from the level of pulmonary trunk bifurcation to the diaphragm, using a predefined attenuation threshold of -190 to -30 Hounsfield Units (HU). We utilized the dilated phase data of each pre-TAVR CT image. The primary outcomes were a composite of all-cause death, cardiovascular death, and heart failure hospitalization, and the secondary outcome was all-cause death.
Results
The mean age of entire cohort was 83.8±6.0 years and 258 patients (62.2%) were female. Balloon-expandable valves were implanted in 271 patients (65.3%). The mean follow-up duration was 882±650 days. The mean EAT attenuation was -74.6±6.7 HU, with a median of -75.0 HU. Based on a median value of EAT attenuation, patients were classified into two groups: High-attenuation EAT group (≥-75.0 HU) and Low-attenuation EAT group (<-75.0 HU). Kaplan-Meier analysis revealed that primary and secondary outcomes were significantly higher in the High-attenuation group (both log-rank p <0.01) (Figure).
Conclusions
In this single-center TAVR cohort, higher pre-procedural CT-derived EAT attenuation was associated with significantly worse long-term outcomes. These findings suggest that EAT attenuation may serve as a simple imaging biomarker for post-TAVR risk stratification and long-term prognostication.Kaplan-Meier Curve