Characteristics and outcomes of patients with eccentric aortic valve calcification undergoing transcatheter aortic valve implantation (TAVI)

Abstract Background TAVI has become the standard therapy for most patients with aortic valve stenosis. Eccentric aortic valve calcification may pose a challenge during TAVI by interfering with precise valve implantation and full valve expansion. This study aims to compare definitions for eccentric aortic valve calcification and to evaluate their impact on the clinical characteristics and outcomes of TAVI patients. Methods We conducted a multicentric study analysing 1748 patients with severe and symptomatic aortic valve stenosis who underwent transfemoral TAVI with a tricuspid aortic valve between June 2012 and June 2024. Preprocedural CT scans were analysed using 3mensio Medical Imaging to measure the calcium volume of each valve cusp. We compared ∆Calcium Score (Nakajima et al., 2021) and Eccentricity Index (Li et al., 2021). To address the limitations of CS and EI, we developed a new scoring system based on the eccentricity and overall calcium distribution, calculated by subtracting cusp calcification from the patient's mean and adding the squared differences (Eccentricity Score = (mean - LCC)² + (mean - RCC)² + (mean - NCC)²). Z-standardization was applied for uniform scaling. This score was further tested in Cox regressions and generalized linear models (GLMs) to assess its predictive value. Endpoints included VARC-3 criteria for new pacemaker implantation, paravalvular leakage, conduction disturbances and survival. Results Comparing CS and EI, we found that EI mainly classified patients with low total calcium as eccentric, while CS focused on the most and least calcified cusps, resulting in the inclusion of many patients with two heavily calcified cusps. The EI eccentric group had larger aortic valve areas (AVA) at baseline (0,76 cm vs. 0,73 cm, p = 0,008). The CS eccentric group was younger (81,7 vs. 82,6 years, p = 0,037), had less atrial fibrillation, fewer cases of CAD, hypertension and dyslipidemia (p 0,05). EI showed no significant differences in outcomes or survival, whereas the CS eccentric cohort showed improved 1-year survival (86% vs. 83%, p = 0,001). While our score did not predict overall mortality, it was associated with lower mortality in cases where the right coronary cusp was the most calcified (OR =0,166, p = 0,0424). GLMs revealed a negative correlation between the score and paravalvular leakage (OR = 0,7, p = 0,04). Further GLMs found no significant associations between the score and other tested outcome factors. Conclusion CS and EI take heterogenous patient groups into account, with EI excluding and CS emphasizing high calcium levels. To adress this, we propose a revised eccentricity defintion using a continuous score for a more precise risk assessment. However, the effect of eccentricity heavily depends on the definition applied and is likely a common finding in calcific aortic stenosis without significant clinical impact.