An Exploratory CT‐Based Anatomical Scoring System to Evaluate Suitability for Off‐Label Valve Implantation in Aortic Regurgitation

Objectives We developed an exploratory computed tomography (CT)–based anatomical scoring system for the aortic root in pure aortic regurgitation (PAR), first performed a descriptive device‐level comparison between the on‐label J‐Valve cohort and the combined off‐label Venus A‐valve cohort, and then explored CT anatomical features associated with procedural risk within the off‐label Venus A‐Valve cohort. Methods From January 2017 to January 2020, 300 patients with severe AR treated with TAVR were enrolled in a multicenter registry (J‐Valve group, n = 143; Venus A‐Valve subgroup I, score 4–7, n = 86; Venus A‐Valve subgroup II, score 0–3, n = 71). Device selection was Heart‐Team‐driven and nonrandomized. The primary analysis was a descriptive device‐level comparison between the on‐label J‐Valve cohort and the combined off‐label Venus A‐Valve cohort. Secondary analyses explored outcomes within the Venus A‐Valve cohort according to the empirical CT score and examined anatomical correlates of procedural risk. No adjusted or causal comparison between devices was performed. Results In the primary descriptive device‐level comparison, a 30‐day device success was lower in the combined Venus A‐Valve cohort than in the J‐Valve cohort (95/157 [60.5%] vs 132/143 [92.3%], p < 0.001). In the anatomically favorable Venus A‐Valve subgroup I, a 30‐day device success was higher than in subgroup II but remained lower than in the J‐Valve cohort (71/86 [82.6%] vs 132/143 [92.3%], p = 0.031); the low‐score subgroup II had markedly reduced device success (24/71 [33.8%]). A 1‐year all‐cause mortality was 6.4% in the combined Venus A‐Valve cohort versus 5.6% in the J‐Valve cohort ( p = 0.969). Off‐label Venus A‐Valve use was also associated with substantially higher rates of device malposition requiring a second valve (30.6% vs 1.4%, p < 0.001), new LBBB at discharge (44.6% vs 18.9%, p < 0.001), and permanent pacemaker implantation (12.7% vs 3.5%, p = 0.007). Within the Venus cohort, low‐score anatomy was associated with higher procedural hazard. Conclusion In this registry, device‐level comparisons between J‐Valve and Venus A‐Valve should be interpreted as descriptive only because treatment allocation was nonrandomized, anatomy‐driven, and unadjusted. Off‐label Venus A‐Valve implantation may be feasible in selected patients with favorable CT anatomy, but clinically important safety differences remain, particularly malposition requiring a second valve and conduction disturbances. The proposed CT score is exploratory and descriptive only; formal derivation and external validation are required before routine clinical use. Trial Registration: Chinese Clinical Trial Registry: ChiCTR‐OPC‐17013149