A TAVR sizing method for bicuspid aortic valve (BAV) anatomy using the level of implantation at the raphe (LIRA) approach demonstrated favorable procedural and clinical outcomes and good reproducibility across centers, according to results from the SUBLIME study published Sept. 28 in JACC: Cardiovascular Interventions.

The retrospective analysis included 215 patients (mean age, 80 years; 38% women; median Society of Thoracic Surgeons score, 3.25%) with severe aortic stenosis and either Sievers type 1 (n=207) or type 2 (n=8) BAV who underwent TAVR using the LIRA method at six Italian centers.

"The rationale of the LIRA method is based on the observation that in bicuspid anatomy, the plane of maximal raphe protrusion (LIRA plane) represents the level with the greatest interaction between the [transcatheter heart valve] and the surrounding anatomy, where anchoring predominantly occurs," write study authors Barbara Bellini, MD, and colleagues. "By relying on a single integrative measurement, the LIRA perimeter, which is not directly affected by raphe length or calcification pattern, this method may reduce operator-dependent variability and facilitate standardized decision-making in bicuspid anatomy."

Use of the LIRA method resulted in prosthesis downsizing in 74.4% of cases, while technical success was achieved in 97.2% of cases and prosthesis-patient mismatch occurred in just 0.9%.

At 30 days, results showed a 93% device success rate, with moderate or greater paravalvular leak, new permanent pacemaker implantation and stroke occurring in 4.6%, 10.7% and 1.9% of patients, respectively. An exploratory subgroup analysis did not show any significant differences based on valve platform or Sievers type.

Among 140 patients with one-year follow-up, Valve Academic Research Consortium-3 clinical efficacy was observed in 92.8% of cases, while all-cause mortality was 5.7%. In a reproducibility analysis of 61 patients, investigators observed high prosthesis-sizing agreement (90.2% of cases, Cohen's κ=0.85) and perimeter concordance between core laboratory and external centers (intraclass correlation coefficient=0.763; p<0.001). Clinical outcomes were also consistent across centers (p=0.069).

"It must be acknowledged that the LIRA method is relatively easy to adopt and does not require specific imaging skills, provided that image reconstructions in the systolic phase can be clearly identified," write Won-Keun Kim, MD, and Derk Frank, MD, in an accompanying editorial comment. However, "until there is randomized validation for supra-annular sizing, outcomes obtained using the LIRA method remain hypothesis generating and at best confirm its feasibility and safety."