Transcatheter electrosurgical leaflet splitting to prevent TAVI-induced coronary obstruction in stented vs. stentless bioprosthetic aortic valves: Results from the multicenter EURO-BASILICA registry

Background: Transcatheter electrosurgical leaflet splitting prevents ostial coronary artery obstruction following transcatheter aortic valve implantation (TAVI) and is more commonly applied when TAVI is used for treatment of failed bioprosthetic tissue valves (valve-in-valve). Stented and stentless biological valves have distinct morphological and fluoroscopic characteristics, which may differentially impact procedural technique and outcomes of patients undergoing leaflet splitting, but comparative data are currently lacking. 

Objectives: To compare procedural characteristics and outcomes of stented vs. stentless surgical valves in patients undergoing bioprosthetic or native aortic scallop intentional laceration to prevent iatrogenic coronary artery obstruction (BASILICA) and TAVI in a multicenter registry.

Methods: A total of 93 patients undergoing BASILICA and valve-in-valve TAVI from the multicenter EURO-BASILICA registry were included. Patients were stratified according to bioprosthetic valve type (stented vs. stentless), and outcomes were described according to the Valve Academic Research Consortium 3 (VARC-3) recommendations. 

Results: Overall, 82 (88.2%) patients had a stented surgical valve (most common types Sorin Mitroflow in 29 patients [35.37%] and St. Jude Medical Trifecta in 28 patients [34.2%]) and 11 (11.8%) had a stentless surgical valve (most common type Medtronic Freestyle in four patients [36.4%]). Risk assessment for TAVI-induced coronary artery obstruction (CAO) showed similar valve-to-coronary (VTC) (3.0 mm [interquartile range (IQR) 2.4-3.6] vs 3.71 mm [IQR 2.7-4.1], p=0.260) and valve-to-sinutubular-junction (VTSTJ) distances (1.5 mm [IQR 0.0-2.9] vs 1.5 mm [1.0-2.3], p=0.749), but coronary height was lower in patients with stented valves (6.8±2.6 mm vs 9.0 ±2.2 mm, p=0.006). Stentless valves had significantly longer leaflets compared to stented valves (17.9 mm [IQR 16.2-20.4] vs 14.0 mm [IQR 12.5-15.8, p<0.001), but the predicted mechanism for CAO was similar. Double-BASILICA was performed in 12 (12.9%) patients. Procedural details were similar, but procedures were longer for stentless valves compared with stented valves (150 min [IQR 109.09-181.00] vs 125 min [IQR 90.00-160.25], p=0.214) and there was a higher use of contrast medium in patients with stentless valves (235 ml [IQR 162.0-300.0] vs 90 ml [IQR 61.5-110.5], p<0.001). Technical success was similar, but stentless valves were associated with a significantly higher rate of target leaflet-related coronary obstruction compared with stented valves (36.4% vs 3.7%, p=0.021), mainly due to a higher rate of partial coronary obstruction (25.0% vs. 1.1%, p<0.001). Other in-hospital complications were similar. Freedom from 30-day-safety endpoints was higher with stented valves compared with stentless valves (86.6% vs. 54.5%, p=0.020).

Conclusions: BASILICA is feasible for both stented and stentless bioprosthetic tissue valves. However, stentless valves are associated with a higher residual risk for coronary obstruction after BASILICA and TAVI and may therefore require additional or alternative coronary artery protection measures in selected patients.