Cardiac Computed Tomography-Guided Procedural Planning for Percutaneous Mitral Paravalvular Leak Closure: Impact on...

Mitral paravalvular leaks (PVLs) are a recognized complication of prosthetic valve implantation and may cause heart failure or hemolysis, often requiring percutaneous closure in patients at high surgical risk. Although transesophageal echocardiography provides essential intraprocedural guidance, PVL crossing remains a technically challenging step. Cardiac computed tomography angiography (CCTA) may facilitate this phase by enabling anatomically aligned fluoroscopic projection planning.

This single-center retrospective study included consecutive patients undergoing percutaneous mitral PVL closure after systematic CCTA-based pre-procedural planning. CCTA was used to localize PVLs and derive optimal fluoroscopic projections for crossing. Procedures were classified according to whether CCTA-derived projections were reproducible in the catheterization laboratory. The primary endpoint was PVL crossing time. Secondary endpoints included total fluoroscopy time, radiation dose, and procedural success.

Twenty-eight patients with 38 mitral PVLs were treated. CCTA-derived fluoroscopic projections were feasible in 30 PVLs (78.9%) and not feasible in 8 (21.1%) because of extreme cranio-caudal angulations. PVL crossing was successful in 97.4% of cases and device implantation in 94.7%. Median crossing time was shorter when CCTA-derived projections were feasible compared with non-feasible cases (10.5 vs 32.0 minutes; P = 0.018). Posterior PVLs were associated with the longest crossing times (median 40.0 minutes; P < 0.001) and higher radiation exposure. Total fluoroscopy time was numerically lower in the feasible group but did not reach statistical significance.

CCTA-based pre-procedural planning improves procedural efficiency during percutaneous mitral PVL closure reducing PVL crossing time. Posterior PVLs remain the main determinant of procedural complexity, reflecting intrinsic anatomic and mechanical constraints.