Paravalvular leak closure: Still a challenge with unpredictable results

Abstract

IntroductionParavalvular leak (PVL) is a common serious complication associated with prosthetic valve implantation. ObjectiveThe aim of this study was to report our single-center experience in a retrospective review and to analyze possible predictors of success. MethodsWe performed 33 percutaneous PVL closures in 26 patients (54% female, mean age 65±13 years). All mitral prostheses were studied previously with 3D transesophageal echocardiography (TEE), and aortic prostheses with 2D/3D TEE. 3D TEE and fluoroscopy were used for the assessment, planning, and guidance of the interventions. Twelve patients also underwent computed tomography angiography for better characterization of anatomic details. ResultsEighteen patients (69.2%) were admitted due to heart failure (New York Heart Association [NYHA] III or IV, seven (26.9%) because of heart failure and hemolysis, and one (3.8%) due to hemolysis only. Regarding the leaks, 46.2% were in aortic and 53.8% in mitral prostheses, 88.5% in mechanical and 7.7% in biological prostheses, and 3.8% in transcatheter aortic valve implants. All the aortic patients had severe aortic regurgitation. Furthermore, all mitral patients but one had moderate to severe or severe mitral regurgitation. Closure was successful in 17 patients (65.4%), partially successful in four (15.4%) and unsuccessful in five (19.2%). After the procedure, 69% were in NYHA I-II. Hemolysis worsened in three patients despite successful closure; all required further valvular surgery and two died. Regarding angiographic and echocardiographic procedural success, we analyzed age, gender, type of prosthesis (mechanical or biological), location (aortic or mitral), clinical data, maximum leak diameter, anatomic regurgitant orifice, leak location (anterior, posterior, inferior and lateral for mitral leaks and left, right and non-coronary sinus for aortic leaks), and number of devices (plugs) used for closure. No parameters presented a significant relationship with success excepting previous hemolysis. There was a relationship between clinical improvement and reduction of PVL (p=0.0001). In follow-up, cardiac-related events (new hospital admissions, cardiac valvular surgery, need for transfusion) were more frequent in patients with partially successful or unsuccessful closure (p=0.012). There was a relationship between cardiac-related events and death (p=0.029). ConclusionPercutaneous PVL closure has emerged as an alternative treatment for PVL. Predictors of procedural success are difficult to establish. Survival is related to reduction of regurgitation and improvement in NYHA functional class.