Transcatheter Closure or Surgery for Symptomatic Paravalvular Leaks: The Multicenter KISS Registry

Abstract

Background The optimal treatment of symptomatic paravalvular leak (PVL) remains controversial between transcatheter closure (TC) and surgery. This large‐scale study aimed to retrospectively evaluate the long‐term outcomes of the patients who underwent reoperation or TC of PVLs. Methods and Results A total of 335 (men, 209 [62.4%]; mean age, 58.15±12.77 years) patients who underwent treatment of PVL at 3 tertiary centers between January 2002 and December 2021 were included. Echocardiographic features, procedure details, and in‐hospital or long‐term outcomes were assessed. The primary end point was defined as the all‐cause death during follow‐up. The regression models were adjusted by applying the inverse probability weighted approach to reduce treatment selection bias. The initial management strategy was TC in 171 (51%) patients and surgery in 164 (49%) cases. Three hundred cases (89.6%) had mitral PVL, and 35 (10.4%) had aortic PVL. The mean left ventricular ejection fraction was 52.03±10.79%. Technical (78.9 versus 76.2%; P =0.549) and procedural success (73.7 versus 65.2%; P =0.093) were similar between both groups. In both univariate and multivariable logistic regression analysis, the in‐hospital mortality rate in the overall population was significantly higher (15.9 versus 4.7%) in the surgery group compared with the TC group (unadjusted odds ratio, 3.13 [95% CI, 1.75–5.88]; P =0.001; and adjusted odds ratio (inverse probability‐weighted), 4.55 [95% CI, 2.27–10.0]; P <0.001). However, the long‐term mortality rate in the overall population did not differ between the surgery group and the TC group (unadjusted hazard ratio [HR], 0.86 [95% CI, 0.59–1.25]; P =0.435; and adjusted HR (inverse probability‐weighted), 1.11 [95% CI, 0.67–1.81]; P =0.679). Conclusions The current data suggest that percutaneous closure of PVL was associated with lower early and comparable long‐term mortality rates compared with surgery.