Conventional redo biological valve replacement over 20 years: Surgical benchmarks should guide patient selection for transcatheter valve-in-valve therapy

Abstract

ObjectivesAlthough primary transcatheter valve interventions have demonstrated acceptable early- and intermediate-term outcomes, data are lacking to guide patient selection for transcatheter valve-in-valve therapy. Furthermore, very few surgical benchmarks have been established for repeat conventional biological valve replacement to refine momentum for broad application of transcatheter intervention for a degenerated bioprosthesis. MethodsFrom January 1993 to July 2014, 694 patients underwent repeat biological valve replacement at our clinic. Median age at repeat operation was 71 years (range, 26-95 years) and there were 437 men (63%). Hypertension was present in 453 patients (65%), diabetes in 128 patients (18%), prior myocardial infarction in 85 patients (12%), and prior stroke in 81 patients (12%). Prior coronary bypass grafting was performed in 212 patients (31%). Median left ventricular ejection fraction was 41% (range, 20-61) and New York Heart Association functional class III or IV was present in 529 patients (76%). ResultsBiological valve re-replacement included most commonly aortic valve in 464 patients (67%) and mitral valve in 170 (24%). Concomitant coronary bypass grafting was performed in 134 patients (19%). Mortality at 30 days occurred in 56 patients (8%). Multivariable analysis with backward stepwise regression identified New York Heart Association functional class (per 1 increment) (hazard ratio, 2.1; 95% confidence interval, 1.06-4.3; P = .03) and prior coronary bypass grafting (hazard ratio, 3.5; 95% confidence interval, 1.2-10.9; P = .03) as independent predictors of early death. Patients with the combination of prior coronary bypass grafting and New York Heart Association functional class III or IV accounted for 26 out of 56 early deaths (46%) and in the absence of this combination, early death in the cohort was 30 out of 694 (4%). Follow-up was available in 602 out of 638 early survivors (94%) for a median of 45 months (range, 1 month-23.4 years). Survival at 5 and 10 years was 63% and 34%, respectively. For patients who died during follow-up, 2-dimensional scatter plots demonstrate durable length of postoperative survival (median, 5.5 years; maximum, 22 years). ConclusionsIn a large population of patients undergoing repeat biological valve replacement, prior coronary bypass grafting and advanced New York Heart Association functional class were associated with increased 30-day mortality, with the remaining population having a low 30-day mortality of 4%. This study could serve as a surgical benchmark to guide patient selection for transcatheter valve-in-valve technology rather than employing a broader application of these techniques to those who may otherwise have low early risk of mortality and durable long-term survival after conventional valve surgery.