LAUSR

Background Surgical aortic bioprosthetic valves, over time, are associated with structural deterioration. Valve-in-valve (ViV) transcatheter aortic valve implantation (TAVI) has emerged as a less-invasive alternative to redo surgery for aortic bioprosthetic valve failure, especially in elderly patients or those with multiple comorbidities at elevated redo operative risk. ViV TAVI poses technical challenges that vary depending on the original aortic bioprosthesis and the TAVI bioprosthesis intended for use. We report indications, procedural characteristics and mid-long-term outcomes of all patients undergoing ViV-TAVI for aortic bioprosthetic failure, in a single centre. Methods 21 consecutive patients with failing stentless (n=5), stented porcine valve (n=4), stented pericardial valve (n=12) were treated with ViV-TAVI between 2010 and 2017. Baseline characteristics included a mean age of 80±8.6 years, mean logistic EuroScore of 31.4%±18% and mean time from index operation to ViV-TAVI 11±3.9 years (33–264 months). All patients were symptomatic at NYHA class 3 or 4. 60% of patients had severe bioprosthetic regurgitation, 19% bioprosthetic stenosis and 21% mixed disease. 24% of patients had moderate or worse left ventricular systolic dysfunction. 33% patients had undergone prior coronary bypass grafting, 10% had a history of significant pulmonary disease and 9% had suffered a prior stroke. All patients underwent TAVI with a self-expanding device, performed percutaneously via the femoral artery under local anaesthetic. Results Perioperative and 30 day mortality was 0%, with 1 year mortality of 14%. 19% (n=4) patients died during a median follow-up period of 24±28.7 months. Only one of these deaths was cardiac in nature, occurring due to refractory heart failure in the context of a peak gradient of 70 mmHg, after ViV-TAVI within a small 21 mm Sorin Mitroflow prosthesis. Mean aortic gradient, after ViV-TAVI, was 14±11 mmHg. 9.5% (n=2) patients had moderate paravalvular regurgitation on pre-discharge post-TAVI transthoracic echocardiogram (TTE). 0% of patients required a permanent pacemaker, suffered cardiac tamponade, or had a major vascular access injury. There was only one recorded stroke. Device embolization or migration was not observed in any patient. Mean TTE follow-up was 23 months, for which the mean aortic gradient remained at 14±11 mmHg, and there was 0% incidence of structural valve deterioration. Only one patient had a cardiac readmission within 1 year of ViV-TAVI. Mean NT-proBNP level prior to ViV-TAVI was 5567 pg/L, falling to 1463 pg/L one year after the procedure. Conclusion Our experience demonstrates that ViV-TAVI is an effective alternative procedure for high-risk redo surgical patients, associated with low peri-procedural, 30 day and 1 year mortality. Our echocardiographic outcomes suggest good durability of the ViV bioprosthesis, and our low readmission rate, combined with the NT-proBNP trend suggest a favourable clinical response also.