Immediate, Early and Late Failure after Transcatheter Aortic Valve Implantation: How to Deal with the Inoperable?

BACKGROUND: Currently, the use of transcatheter aortic valve implantation (TAVI) is constantly increasing, whilst cardiosurgical back-up varies substantially. Besides immediate conversion to surgical aortic valve replacement (SAVR) for periprocedural complications, SAVR for TAV failure may be necessary within the early or late post-implant course. The etiology, incidence, risk-stratification, management and outcome for both scenarios are largely unclear. The study aim was to provide details of the authors' experience of SAVR after the failure of TAVI at a single institution.

METHODS: Nineteen patients (14 males, five females) underwent SAVR after TAVI at the authors' institution between June 2008 and December 2015. The patients' initial EuroSCORE II was 8.54 ± 9.81. In eight cases (42%; 50% transfemoral) an immediate conversion was necessary due to paravalvular leakage and insufficiency (n = 1), valve-malpositioning (n = 1), valve dislocation (n = 3), valve-trapping in mitral chordae (n = 1), and annular rupture (n = 2). The 50% transfemoral EuroSCORE II was 19.06 ± 8.61. In 11 patients transcatheter valve failure occurred at a mean of 18 ± 17 months after TAVI (two patients with structural valve failure and one with severe paravalvular leakage, seven with prosthetic valve endocarditis, and one patient with aortic aneurysm); the mean EuroSCORE II was 13.42 ± 13.06.

RESULTS: For immediate conversion, the cardiopulmonary bypass (CPB) time and aortic cross-clamp time were 104 ± 40 min and 60 ± 16 min, respectively. Concomitant procedures were necessary in two patients, one patient required hypothermic circulatory arrest (HCA) and one died intraoperatively. For early and late failure, the CPB and cross-clamp times were 115 ± 32 min and 82 ± 20 min, respectively. HCA was necessary in one patient, and concomitant procedures in seven patients. The 30-day survival was 63% for immediate SAVR and 100% for early and late SAVR, even though one more patient died on postoperative day 31 after immediate SAVR. Besides, the longest follow up periods were 29 ± 15 months and 19 ± 14 months for immediate and early/late failure, respectively. In both groups, one patient died from cardiovascular-related causes, and one from non-valve-related causes.

CONCLUSIONS: SAVR after previous TAVI will become increasingly relevant. Due to the increasing use of TAVI in medium- or lower-risk patients, adequate strategies must be established since, in comparison to multimorbid patients, not taking action in these patients is not an option. Due to potentially high-risk patients and unique technical implications, SAVR after TAVI differs from conventional (redo) AVR. Under optimal conditions, acceptable survival rates can be achieved, but effective interdisciplinary approaches are essential.