By Michael H. Crawford, MD, Editor

SYNOPSIS: A large registry study of transcatheter mitral valve replacement showed excellent results in failed bioprostheses but less overall success in failed mitral valve repairs with rings and conditions associated with mitral annular calcium.

SOURCES: Yoon SH, Whisenant BK, Bleiziffer S, et al. Outcomes of transcatheter mitral valve replacement for degenerated bioprostheses, failed annuloplasty rings, and mitral annular calcification. Eur Heart J 2019;40:441-451.

Maisano F, Taramasso M. Mitral valve-in-valve, valve-in-ring, and valve-in-MAC: The good, The bad, and the ugly. Eur Heart J 2019;40:452-455.

Surgery remains the gold standard for patients with mitral valve regurgitation (MR). Transcatheter valves have yet to be developed that are successful in degenerative or functional MR. However, degenerated mitral bioprostheses, failed mitral valve repairs, and mitral annular calcium (MAC) problems are associated with high surgical morbidity and mortality. The use of transcatheter valve replacement in these situations is increasing, but little has been reported about the procedural and clinical outcomes of this practice. An international transcatheter mitral valve replacement (TMVR) registry was created for patients undergoing TMVR for degenerated bioprostheses (valve in valve, or ViV), failed annuloplasty rings (ViR), or severe MAC (ViMAC). The transcatheter valves used were those available for aortic valve replacement. The primary endpoint was all-cause mortality at 30 days and one year. Several technical and clinical secondary endpoints also were evaluated. Among 521 patients from 40 centers studied between February 2008 and April 2018, 62% underwent TMVR for ViV, 27% for ViR, and 11% for ViMAC. The majority were high risk for valve re-operation, with a Society of Thoracic Surgeons score average of 9%. Baseline characteristics differed among the three groups. ViMAC patients were more likely women with stenosis of the mitral valve. ViR patients were more likely to have undergone prior coronary bypass with mitral regurgitation due to failed repair.

Overall technical success was 87%, but was higher for ViV patients than for ViR and ViMAC patients (94% vs. 81% vs. 62%; P < 0.001). Left ventricular outflow tract obstruction (LVOT) was more common in ViMAC patients compared to ViR and ViV patients (40% vs. 5% vs. 2%; P < 0.001). Thirty-day mortality was highest in the ViMAC group compared to the ViR and ViV groups (35% vs. 10% vs. 6%; P < 0.001). One-year all-cause mortality was highest in the ViMAC group, followed by the ViR and ViV groups (63% vs. 31% vs. 14%; P < 0.001). Patients with post-procedure moderate or worse MR were twice as likely to die vs. those with mild or no MR (42% vs. 21%; P = 0.01). The authors concluded that in patients with high surgical risk, TMVR was associated with excellent outcomes when used for degenerated bioprosthetic mitral valves. ViR and ViMAC patients experienced significantly higher rates of complications and mortality.


The principle message of this observational registry study is that TMVR probably is the treatment of choice for failed mitral valve bioprostheses. The implication of this conclusion is that there are few reasons to ever place a mechanical prosthesis in the mitral position, even in young patients. The 30-day mortality rate for ViV patients (6%) compares favorably to the 9-12% mortality rate reported previously for redo mitral valve surgery. The centers in this study used the transatrial-septal approach in 60% of patients, and there were no differences in outcomes compared to the patients who underwent TMVR via the apical approach. Thus, there is no reason not to use the less traumatic septal approach. Also, most patients received oral anticoagulation after TMVR (70%); the rest received antiplatelet therapy. Valve thrombosis occurred in 3% of ViV patients over two years; most were not on anticoagulants. Based on this limited nonrandomized experience, the authors recommended oral anticoagulation after mitral ViV.

The results for ViR patients were much less robust. Paravalvular leaks were common with post-procedure moderate or worse MR (18%). Often, these leaks were plugged via other catheter-delivered devices. When all complications were considered, procedural success was only 57%, with 12% of patients needing a second valve insertion. Also, the 30-day mortality rate for ViR patients is similar to reported surgical mortality for failed repairs (9%). Considering these results, the authors of an accompanying editorial suggested that mitral leaflet clipping may be a better option. Of course, a device designed for the aortic valve was used here. Future devices designed for the mitral valve may alleviate some of these issues.

When it came to TMVR for ViMAC patients, the results were even worse. The major complication was LVOT obstruction, which occurred in 40% of patients, necessitating subsequent alcohol septal ablation in 12% of patients. Overall procedural success was 41%. The 30-day mortality rate was a whopping 35%. These results are especially disappointing since MAC patients showed the highest surgical mortality rate of the three groups. A new study has been launched to see whether percutaneously lacerating the anterior leaflet can abrogate the obstruction issue. Also, new low-profile devices that are fully retrievable for the mitral position are in development.

The results from this large international registry are a snapshot of time in a fast-moving field with several new studies of alternative techniques and devices underway. At this time, only TMVR for degenerated bioprosthetic valves appears to be a clear alternative to surgery.