Results:

Between January 2014 and May 2018, 312 patients with severe or greater symptomatic TR underwent TTVI and were included in the registry across 18 centers (in Switzerland, Germany, France, Italy, United States, and Canada). Mean age of the treated population was 76 ± 8.5 years, with a 56% prevalence of women. Mean EuroSCORE II was 9 ± 8%. Etiology of TR was functional in 92% of the patients (288 patients). Most of the patients had severe symptomatic TR at admission: 95% were in New York Heart Association (NYHA) functional class III/IV; and 71% of the patients had a history of previous admission for right ventricular (RV) failure. The main location of the TR jet was central in 75% of the patients (n = 234). Mean tricuspid annular dimension was 46.9 ± 9 mm, with a coaptation depth of 9.4 ± 4.1 mm. Mean systolic pulmonary artery pressure (sPAP) was 41 ± 14.8 mm Hg.

Implanted devices included: MitraClip in 210 cases, Trialign in 18 cases, TriCinch first generation in 14 cases, caval valve implantation in 30 cases, FORMA in 24 cases, Cardioband in 13 cases, NaviGate in six cases, and Pascal in one case.

One patient underwent combined MitraClip and Trialign during the same procedure. Isolated TTVI was performed in 64% of the cases (n = 202), whereas in the remaining, TTVI was performed concomitantly during transcatheter mitral repair (108 cases), transcatheter aortic valve replacement (one case), or paravalvular leak closure (one case). Overall mean procedural time was 133 ± 66 minutes, whereas it was 129 ± 72 minutes in isolated TTVI.

Procedural success (defined as patient alive at the end of the procedure with the device successfully implanted, delivery system retrieved, and a residual TR grade ≤2+) was achieved in 72.8% of cases, with no differences among the different devices (p = 0.20). A baseline coaptation depth >1 cm was identified as the best cutoff to predict the risk for procedural failure.

Thirty-day mortality was 3.6% (10 patients: two sepsis, two respiratory insufficiency, six progressive RV failure), and it was significantly lower among patients with procedural success (1.9% vs. 6.9%; p = 0.04). Overall incidence of major adverse events at 30 days was 10.3% (29 events), including: 10 deaths (3.6%), five major bleedings (1.7%), three strokes (1%), two acute myocardial infarctions requiring right coronary artery stenting (0.7%), four conversions to surgery (1.4%: in two cases emergent conversion to surgery was required, in two cases tricuspid surgery was performed “elective” due to procedural failure), two respiratory failure (0.7%), one device detachment (0.3%), one ventricular arrhythmia (0.3%), and one aortic prosthetic valve thrombosis (0.3%).

Clinical improvement was observed at 6 months of follow-up, with 54% of the patients in NYHA functional class I-II (p = 0.04). Actuarial survival at follow-up was significantly better in patients in whom acute procedural success was achieved (70.3 ± 8% vs. 90.8 ± 4% at 1 year; p = 0.0002). Superior survival according to procedural success was observed also when only isolated TTVIs were considered (69 ± 9% vs. 89 ± 5% at 1 year; p = 0.004). Procedural success and higher values of sPAP at baseline were independently associated with increased mortality at follow-up.