Predictors of outcome in patients undergoing MitraClip implantation: An aid to improve patient selection

•Predictors associated with prognosis after MitraClip implantation were determined.•Heart failure, multiple valve disease and reduction of mitral regurgitation predict prognosis.•Presence of ≥ 2 predictors results in poorer survival after MitraClip implantation.•These predictors may be used to optimize patient selection for MitraClip implantation.

BackgroundMitraClip implantation (MCI) reduces mitral regurgitation (MR) and symptoms in patients considered inoperable or with high-surgical risk. Data to determine the benefit from MCI for an individual patient are limited. The aim of this study is to determine predictors associated with the prognosis after MCI to improve the patient selection for this procedure.MethodsWe included 84 consecutive patients (age: 76 ± 10 years, 51% male) who underwent MCI in our institution for symptomatic severe MR. All patients underwent transthoracic echocardiography before MCI; clinical and echocardiographic follow-up was obtained after MCI.ResultsThe 2-year survival was 81%. Predictors for two-year mortality in multi-variate analysis were baseline NT-proBNP ≥ 5000 μg/L (HR: 5.4, 95% CI: 1.8-16.2), previous valve surgery (HR: 4.5, 95% CI: 1.7-12.2), tricuspid regurgitation (TR) ≥ grade 3 prior to MCI (HR: 2.8, 95% CI: 1.2-6.8) and absence of MR reduction after MCI (HR: 2.1, 95% CI: 1.2-3.8). The 2-year survival of patients with 0, 1 or ≥ 2 of these predictors was: 87%; 78% and 38% respectively (log-rank p < 0.001). The functional class at 1 month and mid-term follow-up was worse in patients with two or more of these predictors present at baseline compared to patients with zero or one of these predictors (1 month: p = 0.007 and mid-term: p < 0.001).ConclusionHeart failure, previous valve surgery, co-presence of TR and the degree of MR reduction after MCI are the independent predictors of survival and functional status after MCI in high risk patients. The pre-procedural characteristics may be used to optimize patient selection, while maximal MR reduction should be attempted to optimize the outcome of MCI.