Hemodynamic classification of paravalvular leakage after transcatheter aortic valve implantation compared with angiographic or echocardiographic classification for prediction of 1-year mortality.

OBJECTIVES

We sought to assess angiographic, echocardiographic and hemodynamic grading of paravalvular leakage (PVL) after transcatheter aortic valve implantation (TAVI) with respect to prediction of 1-year mortality.

BACKGROUND

Meaningful criteria for the severity of PVL are needed to allow intraprocedural guidance and patient management after TAVI.

METHODS

We pooled the prospective TAVI databases of 2 German centers. During TAVI, PVL was assessed angiographically and by the aortic regurgitation index (ARI). ARI was calculated as ratio of the gradient between diastolic blood pressure and left ventricular end-diastolic pressure to systolic blood pressure times hundred. In addition, we performed transthoracic echocardiography before discharge.

RESULTS

A total of 723 patients undergoing TAVI with self-expandable (20.9%) or balloon-expandable (79.1%) valves were included. Grades of PVL as assessed during the procedure by angiography or ARI (below the previously defined cut-off of 25) did not show a significant association with 1-year mortality (P = 0.312 and 0.776, respectively). One-year mortality was 15.7% (39/249) in patienths with an ARI < 25 and 16.5% (71/430) in patients with an ARI ≥ 25. Echocardiographic classes of PVL at discharge showed a significant (P = 0.029) association with 1-year mortality, which was 11.5% (37/322) in patients with no/trace PVL, 18.0% (62/345) in patients with mild PVL and 23.1% (6/26) in patients with more than mild PVL. These findings prevailed after multivariable adjustment.

CONCLUSIONS

ARI did not help identify PVLs that are relevant to 1-year survival. Angiographic assessment during the procedure was less predictive than echocardiographic assessment before discharge.