Objective. To assess risk factors for paravalvular leak (PVL) after transcatheter aortic valve implantation (TAVI) in a large single-center cohort, including measurement of aortic valve calcification using a reproducible method. Methods. We retrospectively analyzed preoperative contrast-enhanced multidetector computed tomography (MDCT) scans of patients who underwent TAVI in our center between 2009 and 2016. Calcium volume was calculated for each aortic cusp in the aortic valve (AV), left ventricular outflow tract (LVOT) and device-landing zone (DLZ). Results. Overall, 539 patients were included in the study (Edwards SapienXT, n=192; Edwards Sapien3, n=206; Medtronic CoreValve EvolutR, n=44; Symetis Acurate, n=97). Median calcium volume in the DLZ was 757 mm 3 , with no significant differences among the four prosthesis groups. None of the patients had severe PVL. The overall incidence of mild-to-moderate PVL was 15.8% (95% CI: 12.8-19.1%). On multivariate logistic regression, DLZ calcification (p=0.00006; OR for an increase of 100 mm 3 1.08; 95% CI: 1.04-1.13) and use of the CoreValve (p=0.0028; OR 4.1; 95% CI: 1.6-10 with SapienXT as reference) prosthesis were found to be associated with ≥mild PVL. In contrast, degree of oversizing (p=0.002; OR 0.97; 95% CI: 0.95-0.99), and use of Sapien3 (p=0.00005; OR 0.23; 95% CI: 0.11-0.47 with SapienXT as reference) were associated with a lower incidence of ≥mild PVL. Conclusions. Aortic calcification volume in the DLZ is associated with residual PVL after TAVI. When taking calcification into account, the balloon-expandable prosthesis Sapien3 seems to be associated with a lower incidence of PVL.