We compared the measurements of the aortic valve area (AVA) using dual-source computed tomography (DSCT) in patients with mid to severe aortic stenosis to measurements using transthoracic echocardiography (TTE) and invasive hemodynamic assessment. A total of 50 patients (mean age 73 +/- 10 years) with suspected aortic stenosis were included. The computed tomographic data were acquired using DSCT with standardized scan parameters (2 x 64 x 0.6 mm collimation, 330-ms rotation, 120-kV tube voltage, 560 mA/rot tube current). After injection of 35 ml contrast agent (flow rate 5 ml/s), a targeted volume data set, ranging from the top of the leaflets to the infundibulum, was acquired. Ten cross-sectional data sets (slice thickness 1 mm, no overlap, increment 0.6 mm) were reconstructed during systole in 5% increments of the R-R interval. The AVA determined in systole by planimetry was compared to the calculated AVA values using the continuity equation on TTE and the Gorlin formula on catheterization. DSCT allowed the planimetry of the AVA in all patients. The mean AVA using DSCT was 1.16 +/- 0.47 cm(2) compared to a mean AVA of 1.04 +/- 0.45 cm(2) using TTE and 1.06 +/- 0.45 cm(2) using catheterization, with a significant correlation between DSCT/TTE (r = 0.93, p <0.001) and DSCT/cardiac catheterization (r = 0.97, p <0.001). However, DSCT demonstrated a slight, but significant, overestimation of the AVA compared to TTE (+0.12 +/- 0.17 cm) and catheterization (+0.10 +/- 0.12 cm(2)). In conclusion, DSCT permits one to assess the AVA with a high-image quality and diagnostic accuracy compared to TTE and invasive determination.