Accuracy and efficiency of left ventricular ejection fraction analysis, using multidetector row computed tomography:...

The aim of the present study was to assess the accuracy and efficiency of left ventricular ejection fraction (LVEF) analysis by multidetector row computed tomography (CT).

The CT data of 21 patients were analyzed by semi-automated software on a workstation. In analysis of LVEF using systolic volumes in the 6 phases (30-55%), systolic images of 10% interval (35, 45, 55%) were underestimated with a mean measurement error of -0.4% and the standard error of the estimate (SEE) of 0.6%, compared with the LVEF using images of 5% interval. In analysis of LVEF using 3-slice thicknesses of axial images (1, 2, and 3 mm), and 3-slice numbers of short-axis sections (10 14, and 30 slices), there was no significant difference between the SEE of the intraobserver reproducibility and that of the analysis with 30 short-axis sections using axial images of 1-, 2- and 3-mm thickness, and that with 14 short-axis sections using axial images of 1- and 2-mm thickness. The mean analysis times were 96.9 s, 119.8 s, and 227.0 s for the analysis with 10, 14, and 30 short-axis sections, respectively.

The proper selection of the reconstruction interval in the cardiac phase, the slice thickness of the axial images, and the number of short-axis sections reduces the analysis time, maintaining the accuracy of LVEF analysis. This will be acceptable for practical use at present, although more accurate analysis is preferable.