Two- and Three-Dimensional CT Ventriculography

Publisher Website
Google Scholar
Abstract
OBJECTIVE. We propose a new application of helical CT, CT ventriculography, which can produce two-dimensional (2D) and three-dimensional (3D) images of different cardiac phases (plus animation). We sought to determine the accuracy of CT ventriculography for assessing left ventricular volumes. MATERIALS AND METHODS. With a single breath-hold, the patient's entire heart was scanned with an ECG-gating technique (3-mm-thick collimation, 2 mm per rotation table speed, 0.8 sec per rotation, and 50 rotations through 10 cm in total). Using a 0.2-mm (0.08-sec) interval (10 slices per rotation) overlapping reconstruction, about 500 axial slices were obtained and reordered to separate different cardiac cycles. Then, 2D cardiac axes and 3D images were reconstructed and animated movies of the 2D and 3D images were produced. In 21 patients, the left ventricular end-diastolic volume, end-systolic volume, and ejection fraction were assessed and compared with left ventriculography. Correlations and agreements between CT and left ventriculography were determined. RESULTS. Close correlations between CT and left ventriculography were obtained (r = 0.95, 0.98, and 0.91, for end-diastolic volume, end-systolic volume, and left ventricular ejection fraction, respectively; p ≪ 0.0001 for all values). The limits of agreement between CT and left ventriculography were 44.3 to -44.5 ml for end-diastolic volume, 19.8 to -29.0 ml for end-systolic volume, and 19.7% to -9.5% for left ventricular ejection fraction. CONCLUSION. This cardiac application of helical CT provides a clear morphology along the cardiac axes and 3D images and an assessment of left ventricular volumes (end-diastolic volume, end-systolic volume, and left ventricular ejection fraction).