Two-phase reconstruction for the assessment of left ventricular volume and function using retrospective ECG-gated MDCT: Comparison with echocardiography

OBJECTIVE. The aims of our study were to investigate the clinical feasibility of a two-phase reconstruction method based on ECG to evaluate left ventricular (LV) volume and function using cardiac MDCT and to compare these results with those from echocardiography. SUBJECTS AND METHODS. The LV end-diastolic and end-systolic volumes, stroke volume, and ejection fraction were measured using two different methods of cardiac MDCT in 19 patients who had undergone cardiac MDCT and echocardiography. The first was a two-phase reconstruction method based on retrospective ECG-triggering: The end-systolic phase was reconstructed when the reconstruction window was located halfway in the ascending T wave on ECG, and the end-diastolic phase was reconstructed when the reconstruction window was located at the starting point of the QRS complex on ECG. The second was a multiphase reconstruction method: 20 series of images were reconstructed at every 5% throughout the cardiac cycle. The LV volumes and function determined by the two reconstruction methods were compared. The results measured by cardiac MDCT were compared with those obtained by echocardiography. RESULTS. The LV end-diastolic and end-systolic volumes, stroke volume, and ejection fraction measured by the two-phase reconstruction method correlated well with those measured by the multiphase reconstruction method (r = 0.984, 0.978, 0.969, 0.969, respectively). There were no significant differences between the results of the two different reconstruction methods (p > 0.05). The LV volumes showed moderate to good correlation between cardiac MDCT and echocardiography (0.766 < r < 0.940). Ejection fraction measured by cardiac MDCT yielded a significant overestimation of 2.9% +/- 8.7% (mean +/- SD) compared with that measured by echocardiography. CONCLUSION. A two-phase reconstruction method on cardiac MDCT is relatively simple and can provide an objective standard for reconstructing the appropriate image sets for end-diastole and end-systole without the need to review serial preview images.