Comparison of computer and non-computer-assisted technologies in noninvasive cardiac imaging

We compared gated magnetic resonance imaging (MRI) and dynamic computed tomography (DCT) with two-dimensional cardiac ultrasound (ECHO) to assess differences in diagnostic information. Magnetic resonance imaging was performed in 41 patients; ECHO in 36, and DCT in 28 patients with various pathologic conditions. We measured the left and right ventricular (LV, RV) long and short axes (LA, SA), LV free wall and septal thickness (WT, ST) at end systole (ES) and end diastole (ED) on the apical four-chamber view (ECHO) or appropriate transaxial slice (MRI, DCT) on a subset of 14 patients. Paired-sample analysis of these three techniques, in this preliminary data, yields statistically different results as follows: LV SA: MRI versus ECHO at ED and ES (P less than 0.001 and 0.005); WT: MR versus ECHO at ES (P less than 0.002); CT versus ECHO at ED and ES (P less than 0.05 and 0.01); ST: MRI versus ECHO at ED and ES (P less than 0.001), and CT versus ECHO at ES (P less than 0.05). Thus, CT and MRI yield similar quantitative data, but both differ in varying degree from ECHO measurements. Differential accuracies and utilities of these techniques warrant further careful investigation.