OBJECTIVE To investigate the clinical value of 64-slice spiral CT (64SCT) in assessing global left ventricular function in patients with old myocardial infarction (OMI), with magnetic resonance imaging (MRI) as the reference standard. METHODS A total of 28 patients (23 men and 5 women) with OMI underwent contrast-enhanced 64SCT with retrospective electrocardiographic-gating, MRI, and transthoracic echocardiography (Echo). The data sets of 64SCT and MRI were reconstructed at both end-diastole and end-systole to measure left ventricular end-diastolic volume (EDV), end-systolic volume (ESV), stroke volume (SV), ejective fraction (EF), and myocardial mass at end-diastole (MM). The data acquired with 64SCT and Echo were compared with the results obtained on MRI as the standard of reference respectively. RESULTS The parameters of global left ventricular function obtained with 64SCT were significantly correlated with the MRI data (r = 0.788-0.976, P < 0.001). EDV, ESV, SV, EF, and MM were slightly overestimated by 64SCT compared with MRI; however, there was no significant difference among the measurements. 64SCT was in good agreement with MRI. For the EDV, ESV, SV, and EF, the limits of agreement with Echo were significantly larger than with 64SCT compared with MRI (P < 0.05). The sensitivity, specificity, and accuracy of 64SCT to identify EF value less than 50% were 84.6%, 100% and 92.9%, respectively, whereas those of Echo were 61.5%, 66.7% and 64.3%, respectively. The accuracy of 64SCT was significantly higher than that of Echo (P < 0.01). There was a significantly larger overestimation of EDV, ESV, and SV with Echo than with 64SCT compared with MRI (P < 0.05), whereas EF was not significantly different. CONCLUSIONS A strong correlation between 64SCT and MRI is found for all parameters. 64SCT agrees well with MRI, and allows more reliable and accurate evaluation of global left ventricular function in patients with OMI than Echo compared with MRI.