Estimation of global and regional cardiac function using 64-slice computed tomography: a comparison study with echocardiography, gated-SPECT and cardiovascular magnetic resonance.

Background: Sixty-four-slice multidetector spiral computed tomography (CT) has improved temporal resolution and reduced acquisition time. We aimed to evaluate the functional analysis using 64-slice CT comparing with echocardiography, electrocardiographically gated single-photon emission tomography (SPECT) and cardiovascular magnetic resonance (CMR).
Methods: Six-three patients (77.4+/-18.6 bpm) underwent 64-slice CT and CMR (echocardiography in 55; SPECT in 33) within 2 weeks were retrospectively reviewed. The left ventricular volumetric data from different methods were compared with CMR. Regional wall motion was compared between CT and CMR in a 17-segment and 4-point system (1=normal to 4=akinesis/dyskinesis).
Results: Ejection fraction (EF), end-diastolic volume (EDV) and end-systolic volume (ESV) by CT agreed well with CMR (bias+/-SD, -0.22%+/-4.18, r=0.97;-0.59 mL+/-15.21, r=0.98; 1.09 mL+/-10.61, r=0.99) over a wide range of left ventricular (LV) function (EF 18-76% by CMR). Our results also showed good correlation of EF measured by CT and echocardiography (r=0.87) or SPECT (r=0.91, all P<0.0001); however, standard deviation of EF difference between CT and CMR was significantly less than echocardiography or SPECT (P<0.005). For regional wall motion, an exact agreement of 97% (kappa=0.91) was found between CT and CMR.
Conclusion: Sixty-four-slice CT agreed well with CMR in LV function assessment, and had a superior accuracy than echocardiography and SPECT on EF estimation. Sixty-four-slice CT is considered a clinically acceptable and robust method to evaluate LV function.