Left ventricular ejection fractions (LVEF) are routinely measured and result in critical decision-making algorithms in cardiology. This study was conducted to compare the accepted standard two-dimensional (2-D) echocardiogram ejection fraction (EF(ECHO)) with single photon emission computed tomography rest ejection fraction (EF(SPECT)). Data were acquired and analyzed from 51 inpatients. EF(ECHO) was obtained using modified Simpson's rule in the four-chamber apical view, and gated EF(SPECT) was computed by an automated method (Siemens ICON software). Comparison between EF(ECHO) and EF(SPECT) was done by linear regression, Bland-Altman, and receiver operator characteristic (ROC) analyses. Linear regression analysis revealed EF(SPECT) = 1.12 x EF(ECHO) - 3.6%, r = 0.93, n = 51, P < 0.0001. Bland-Altman analysis showed that the limits of 95% confidence for the difference between EF(SPECT) and EF(ECHO) were - 12% to 18%. Ninety-eight percent of the datapoints were within the limits of confidence. The ROC analysis showed that the sensitivity and specificity for detecting abnormal EF (< or = 50%) were 85% and 86% with EF(SPECT), and 91% and 90% with EF(ECHO). This study showed good correlation as well as agreement between SPECT and two-dimensional echocardiograms in measuring EF.