The feasibility of F-18-FDG gated positron emission tomography (PET) for left ventricular dyssynchrony assessment in comparison with Tc-99m-MIBI gated single-photon emission computed tomography (SPECT) among patients with prior myocardial infarction

Background: Phase analysis by( 99m)Tc-MIBI gated single-photon emission computed tomography (GSPECT) has been considered to be an adequate method in the validation of left ventricular (LV) dyssynchrony. Compared with GSPECT, prior myocardial infarction patients with myocardial perfusion defects but myocardial viability usually show preserved uptake of F-18-FDG, and extensive myocardium is detected by F-18-FDG gated positron emission tomography (GPET). Thus, theoretically, it should be more accurate. The aim of this study was to investigate the feasibility of GPET for LV dyssynchrony assessment in comparison with GSPECT among infarction patients. Methods: A total of 146 patients with infarction underwent 2 consecutive days of GSPECT and GPET examinations. Quantitative gated SPECT-derived LV phase analysis was applied to GPET and GSPECT data to assess the presence of LV dyssynchrony via histogram bandwidth (BW) and phase standard deviation (SD). The correlation and agreement of BW and SD between GSPECT and GPET were examined. Factors (i.e., total perfusion defect, scar and mismatch) related to the discrepancies of LV dyssynchrony (i.e., BW and SD) in GPET and GSPECT were assessed by univariate and multivariate regression analysis. Results: A moderate correlation between GPET and GSPECT was found in the measurements of BW (r=0.554) and SD (r=0.537). Bland-Altman analysis revealed that GPET overestimated both BW and SD (20.5 degrees and 9.5 degrees, respectively). In addition, the BW and SD measured by GPET were still overestimated after subgroup analysis. Between GPET and GSPECT, multivariate regression analysis revealed that total perfusion defects were related to the difference in BW measurement (P<0.001), and mismatch was associated with the difference in SD measurement (P<0.01). Conclusions: In patients with infarction, GPET moderately correlated with GSPECT in assessing LV dyssynchrony. GPET overestimated both BW and SD, so these analyses should not be interchangeable in individual patients.