Technical factors affecting ⁸²Rb PET myocardial blood flow quantitation

Objectives Factors such as age & gender are known to contribute to normal variability of myocardial blood flow (MBF). Among technical factors that can diminish reliability of left ventricular (LV) MBF measurements derived from dynamic PET data is the highly variable MBF of basal segments (Gould et al, Circulation 1994;89:1530-8). Our investigation was conducted to determine the magnitude of the effect of excluding basal-septal segments in calculating global MBF.

Methods Data were examined retrospectively for 196 patients who underwent rest/ regadenoson-stress ⁸²Rb PET/CT scans for CAD. MBF was computed individually for each of the 17 standard AHA/ACC LV segments using factor analysis of time-activity curves applied to a 2-compartment model. Means of MBF values for all 17 segments (M17) were compared to means of MBF excluding the 3 basal-septal segments (M14). MBF inhomogeneity was computed as the ratio of standard deviation to the mean of segmental MBF values.

Results M17 correlated strongly with M14 for both rest & stress MBF (r = 0.99, p < 0.0001), but Bland-Altman plots of MBF differences versus means exhibited significant correlation (r = -0.64, p < 0.0001), with slopes of -7.2±0.6% & -8.3±0.7% (p < 0.0001) for rest & stress, so that M17 MBF was consistently 7-8% lower than M14 MBF. Perfusion inhomogeneity was greater if basal-septal segments were included for rest (38±9% versus 31±10%, p < 0.0001) & stress (45±11% versus 35±11%, p < 0.0001).

Conclusions Because including basal septal segments when calculating global MBF results in lower MBF values, and significantly greater computed MBF inhomogeneity, these segments should not be included for global MBF determinations.

• Download figure
• Open in new tab
• Download powerpoint