Background: There are limited data on the effect of tracer dose on the reproducibility and accuracy of left ventricular (LV) mechanical dyssynchrony indices by phase analysis of gated single photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI).
Methods and results: We measured LV dyssynchrony in 54 patients with normal LV ejection and perfusion (group 1) and 54 age and gender matched patients with LV ejection <35% (group 2) using phase analysis of gated SPECT MPI from stress (high dose) and rest (low dose) studies with 2 software programs: Corridor4DM (4DM) and Emory Cardiac Toolbox (ECTb). Although the correlation between rest- and stress-derived standard deviation was good (R = 0.76, P < .0001, with both software programs), there was considerable variability between the measurements (P < .0001 by paired t test). In addition, the rest standard deviation was significantly higher than stress in group 1 (10.2° ± 4.6° vs 6.1° ± 2.5°, and 12.2° ± 6.4° vs 7.9° ± 4.6°, with 4DM and ECTb, respectively, P < .0001 for both) and group 2 patients (44.0° ± 18.0° vs 35.9° ± 21.0° and 47.3° ± 19.2° vs 38.8° ± 19.8°, with 4DM and ECTb, P = .03 and .02, respectively). Similarly, the rest standard deviations were higher than the stress values irrespective of the type of stress test (i.e., exercise vs pharmacological), and the body mass index. Finally, using rest-derived dyssynchrony indices was associated with 9%-13% and 22%-26% false positive rate of significant mechanical dyssynchrony using different cut-off values for groups 1 and 2, respectively.
Conclusion: LV mechanical dyssynchrony indices by phase analysis have more variation and are significantly higher if derived from rest gated SPECT images obtained with low-dose tracer.