Repeatability of treadmill exercise ejection fraction and wall motion using technetium 99m-labeled sestamibi first-pass radionuclide ventriculography

Background: Peak treadmill exercise radionuclide ventriculography (RVG) with technetium 99m has recently been validated for determination of left ventricular ejection fraction (LVEF). However, the repeatability of this technique for determination of both LVEF and regional wall motion has not been reported. Methods and Results: Each of 27 clinically stable patients underwent two treadmill exercise RVG studies within 40±51 days. The level of exercise achieved in the two tests was similar (double product: 26357±3877 vs 26621±4287), and there was no change in clinical or treatment status between the studies. Acquisition and processing were accomplished with a mobile multicrystal camera and a new version of a commercial software (Scinticor SIM 400 V. 4.1 BETA, Milwaukee, Wis.) that uses two left ventricular regions of interest. The two tests were compared to assess agreement (repeatability) on both an automatically calculated LVEF and wall motion in five left ventricular segments (basal anterior, distal anterior, apical, distal inferior, and basal inferior), with a 3-point semiquantitative visual score. Intraobserver and interobserver agreements (reproducibility) also were assessed on quantitative exercise LVEF derived from the same RVG test from a separate group of 20 patients with a broad range of exercise LVEF. The first and second treadmill exercise LVEFs were highly correlated (r=0.92, SEE=3.96, y=0.97x+0.58; and r=0.99, SEE=1.32, y=0.99x+0.25 respectively). Results of segmental visual score agreement between the first and the second treadmill first-pass studies were as follows: overall, 86% (116/135, kappa=0.74); basal anterior, 85% (23/27, kappa=0.72); distal anterior, 85% (23/27, kappa=0.84); apical, 93% (25/27, kappa=0.85); distal inferior, 93% (25/27, kappa=0.80); and basal inferior, 67% (18/27, kappa=0.64). Conclusion: Treadmill exercise first-pass RVG is a highly repeatable and reproducible test for quantitative LVEF and visual regional wall motion analysis. Our results imply the procedure may be useful for serial follow-up of patients with coronary artery disease and for the evaluation of the efficacy of medical or interventional treatment.