Reproducibility of measurements of left ventricular function with gated myocardial perfusion SPECT and comparison with blood pool radionuclide ventriculography

Introduction and objective This study was designed to evaluate the reproducibility of ejection fraction (EF) and ventricular volume measurements obtained with single photon emission computed tomography (gated-SPECT), and to assess the correlation between EF values obtained with this method and blood pool planar radionuclide ventriculography. Patients and method A total of 55 patients were included (37 men, mean age 61.3 years) upon referral to two nuclear cardiology units for diagnosis (50%) or follow-up of known coronary artery disease. In a standard 2-day protocol, patients received a dose of [ 99m Tc]tetrofosmin (800 MBq) at stress and at rest. Two resting gated-SPECT studies were performed. QGS software was used to obtain left ventricular EF, end-diastolic volume (EDV) and end-systolic volume (ESV). Forty-nine patients agreed to undergo blood pool ventriculography on the third day. Results Interobserver variability was 0.5 (2.6)% ( r = 0.99) for EF, 1.9 (10.7) mL for EDV ( r = 0.98) and 0.5 (5.4) mL for ESV ( r =,0.99). Interassay variability was 2 (5.1)% ( r = 0.94) for EF, 4.5 (8.6) mL for EDV ( r = 0.99) and 3.4 (6.6) mL for ESV ( r = 0.99). The correlation between gated-SPECT EF and blood pool EF was suboptimal ( r = 0.75, 95% Cl, 0.59-0.85). Conclusions There was excellent interobserver and interassay reproducibility for left ventricular functional parameters measured with gated-SPECT and QGS software, and this method can be used for serial evaluations of ventricular function. Although the correlation between values obtained with gated-SPECT and blood pool ventriculography was acceptable, the differences show that the two techniques cannot be considered equivalent.