Rubidium-82 PET versus Thallium-201 SPECT for detection of coronary artery disease in the Alternative Radiopharmaceutical for Myocardial Imaging (ARMI) trial

241 Objectives Rubidium (Rb-82) positron emission tomography (PET) may be a more accurate alternative than thallium (Tl-201) single-photon emission tomography (SPECT) for myocardial perfusion imaging (MPI) when Tc-99m supply is limited. This study evaluated the accuracy of Tl-201 SPECT vs. Rb-82 PET for diagnosis of obstructive coronary artery disease (CAD). Methods Eighty normal subjects (N=40 male) were identified with low-likelihood (LLK) of disease who underwent Rb-82 PET or Tl-201 SPECT dipyridamole stress perfusion imaging during the Tc-99m isotope crisis. Normal mean and SD database polar-maps of relative uptake (0-100%) were created separately for both tracers. Fifty PET and 74 SPECT patients with normal LVEF≥40% and no previous revascularization were used to assess sensitivity and specificity for detection of 50% and 70% stenosis, as confirmed using invasive coronary angiography (ICA). Standard 17-segment sum-stress-score (SSS) 4 was considered scan-positive for CAD. Normalcy was validated in an independent LLK group of 35 PET and 52 SPECT subjects, to account for ICA post-test referral bias. Results Normalcy in the LLK subjects was similar between tracers at 94% for Rb-82 and 93% for Tl-201, confirming adequate normal database specificity. Rb-82 PET sensitivity and specificity were 83% and 94% for 50% stenosis, and 90% and 93% for 70% stenosis, using the pooled CAD and LLK subjects. Corresponding Tl-201 SPECT values were lower with sensitivity and specificity of 56% and 85% for 50% stenosis, and 57% and 83% for 70% stenosis. Areas under the receiver operating characteristic curves were higher for Rb-82 (AUC ≥ 0.88) vs. Tl-201 (AUC ≤ 0.76). Conclusions Rb-82 PET demonstrated superior accuracy to Tl-201 SPECT for the diagnosis of obstructive CAD. Patient follow-up is underway to determine the potential impact on downstream resource utilization, cost-effectiveness and clinical prognosis for adverse cardiac events. Research Support CIHR Grant MIS100935