Purpose: We compared the quality, interpretive confidence and interreader agreement between SPECT and PET myocardial perfusion imaging (MPI) in the same group of patients.
Methods: The study group comprised 27 patients (age 55 ± 8.5 years, 12 men) with known or suspected coronary artery disease (CAD) who had undergone gated rest/stress MPI with (99m)Tc-labelled agent SPECT (with and without attenuation correction, AC), and subsequent clinical confirmation with (82)Rb PET. Three experienced readers blinded to the clinical information interpreted all MPI studies.
Results: Interreader agreement was significantly superior for PET studies than for SPECT studies. Following consensus interpretation, the quality of 22 % of the non-AC SPECT studies, 33 % of the AC SPECT studies and 63 % of the PET studies was assessed as excellent or good (p = 0.016). Interpretations were definitely normal or abnormal in 7 % of non-AC SPECT studies, 30 % of AC SPECT studies and 85 % of PET studies (p = 0.046). In 13 patients who had received either invasive coronary angiography or CT angiography with no significant CAD, the true-positive rate for significant CAD was higher for PET, and the true-negative rate was equal for PET and AC SPECT, and lower for non-AC SPECT.
Conclusion: (82)Rb PET MPI, used as a confirmatory test after SPECT, offers improved image quality, interpretive confidence and interreader agreement.