Computed tomography attenuation correction improves the risk stratification accuracy of myocardial perfusion imaging

Abstract

Computed tomography attenuation correction (CT-AC) may improve the image quality and diagnostic accuracy of myocardial perfusion imaging. We evaluated the prognostic value of single-photon emission computed tomography (SPECT) myocardial perfusion imaging with CT-AC. Technetium-99m sestamibi stress/rest myocardial perfusion imaging was performed with SPECT and CT-AC in 935 patients. Images without and with AC were rated using the summed stress score (SSS) and classified as normal (SSS, 0-3), mildly abnormal (SSS, 4-8), or moderately or severely abnormal (SSS>8). All patients were followed up for a major adverse cardiac event (MACE). At the end of a mean follow-up period of 2.2 ± 0.8 years, there had occurred 42 MACEs [17 all-cause deaths (2%) and 25 nonfatal myocardial infarctions (3%)]. The annual frequency of MACEs in patients with normal SSS was 0.5%, that in patients with mildly abnormal SSS was 2%, and in patients with moderately or severely abnormal SSS was 8%. With AC, more studies were categorized as definitely normal, and the number of patients with moderately to severely abnormal perfusion on CT-AC was reduced (κ=0.32, P ≤ 0.001). The annual frequency of MACEs was similar between studies without AC and those with CT-AC for patients with normal or mildly abnormal SSS, whereas for the moderately to severely abnormal group the annual frequency increased significantly after CT-AC (4.5 vs. 8.1%, P=0.03). Cumulative survival without MACE was highest among patients who had normal CT-AC studies (SSS<4) and was the least among patients who had moderately to severely abnormal studies (SSS>8; P ≤ 0.01). CT-AC allows improved risk stratification for MACEs because there is a more clear separation between low risk and moderately to severely abnormal findings.