Incremental diagnostic accuracy of computed tomography myocardial perfusion imaging over coronary angiography stratified by pre-test probability of coronary artery disease and severity of coronary artery calcification: The CORE320 study.

2015 
Abstract Background Myocardial CT perfusion (CTP) has been validated as an incremental diagnostic predictor over coronary computed tomography angiography (CTA) in assessing hemodynamically significant stenosis. Objectives To assess the diagnostic performance of CTA and CTP alone versus combined CTA–CTP stratified by Morise's pre-test probability and coronary artery calcium (CAC, Agatston) score. Methods 381 individuals (153 low/intermediate-risk for CAD, 83 high-risk, 145 known CAD) were further stratified based on CAC score cut-offs of 1–399 and ≥400. Area under the curve for receiver operating characteristics (AUC) was calculated to assess the diagnostic performance. Reference standards were QCA≥50% stenosis+corresponding SPECT summed stress score ≥1. Results In both pre-test risk groups with an Agatston score of 1–399, AUCs of CTA–CTP were not significantly different than that from CTA alone. In the low/intermediate-risk group with CAC score 1–399, AUC for CTA–CTP (89) was higher than that for CTP (76, p=0.003) alone. In the same group with CAC score ≥400, AUCs were higher for CTA–CTP (97) than that for CTA (88, p=0.030) and CTP (83, p=0.033). In high risk/known CAD patients with CAC 1–399, diagnostic performance for CTA–CTP (77) was superior to CTP (71, p=0.037) alone. In the high risk/known CAD group with CAC score ≥400, AUCs for combined imaging were higher (86) than that for CTA (75, p Conclusions The incremental diagnostic accuracy of CTP over CTA persists in patients across severity spectra of pre-test probability of CAD and coronary artery calcification. In patients with severe coronary calcification (CAC score≥400), combined CTA–CTP has better diagnostic accuracy than CTA and CTP alone.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    30
    References
    26
    Citations
    NaN
    KQI
    []