CT myocardial perfusion and coronary CT angiography: Influence of coronary calcium on a stress-rest protocol.

ABSTRACT

BACKGROUND: High amounts of coronary artery calcium limit image quality and diagnostic accuracy of multidetector computed tomography (MDCT) angiography (CTA) regarding the assessment of obstructive coronary artery disease (CAD). CT myocardial perfusion imaging may represent an opportunity to overcome this limitation. OBJECTIVE: To explore the additive value of CT myocardial perfusion to CTA depending on the patient's calcium score and in comparison to the reference standard of invasive coronary angiography plus fractional flow reserve (FFR) measurement. METHODS: Symptomatic patients with intermediate pretest probability of CAD were prospectively recruited and underwent both cardiac MDCT (64-slice scanner, retrospectively-gated stress-rest protocol) and invasive coronary angiography including FFR assessment. We defined hemodynamically significant CAD by the presence of occlusive or subocclusive (99%) stenosis, >50% stenosis in left main or FFR≤0.80. Stress CT myocardial perfusion imaging was performed in all patients in addition to CTA. The additive value of CT myocardial perfusion to rule in or rule out the presence of hemodynamically relevant stenosis on a per-patient basis was assessed and analyzed relative to the patient's calcium score. RESULTS: 95 patients were included in the analysis (62 ± 8.2 years, 68%males). Hemodynamically significant CAD was present in 42 patients. Sixty-four patients had a fully evaluable CTA examination. Per-patient, CTA alone had a sensitivity, specificity and AUC of 100%, 59% and 0.79 respectively (77% patients correctly classified). Adding CT myocardial perfusion to evaluate uninterpretable vessel territories in CTA in 66 patients with a calcium score>100 yielded a sensitivity of 88%, a specificity of 74% and an AUC of 0.81 (81% patients correctly classified), and in 52 patients with a calcium score>400 sensitivity was 91%, specificity 70%, and AUC 0.80 (82% patients correctly classified, p = 0.733 versus using perfusion imaging in all patients). CONCLUSIONS: From a pragmatic standpoint, limiting the use of CT perfusion to individuals with a calcium score above 400 might be a feasible strategy to optimize the diagnostic accuracy of CT imaging for diagnosis of obstructive CAD.