On the inappropriateness of noninvasive multidetector computed tomography coronary angiography to trigger coronary revascularization: a comparison with invasive angiography.

OBJECTIVES: Our purpose was to evaluate the appropriateness of multidetector computed tomography angiography (MDCTA) as an anatomical standard for decision making in patients with known or suspected coronary artery disease.
BACKGROUND: Although correlative studies between MDCTA and coronary angiography (CA) show good agreement, MDCTA visualizes plaque burden and calcifications well before luminal dimensions are encroached.
METHODS: Pressure-derived fractional flow reserve (FFR) was obtained in 81 patients (116 vessels) who underwent both CA and MDCTA. Segments were visually graded for stenosis severity as: G0 = normal, G1 = nonobstructive (<50% diameter reduction), and G2 = obstructive (> or =50% diameter reduction).
RESULTS: Concordance between segmental severity scores by MDCTA and CA was good (k = 0.74; 95% confidence interval: 0.56 to 0.92). Diagnostic performance of MDCTA for detection of functionally significant stenosis based on FFR was low (sensitivity 79%; specificity 64%; positive likelihood ratio 2.2; negative likelihood ratio 0.3). Revascularization was considered appropriate in the presence of reduced FFR (< or =0.75). Decision making based on MDCTA guidance would result in revascularization in the absence of ischemia in 22% of patients (18 of 81) and inappropriate deferral in 7% (6 of 81), while revascularization in the absence of ischemia would be 16% (13 of 81) and inappropriate deferral 12% (10 of 81) with decisions guided by CA. Combined evaluation of stenosis severity using both anatomy (with either CA or MDCTA) and function (with FFR) yields the highest proportion of appropriate decisions: 90% and 91%, respectively (p = 0.0001 vs. CA only, p = 0.0001 vs. MDCTA only).
CONCLUSIONS: Similar to CA, anatomical assessment of coronary stenosis severity by MDCTA does not reliably predict its functional significance.