Assessment of Coronary Artery Disease by Computed Tomography

Abstract

BACKGROUND Computed Tomography Coronary Angiography (CTCA)is a technique for imaging coronary arteries with increasing indications in clinical cardiology. AIMS 1.Develop a heart rate (HR) lowering regime for CTCA and to measure its association with image quality. 2.Examine the diagnostic accuracy of 64 slice CTCA (CTCA64) in patients with known coronary artery disease (CAD). 3.Examine the diagnostic accuracy of CTCA64 for assessment of stent restenosis 4.Demonstrate utility of CTCA as an endpoint in assessment of novel diagnostic biomarkers of CAD. METHODS I developed a HR reducing strategy using metoprolol and assessed its effectiveness for improving CTCA64 image quality. The diagnostic value of CTCA in patients with suspected angina was evaluated by comparison with invasive coronary angiography. The diagnostic value of CTCA for quantifying stent restenosis was evaluated by comparison with intravascular ultrasound. The utility of CTCA for evaluating the diagnostic value of B-type natriuretic peptide (BNP) and high sensitivity cardiac troponin I (hs- TnI) was evaluated by blood sampling in patients with suspected angina who subsequently underwent CTCA. RESULTS 1.In 121 patients undergoing CTCA, 75 required rate control. This was achieved (rate ≤60 bpm) in 83% using a systematic regimen of oral and IV metoprolol (n=71) or verapamil (n=4). I demonstrated a significant relation between HR reduction and graded image quality (p<0.001). 2.80 patients underwent CTCA64 and invasive coronary angiography. 724 coronary arterial segments were available for analysis. The sensitivity and specificity of CTCA for significant luminal stenosis was 83.3% (95% CI 67.1-92.5%) and 96.7% (95% CI 95.1-97.9%), respectively, but the positive predictive value was only 63.5% (95% CI 50.4-75.3%). 3.80 patients with 125 stented segments underwent CTCA64 and invasive coronary angiography. Additional intravascular ult rasound (IVUS) examination of stented segments was performed in 48 patients. Using IVUS as the gold-standard for stent restenosis, CTCA and invasive coronary angiography had comparable diagnostic specificities for binary stent restenosis: 82.7% (95% confidence intervals 69.7- 91.84%)and 78.9% (95% confidence intervals 65.3-88.9%), respectively. Sensitivities were lower, particularly the sensitivity of CTCA which was only 11.8% (95% confidence intervals 1.5-36.4%) compared with 58.8% (95% confidence intervals 32.9-81.6%) for invasive coronary angiography. 4. In 93 patients with suspected angina CTCA64 provided a useful endpoint for assessing the diagnostic value of novel circulating biomarkers. BNP levels were higher in the 13 patients shown to have significant (≥50% stenosis) coronary artery disease compared with patients who had unobstructed coronary arteries (18.08pg/ml (IQR 22) vs 9.14pg/ml (IQR 12.62), p=0.024) and increased significantly with exercise, particularly in the group with anatomic coronary artery disease (2.73 ± 5.69 pg/ml vs 1.27±3.29 pg/ml, p=0.16). Conversely I found no association between hs-TnI and the presence of CAD. CONCLUSION Image quality of CTCA64 is enhanced by heart rate reduction below 60 bpm which can be achieved safely by a regimen of oral and intravenous metoprolol. Although CTCA64 is a useful non-invasive method for diagnosis of coronary artery disease, it has a low positive predictive value for identifying severe (≥50%) luminal stenosis which limits its clinical value. Its value for assessment of stent restenosis is even more limited but it finds useful application as an endpoint for diagnostic evaluation of novel biomarkers, allowing confirmation of an association between circulating BNP levels and stable coronary artery disease