Coronary flow reserve by CT perfusion

Abstract

Cardiovascular computed tomography (CT) is one of the most impressive advances in the non-invasive diagnosis of cardiovascular disease in the last decade. Going beyond coronary calcium scoring, cardiovascular CT is capable of identifying calcified and non-calcified plaque, percent stenosis, cardiac structure and morphology, and left ventricular systolic function. The abilities to detect subclinical and obstructive atherosclerosis and exclude disease with high diagnostic certainty are its greatest attributes. Single-center and multi-center studies in various cohorts have established that CT coronary angiography (CTA) is an accurate non-invasive test for determining percent stenosis which can be performed with low radiation dose. However, it is important to point out that coronary CTA is an anatomical test that does not provide important physiological data. In fact, the value of CTA for predicting the presence of myocardial ischemia is limited. While coronary CTA has an excellent negative predictive value, it is more limited in determining the significance of stenoses in patients with disease. Thus, caution needs to be exercised when adopting cardiac CT clinically. The importance of coronary physiology for diagnosis, prognosis, and guidance of therapies must not be ignored. Invasive measurements such as fractional flow reserve have been shown to impact outcomes in patients with coronary stenoses and guide the appropriateness of invasive therapies such as intracoronary stenting. Radionuclide myocardial perfusion imaging is an established diagnostic tool for detecting obstructive coronary artery disease and is proven to accurately predict prognosis and guide therapy. Furthermore, it has been shown to add incremental prognostic value above and beyond what is provided by invasive angiography. While the major focus of cardiac CT research has been its application in coronary angiography, our group and others have recognized that X-ray computed tomography using iodinated contrast as a tracer is capable of providing measurements of myocardial blood flow and blood volume. In this issue of the Journal of Nuclear Cardiology, Christian and colleagues use the upslope integral method, a previously described method published in the magnetic resonance perfusion imaging literature, to demonstrate cardiac CT’s potential to quantify coronary flow reserve (CFR) independent of the arterial input function. In summary, they performed dynamic CT perfusion imaging in a canine model that received intracoronary adenosine in one vessel supplying a hyperemic territory, while the remote territory was kept at baseline conditions. In addition, several experiments employed CT perfusion imaging during temporary occlusion of a vessel. Images were analyzed by measuring the attenuation changes over time in the remote and hyperemic territories and plotting time-attenuation curves. They then calculated the area under each myocardial curve during the upslope to peak of contrast enhancement and calculated CFR by taking the ratio of area under the curve for the hyperemic territory and the area under the curve for the remote territory. The reported results are quite impressive. Compared to the gold standard, CT-derived CFR showed an excellent correlation with microsphere-derived CFR (r = .9325). There was also an excellent agreement between CFR by CT vs microspheres, 4.4 ± 1.4 vs 4.1 ± 1.1, respectively, with 95% confidence limits of difference between the measures of 1.08. This study demonstrates that CFR can be accurately derived from CT. The authors suggest that this method could be implemented by setting the highest slope integral value as a normalization standard and calculating CFR relative to this normal value. They postulate that this could be performed with as few as 5 heart beats and independent of the arterial input function, while also acquiring the CT angiogram using a single contrast bolus and scanning session. While these preclinical results look promising, the implementation of this method faces several technical challenges. From the Department of Medicine, Division of Cardiology, Russell Morgan Department of Radiology, Division of Nuclear Medicine, Department of Medicine, Division of Cardiology, and The Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD. Reprint requests: Richard T. George, MD, Department of Medicine, Division of Cardiology, Johns Hopkins University, 600 N. Wolfe Street, Carnegie Building 565C, Baltimore, MD 21287; rgeorge3@ jhmi.edu. J Nucl Cardiol 2010;17:540–3. 1071-3581/$34.00 Copyright 2010 by the American Society of Nuclear Cardiology. doi:10.1007/s12350-010-9250-2