Number 24, 2004
Angina Pectoris

Imaging the coronary vasculature using multislice computed tomography

Back to the Summary

Nico R. Mollet^1,2, Filippo Cademartiri², Pim J. de Feyter
¹Department of Cardiology, Thoraxcenter, Erasmus Medical Center, Rotterdam, The Netherlands
²Department of Radiology, Erasmus Medical Center, Rotterdam, The Netherlands
Correspondence: Pim J. de Feyter, MD, PhD, Erasmus Medical Center, Department of Cardiology and Radiology, Thoraxcenter Bd 410, PO Box 2040, 3000 CA Rotterdam, The Netherlands.
Tel: +31 104635242, fax: +31 104634320, e-mail: p.j.defeyter@erasmusmc.nl

Introduction
Multislice computed tomography (MSCT) is developing rapidly. First generation four-row MSCT scanners were introduced in the late 1990s and have brought promising results, but image quality of the small and rapidly moving coronary arteries was inadequate in a significant number of imaged vessels [1–4]. Current 16-row MSCT scanners have increased temporal and spatial resolution, allowing more reliable evaluation of both coronary lumen and plaques, and left ventricular function can also be easily evaluated from the same scan. Here we summarize the technique and diagnostic performance of this imaging method.

Technique of MSCT coronary angiography
Current state-of-the-art MSCT scanners are equipped with 16 detector rows. The entire heart can be scanned within a single breath-hold of about 20 seconds, which is a manageable for practically all patients. Little iodinated contrast material (100mL, injected at 4mL/s) is needed to perform the scan. The ECG is recorded during the scan and used, retrospectively, to reconstruct images during the mid-to-end diastolic cardiac phase when least motion of the heart is present. This retrospective gating technique ‘freezes’ the contraction of the heart and allows flexibility in the position of reconstruction windows to diminish artefacts related to rapid coronary motion and premature beats. Current MSCT scanners have a temporal resolution that allows reconstruction of nearly motion-free images in patients with a regular, slow heart rate (below 70 beats/min); in the absence of contraindications, patients presenting with faster heart rates receive oral or intravenous β-blockers to reduce the heart rate, thereby improving image quality.
The high radiation exposure during MSCT coronary angiography, reported as between 6.7 and 13.0 mSv [5–8], is a matter of concern. Development of radiation sparing techniques such as prospective x-ray tube modulation (reducing the radiation exposure by nearly 50% in patients with low heart rates [5,8]) is highly desirable.

Diagnostic performance of MSCT coronary angiography
Results with first-generation four-row MSCT scanners in detecting significant coronary stenoses (Table I) are promising in comparison with conventional angiography [1–4]. However, these were obtained after exclusion of approximately 30% of imaged vessels from analysis, because of insufficient image quality. Current-generation 16-row MSCT scanners have increased temporal and spatial resolution that give greater diagnostic performance in evaluating coronary stenoses (Figures 1, 2). In all currently available studies, oral β-blockers were administered to patients with heart rates above 60 or 65 beats/min, and sensitivity and specificity around 90% were reported [9–11]. The number of vessels with low image quality ranged between 7% and 12%, but two studies yielded a high diagnostic accuracy despite including these vessels (Table I) [9,11]. These results indicate that MSCT coronary angiography can be a viable alternative for conventional angiography in selected patients in sinus rhythm, with low heart rates, able to breath-hold for 20 seconds. Patients with irregular or faster (≥70 beats/min) heart rates (which cause motion artefacts on the MSCT scan) and those with extensive coronary calcifications (which cause artefactual blooming masking the coronary lumen) are less suitable candidates. Some investigators perform a low-dose, nonenhanced calcium scoring scan before the contrast-enhanced computed tomography (CT) angiography scan, to exclude patients with an unfavorably high coronary calcium load [12].

Table 1. Studies comparing 4-row and 16-row multislice computed tomography (MSCT) coronary angiography scanners.

Functional parameters derived from MSCT coronary angiography
Important functional parameters regarding left ventricular performance can be extracted from the angiography scan. Studies comparing results of MSCT coronary angiography with those of biplane conventional angiography, magnetic resonance imaging, or two-dimensional echocardiography found good correlation between the ejection fraction and volumes from both techniques [13–15]. However, MSCT is a less reliable technique for evaluating regional wall motion abnormalities, because of limited temporal resolution [15]. Furthermore, β-blockers, often used before the MSCT angiography scan, can negatively affect left ventricular performance.

MSCT coronary plaque imaging
MSCT coronary angiography allows evaluation of both the coronary lumen and coronary plaques (Figure 2). Even plaques without encroachment of the coronary lumen from compensatory outward remodeling of the coronary vessel (‘positive remodeling’) can be detected, and the different plaques can be classified as noncalcified, calcified, and mixed [16–22].
Two published studies compared the diagnostic performances of 16-row MSCT coronary angiography and of intravascular ultrasound, for detection of plaques. They reported sensitivities of 82% and 86% for detection of any plaque, and specificities of 88% and 92% [16,18]. However, lower sensitivity was found for detection of noncalcified plaques, one study reporting a sensitivity of only 53% [16].
Schroeder et al [20] were the first to compare intravascular ultrasound and MSCT; they reported that noncalcified coronary plaques can be further classified as soft or fibrous, on the basis of CT density measurements. These findings were confirmed in ex-vivo and in-vivo studies comparing results of CT density measurements with those of histopathology or intravascular ultrasound [18,21,22], and may have important implications for the identification of high-risk coronary plaques. For instance, Leber et al [23] found significantly more noncalcified plaques in patients with an acute coronary syndrome than in stable patients. However, there is overlap between the range of CT density measurements in soft and fibrous plaques, probably related to the mixture of both tissue components within a single plaque.
Current MSCT technology allows assessment of the extent, severity, and localization of coronary plaques, and classification of plaque tissue components as non-calcified, mixed, and calcified. This might permit more advanced risk stratification by noninvasive assessment of the coronary plaque burden. We evaluated the coronary plaque burden in 40 patients with stable angina pectoris (current research), and found plaques in nearly 60% of all coronary segments ≥2mm and more than six plaques per patient. Plaque composition was: calcified in 62%, noncalcified in 27%, and mixed in 11%; plaques were predominantly located in proximal and mid parts of the main coronary vessels.
It is noteworthy that the spatial resolution of 16-row MSCT scanners allows detection only of advanced coronary plaques; earlier phases of atherosclerosis remain undetected [18]. Therefore, evaluation of the coronary plaques using MSCT coronary angiography will still underestimate the coronary plaque burden when compared with histopathological findings.▪

Back to the Summary

REFERENCES

Back to the Summary