DOI: 10.1148/rg.23si035514
(Radiographics. 2003;23:S111-S125.)
© RSNA, 2003
IMAGING OF THE CORONARY ARTERY |
Current Concepts in Multi–Detector Row CT Evaluation of the Coronary Arteries: Principles, Techniques, and Anatomy1
Harpreet K. Pannu, MD,
Thomas G. Flohr, PhD,
Frank M. Corl, MS and
Elliot K. Fishman, MD
1 From the Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins Medical Institutions, Baltimore, Md (H.K.P., F.M.C., E.K.F.); and the CT Division, Siemens Medical Solutions, Forchheim, Germany (T.G.F.). Presented as an education exhibit at the 2002 RSNA scientific assembly. Received March 5, 2003; revision requested April 16 and received May 21; accepted May 29. T.G.F. is an employee of Siemens Medical Solutions. E.K.F. is a consultant to the CT Advisory Board. Address correspondence to H.K.P., Department of Radiology, Johns Hopkins Hospital, 600 N Wolfe St, Baltimore, MD 21287 (e-mail: hpannu1@jhmi.edu).
 |
Abstract
|
|---|
Cardiac imaging is becoming a practical application of mechanical computed tomography (CT) with the availability of four, eight, and 16 detector row scanners. The role of imaging is progressing from simple determination of the presence of arterial calcifications on nonenhanced scans to demonstration of vascular stenoses on coronary CT angiograms. Optimization of the imaging technique and knowledge of coronary artery anatomy are both important for the development of CT of the heart. Technical factors such as a slow heart rate, a short scanning time, subcentimeter spatial resolution, high temporal resolution, and reconstruction of multiple image data sets at various intervals in the cardiac cycle result in optimal visualization of the coronary arteries. Axial, thin-slab maximum intensity projection, and volume-rendered images are used to display the normal anatomy and anomalies of the coronary arteries. The challenges of CT angiography of the coronary arteries have been partially met and will likely be overcome with continued evolution of the technology.
© RSNA, 2003
Index Terms: Computed tomography (CT), angiography, 54.12116 • Computed tomography (CT), multi–detector row, 54.12116 • Coronary vessels, CT, 54.12116 • Coronary vessels, stenosis or obstruction, 54.76
|
LEARNING OBJECTIVES FOR TEST 5
|
|---|
After reading this article and taking the test, the reader will be able to:
- Discuss the technical factors involved in acquiring and reconstructing contrast-enhanced CT images of the coronary arteries.
- Describe the anatomy of the coronary arteries.
- List the differences between multi–detector row CT, electron-beam CT, and conventional angiography.
|
Introduction
|
|---|
Contrast material–enhanced computed tomography (CT) of the coronary arteries is becoming feasible with mechanical CT scanners as the temporal and spatial resolution improves with the availability of four, eight, and 16 detector row CT. Sections as thin as 0.5 mm and temporal resolution as low as 105–250 msec are possible. Potential applications of coronary angiography with multi–detector row CT are diagnosis of noncalcified plaques, detection and quantification of coronary artery stenoses, and follow-up after surgical bypass therapy. Additional applications include evaluation of myocardial perfusion, scarring, and contractility. If this technique is successful, the key advantages would be the noninvasiveness of the study and the ability to evaluate both the coronary artery lumen and the vessel wall.
Challenges in evaluating the coronary arteries at CT are the small size of the vessels and the location adjacent to the moving heart. The vessels are typically 2–4 mm in diameter and are parallel, oblique, or perpendicular to the axial plane in portions. In addition, they are adjacent to both the atria and ventricles and therefore may be affected by cardiac motion in different phases of the cardiac cycle. Possible solutions are imaging on scanners with an increasing number of rows and faster rotation speeds and reconstructing multiple sets of images obtained in different phases of the cardiac cycle from a volume acquisition. The challenges of CT angiography of the coronary arteries have been partially met and will likely be addressed with the continued evolution of technology.
The technique of multi–detector row CT of the coronary arteries, postprocessing methods, and normal coronary artery anatomy at multi–detector row CT are reviewed. Assessment of the coronary arteries with multi–detector row CT and the role of multi–detector row CT in cardiac imaging are also discussed. Finally, the literature to date on coronary angiography with multi– detector row CT is presented.
|
Technique of Multi–Detector Row CT Coronary Angiography
|
|---|
Imaging is most likely to be successful if the patient has a slow and regular heart rate, the breath-holding and scanning times are short, and the spatial and temporal resolutions are high. The acquisition and processing of data are geared to achieving these goals.
Optimizing Heart Rate
The heart is in constant motion, more so in systole than in diastole. The duration of diastole depends on the heart rate and the time spent in systole (Appendix A) (1). The proportion of the cardiac cycle spent in systole increases with increasing heart rates and decreases with lowering of the heart rate (1,2). With higher heart rates, there is shortening of the end-diastolic interval (3). To obtain nearly motion-free images, acquisition or reconstruction of data is performed during diastole. Diastolic times are higher with heart rates of less than 75 beats per minute (1). A longer diastole increases the window of time available to acquire optimal images, especially if the acquisition time is relatively long. The temporal resolution of multi–detector row CT has improved, but technological limitations still require optimizing the heart rate for satisfactory imaging. Medications can increase diastole by decreasing the heart rate or shortening the systolic component (1). The diastolic time can be increased by giving patients ß-blockers to lower the heart rate. For example, for patients with a heart rate of greater than 65 beats per minute, 100 mg of metoprolol has been administered orally 1 hour prior to CT (4).
Lowering of the heart rate is suggested on the basis of comparison of coronary artery visibility in patients with low versus high heart rates. Most studies so far have been performed on four-section CT scanners with a minimum gantry rotation time of 0.5 seconds and a corresponding temporal resolution of 125–250 msec. In a study of 94 patients, there was an inverse correlation between the number of analyzable vessel segments and the heart rate (5). Vessel visibility was highest if the heart rate was less than 65 beats per minute (5). These results have been confirmed by other studies (3,4). In one study, image quality was diminished in 13% of arteries if the heart rate was less than 70 beats per minute and in 33% of arteries if the heart rate was greater than 70 beats per minute (3). As a result of unassessable segments, the sensitivity of multi–detector row CT for coronary artery stenosis was 62% for those with the lower heart rate and 33% for those with the higher heart rate (3). In another study, the sensitivity for stenosis was 82% for patients with low heart rates but only 32% for patients with rates greater than 80 beats per minute (6). Therefore, experience with established four-section CT scanners suggests that lowering the heart rate with pharmacologic therapy is helpful.
With the latest generation of 16-section CT scanners, gantry rotation times have been further reduced to 0.42 seconds. Whether the corresponding improvement in temporal resolution (105–210 msec) will affect these recommendations and allow reliable cardiac imaging also at higher heart rates has to be evaluated in clinical studies. A recent study of 59 patients who underwent coronary CT angiography with a 16-section system demonstrated 86% specificity and 95% sensitivity. None of the patients had to be excluded (4).
Decreasing Scanning Time
Reducing the scanning time decreases the likelihood of breathing artifacts and enhancement of the cardiac veins (6). The heart rate also tends to decrease or be relatively stable in the first 20 seconds of breath holding but may increase after that (2,6).
Initial studies of CT coronary angiography performed with four detector row scanners required a relatively long breath-hold time of approximately 40 seconds for thin-section collimation (1 mm/1.25 mm) (7). Scanning time can be reduced without compromising spatial resolution if scanners with faster rotation speeds and larger numbers of detectors are used. The area that needs to be covered extends from the aortic root to the apex of the heart. In most individuals, this area is approximately 10–12 cm long (8). With prospective gating, as with electron-beam CT, coverage is hampered by dependency on the heart rate (7). With retrospective gating on a 16 detector row CT unit, the entire heart is covered in approximately 20 seconds with submillimeter sections (4).
Optimizing Spatial Resolution
Spatial resolution is largely dependent on the type of scanner available. The 16 detector row scanners have an inherent advantage over the four detector row scanners. More detectors with smaller widths are available, with the design varying by manufacturer. The smallest detector widths range from 0.5 to 1.25 mm, and thin sections of the entire heart are feasible.
The coronary arteries are small, usually 2–4 mm in diameter (9). The diameter of the lumen is approximately 4 mm at the level of the proximal main segments and 1 mm in the branches (10). In addition to the vessels being small, they also have a complex course on the surface of the heart. The left anterior descending (LAD) artery is nearly parallel to the axial plane, whereas portions of the right coronary artery (RCA) and left circumflex (LCX) artery are perpendicular (11). Superior resolution in both the axial plane and z axis is therefore necessary at CT to evaluate the coronary arteries.
The spatial resolution of four detector row CT is 0.6 x 0.6 x 1.0 mm, that of electron-beam CT is 0.7 x 0.7 x 3 mm, and that of magnetic resonance (MR) coronary angiography is 1.25 x 1.25 x 1.5 mm (12). Spiral CT allows volume acquisition and reconstruction of overlapping sections, which improve z-axis resolution (10,13). The resolution of 16 detector row CT is up to 0.5 x 0.5 x 0.6 mm (10,14,15). This resolution is approaching, but remains inferior to, that of conventional angiography, which is 0.2 x 0.2 mm (16).
Optimizing Temporal Resolution
The temporal resolution is the amount of time it takes to acquire the necessary scan data to reconstruct an image (17). The temporal resolution of electron-beam CT is 100 msec, and that of MR imaging is 100–150 msec (6). For multisection CT, it is primarily dependent on the time taken by the scanner to complete one gantry rotation but can be modified by using partial scan reconstruction techniques. With these techniques, the image is reconstructed by using data acquired from a gantry rotation of approximately 240° (180° plus the total detector fan angle). By using optimized reconstruction algorithms, 180° of data in parallel geometry are extracted from the acquired data and reconstructed (17). This improves the temporal resolution to one-half the gantry rotation time (10). For example, if the rotation time of a four detector row CT scanner is 500 msec, the resulting temporal resolution is 250 msec; if the rotation time of a 16 detector row CT scanner is 420 msec, the resulting temporal resolution is 210 msec.
The temporal resolution equal to one-half the gantry rotation time (rotation time/2) works for low heart rates but can give blurring and stair-step artifacts with higher heart rates (17). The temporal resolution for higher heart rates is improved by using data from more than one cardiac cycle to reconstruct the image (17,18). The data from one, two, or three consecutive cycles are used, depending on the heart rate. The higher the heart rate, the more cycles are used. The resulting temporal resolution can be as good as the gantry rotation time divided by 2N (N = number of cycles) (Appendix B). For example, if the rotation time is 500 msec and if data from two cycles are used, the temporal resolution goes up to 125 msec (17). The temporal resolution with this approach is not constant but depends on the heart rate. A drawback of this technique is that spatial resolution in the z axis may suffer if the pitch is too high for the heart rate and there are gaps in the acquired data (17). If data from a single cardiaccycle are used for generating an image, it is called single-sector reconstruction or single-segment reconstruction. If data from multiple cycles are used, it is called multisector reconstruction or multisegment reconstruction. In general, for heart rates of less than 65 beats per minute, single-sector reconstruction is performed. Overall, the current temporal resolution of multi–detector row CT is approximately 125–250 msec for four- and eight-section systems and 105–210 msec for 16-section systems with a 0.42-second gantry rotation time.
Decreasing Motion Artifact
With multi–detector row spiral CT, the data are acquired as a volume and retrospective electrocardiographic (ECG) gating is performed to reconstruct the images. An ECG tracing is obtained simultaneously with the acquisition, and a certain time interval from the R wave is chosen to start reconstruction.
On an ECG tracing, the P wave signifies atrial activation, the PR segment is due to atrioventricular conduction and the time of atrial contraction, the QRS complex is due to ventricular activation, and the ST-T wave occurs during ventricular recovery. Closure of the aortic valve or the end of ventricular contraction occurs approximately at the end of the T wave. There is rapid, and then slow, filling of the ventricles between the T and P waves.
Images are reconstructed during the time of least motion or between the T and P waves. Images are reconstructed with a predefined temporal offset relative to the R waves of the patient’s ECG signal; this delay can be either relative (given as a certain percentage of the R-R interval time) or absolute (given in milliseconds) and either forward or reverse (Fig 1) (10,13). In the relative delay method, a certain delay from the prior wave is determined as a percentage of the R-R interval. This delay is used to start reconstruction. In the absolute delay method, a fixed time delay after the R wave determines the start of image reconstruction. Finally, in the absolute reverse method, a fixed time before the next R wave is the point for starting reconstruction. The relative delay method is usually used for reconstruction.
View larger version (97K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1a. Techniques of retrospective image reconstruction. (a) ECG strip shows the relative delay method, in which a certain delay from the prior wave is determined as a percentage of the R-R interval. This delay is used to start reconstruction. (b) ECG strip shows the absolute delay method, in which a fixed time delay after the R wave determines the start of reconstruction. (c) ECG strip shows the absolute reverse method, in which a fixed time before the next R wave determines the start of reconstruction.
|
|
View larger version (98K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1b. Techniques of retrospective image reconstruction. (a) ECG strip shows the relative delay method, in which a certain delay from the prior wave is determined as a percentage of the R-R interval. This delay is used to start reconstruction. (b) ECG strip shows the absolute delay method, in which a fixed time delay after the R wave determines the start of reconstruction. (c) ECG strip shows the absolute reverse method, in which a fixed time before the next R wave determines the start of reconstruction.
|
|
View larger version (97K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1c. Techniques of retrospective image reconstruction. (a) ECG strip shows the relative delay method, in which a certain delay from the prior wave is determined as a percentage of the R-R interval. This delay is used to start reconstruction. (b) ECG strip shows the absolute delay method, in which a fixed time delay after the R wave determines the start of reconstruction. (c) ECG strip shows the absolute reverse method, in which a fixed time before the next R wave determines the start of reconstruction.
|
|
However, choosing a single reconstruction time in the cardiac cycle may not be optimal for displaying all of the coronary arteries without motion artifact (11,19). The coronary arteries can move in the axial plane with considerable velocities. A study at the level of the mid-RCA found that the mean velocity was approximately 47 mm/sec (20). The greatest motion was seen for the RCA; the least was seen for the LAD artery. The RCA and LCX artery are closer to the atria than is the LAD artery. As a result, they are affected by atrial contraction. Each of the coronary arteries is most susceptible to motion artifacts in different parts of the cardiac cycle (3,11). Therefore, reconstruction of multiple sets of images in different parts of the cardiac cycle has been attempted to optimize vascular visualization (4).
By reconstructing seven sets of images at 10% intervals in the cardiac cycle, Kopp et al (21) found that the RCA was best seen early in diastole at 40% of the cardiac cycle, the LCX artery was best seen at midcycle, and the LAD artery was best seen slightly later, at 60%–70% of the cardiac cycle (11,21). In another study, multiple reconstructions were performed at 30%–80% of the R-R interval, and a 50% delay was optimal for the RCA, a 50% or 60% delay was optimal for the LAD artery, and a 60% delay was optimal for the LCX artery (2). In practice, multiple reconstructions may be necessary to select the best image data set for each vessel in a particular patient.
An advantage of retrospective gating with multi–detector row CT over prospective triggering with electron-beam CT is that there is flexibility in reconstruction, as the data are acquired as a volume. It is also less sensitive to aberrant heartbeats (10,22).
Decreasing Radiation Dose
The signal-to-noise ratio is high with multi– detector row CT, as adjustments in tube current are possible to maintain image quality (12). However, continuous scanning through the entire cardiac cycle results in a higher radiation dose with multi–detector row CT than with electron-beam CT (12). The dose from multi–detector row CT is approximately 4–7 mSv and is similar to that from uncomplicated conventional coronary angiography (7,11,16,19). In a study of conventional coronary angiography performed by multiple cardiologists, the average dose varied from 3.1 mSv to 8.6 mSv (23). Attempts have been made to modulate the tube output according to the patient’s ECG to decrease the dose given during systole (4,16). By reducing the tube output in systole, dose reductions of 45%–48% can be achieved, depending on the patient’s heart rate (24).
Technique Summary
Initially, the patient’s heart rate is determined (Table 1). If the rate is greater than 65–70 beats per minute, a ß-blocker such as metoprolol can be given orally 1 hour prior to CT (4,25) or the appropriate intravenous dose can be administered just prior to scanning. Nitroglycerin (0.4 mg) has also been given to dilate the coronary arteries but may result in reflex tachycardia (8,11).
A 120-mL dose of nonionic iodinated contrast material is injected intravenously at approximately 4 mL/sec for CT angiography (4,11,25). A saline solution bolus can also be given following contrast material injection to decrease artifact from contrast material in the right heart. If a saline solution bolus is used, the total amount of contrast material injected is reduced. Scanning is triggered once contrast material is seen in the ascending aorta, or a test bolus is administered to calculate the appropriate delay (4,11). Typical delays in one study were 10–18 seconds (13). Detector collimation depends on the type of scanner and is 4 x 1 mm for a four detector row unit and 12 x 0.75 mm or 16 x 0.75 mm for a 16 detector row unit (Siemens Medical Solutions, Malvern, Pa) (Table 1) (4,11). Other parameters are a kilovolt peak of 120, milliamperage of 300–400, pitch factor of 0.28–0.375, and matrix of 512 x 512.
The images are reconstructed by using a medium soft-tissue kernel with retrospective ECG gating. One or multiple image sets are reconstructed in diastole by using the methods described earlier, such as a 40%–70% relative delay or 300–550-msec absolute reverse ECG gating. In general, reconstruction is avoided at 10%–30% or greater than 80% of the R-R interval, as these times are particularly susceptible to motion artifacts (2). On a four detector row unit, 1.25-mm-thick sections are reconstructed at 0.6-mm intervals; on a 16 detector row unit, 1-mm-thick sections are reconstructed at 0.5-mm intervals. On a 16 detector row unit, additional 0.75-mm-thick sections can be reconstructed for detail viewing of severely calcified coronary arteries or stents at the expense of increased image noise.
Display of Multi–Detector Row CT Coronary Angiograms
Three-dimensional and multiplanar views are used to supplement the axial CT images. Portions of all vessels, in particular the left main artery and LAD artery, can be evaluated on the axial images. Longer vessel segments are demonstrated on curved multiplanar reformatted and three-dimensional volume-rendered images. Calcifications and noncalcified plaques are assessed on thin-slab maximum intensity projection and volume-rendered images (11). Virtual angioscopic views are also possible.
In a study that compared axial, virtual angioscopic, volume-rendered, and multiplanar reformatted images, the most stenoses were seen on axial images followed by virtual angioscopic, volume-rendered, and multiplanar reformatted images (25). Use of all four techniques gave the highest sensitivity. A combination of various viewing methods has been used in most studies (4,11). The most effective method for reformation has still to be determined, but thin-slab maximum intensity projection seems to be most widely used.
|
Coronary Artery Anatomy
|
|---|
The four main coronary arteries evaluated at CT are the RCA, left main artery, LAD artery, and LCX artery (Fig 2) (26). The coronary circulation is right dominant if the RCA gives rise to the posterior descending artery (PDA) and posterior left ventricular branches (27). This occurs in 85% of individuals. The circulation is left dominant in 8% of individuals, as the PDA and posterior left ventricular branches arise from the LCX artery. In 7% of individuals, a codominant system exists in which the PDA arises from the RCA and the posterior left ventricular branches arise from the LCX artery. The coronary arteries have also been divided into segments, and different nomenclatures are used (25,28).
View larger version (157K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a. Coronary artery anatomy at CT. In all figures, A = aorta, LA = left atrium, LV = left ventricle, P = pulmonary trunk. (a) Oblique volume-rendered image of the top of the heart shows the origins of the right (arrow) and left (arrowhead) main coronary arteries, which demonstrate high-attenuation calcifications. (b, c) Coronal oblique volume-rendered (b) and maximum intensity projection (c) images of the anterior heart show the right (arrow) and left (arrowhead) coronary arteries.
|
|
View larger version (144K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b. Coronary artery anatomy at CT. In all figures, A = aorta, LA = left atrium, LV = left ventricle, P = pulmonary trunk. (a) Oblique volume-rendered image of the top of the heart shows the origins of the right (arrow) and left (arrowhead) main coronary arteries, which demonstrate high-attenuation calcifications. (b, c) Coronal oblique volume-rendered (b) and maximum intensity projection (c) images of the anterior heart show the right (arrow) and left (arrowhead) coronary arteries.
|
|
View larger version (89K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2c. Coronary artery anatomy at CT. In all figures, A = aorta, LA = left atrium, LV = left ventricle, P = pulmonary trunk. (a) Oblique volume-rendered image of the top of the heart shows the origins of the right (arrow) and left (arrowhead) main coronary arteries, which demonstrate high-attenuation calcifications. (b, c) Coronal oblique volume-rendered (b) and maximum intensity projection (c) images of the anterior heart show the right (arrow) and left (arrowhead) coronary arteries.
|
|
Right Coronary Artery
The RCA arises from the right coronary sinus of the aorta. It runs rightward posterior to the pulmonary outflow tract and then inferiorly in the right atrioventricular groove toward the posterior interventricular septum (Fig 3).
View larger version (146K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. RCA. (a) Axial CT image shows origin of the RCA (arrow) from the aorta and extension into the right atrioventricular groove. The artery is minimally calcified. (b) Coronal oblique volume-rendered image shows the caudal course of the proximal RCA (arrow). (c) Coronal oblique volume-rendered image shows the RCA (arrow) coursing in the groove between the right atrium (RA) and right ventricle (RV). (d) Coronal oblique volume-rendered image shows the course of the RCA (arrow) posterior to the pulmonary trunk. (e) Posterior coronal oblique volume-rendered image shows the RCA (long arrow), PDA branch (short arrow), and posterolateral left ventricular branch (arrowhead).
|
|
View larger version (114K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. RCA. (a) Axial CT image shows origin of the RCA (arrow) from the aorta and extension into the right atrioventricular groove. The artery is minimally calcified. (b) Coronal oblique volume-rendered image shows the caudal course of the proximal RCA (arrow). (c) Coronal oblique volume-rendered image shows the RCA (arrow) coursing in the groove between the right atrium (RA) and right ventricle (RV). (d) Coronal oblique volume-rendered image shows the course of the RCA (arrow) posterior to the pulmonary trunk. (e) Posterior coronal oblique volume-rendered image shows the RCA (long arrow), PDA branch (short arrow), and posterolateral left ventricular branch (arrowhead).
|
|
View larger version (129K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3c. RCA. (a) Axial CT image shows origin of the RCA (arrow) from the aorta and extension into the right atrioventricular groove. The artery is minimally calcified. (b) Coronal oblique volume-rendered image shows the caudal course of the proximal RCA (arrow). (c) Coronal oblique volume-rendered image shows the RCA (arrow) coursing in the groove between the right atrium (RA) and right ventricle (RV). (d) Coronal oblique volume-rendered image shows the course of the RCA (arrow) posterior to the pulmonary trunk. (e) Posterior coronal oblique volume-rendered image shows the RCA (long arrow), PDA branch (short arrow), and posterolateral left ventricular branch (arrowhead).
|
|
View larger version (140K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3d. RCA. (a) Axial CT image shows origin of the RCA (arrow) from the aorta and extension into the right atrioventricular groove. The artery is minimally calcified. (b) Coronal oblique volume-rendered image shows the caudal course of the proximal RCA (arrow). (c) Coronal oblique volume-rendered image shows the RCA (arrow) coursing in the groove between the right atrium (RA) and right ventricle (RV). (d) Coronal oblique volume-rendered image shows the course of the RCA (arrow) posterior to the pulmonary trunk. (e) Posterior coronal oblique volume-rendered image shows the RCA (long arrow), PDA branch (short arrow), and posterolateral left ventricular branch (arrowhead).
|
|
View larger version (140K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3e. RCA. (a) Axial CT image shows origin of the RCA (arrow) from the aorta and extension into the right atrioventricular groove. The artery is minimally calcified. (b) Coronal oblique volume-rendered image shows the caudal course of the proximal RCA (arrow). (c) Coronal oblique volume-rendered image shows the RCA (arrow) coursing in the groove between the right atrium (RA) and right ventricle (RV). (d) Coronal oblique volume-rendered image shows the course of the RCA (arrow) posterior to the pulmonary trunk. (e) Posterior coronal oblique volume-rendered image shows the RCA (long arrow), PDA branch (short arrow), and posterolateral left ventricular branch (arrowhead).
|
|
The first branch of the RCA is the conus artery. It arises from the RCA or has a separate origin directly from the right coronary sinus (29). The sinus node artery also arises from the proximal RCA in 60% of individuals within a few millimeters of the RCA origin and runs superiorly and posteriorly (29,30). In the remaining cases, it arises from the proximal LCX artery. Next, several anterior branches supply the free wall of the right ventricle. The branch to the right ventricle at the junction of the mid- and distal RCA is called the acute marginal (30).
The distal RCA divides into the PDA and the posterior left ventricular branches (Fig 4). The PDA runs in the posterior interventricular groove. If the LAD artery, which usually supplies the apex of the heart, is small, the PDA can extend around the apex to supply one-third of the anterior interventricular septum (29).
View larger version (139K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. PDA. (a, b) Coronal oblique maximum intensity projection images show the PDA branch (arrow) and posterolateral left ventricular branches (arrowheads) of the RCA. (c, d) Axial oblique (c) and posterior coronal (d) volume-rendered images show the PDA (arrow) coursing in the posterior interventricular groove and the posterolateral left ventricular branch (arrowhead). RV = right ventricle.
|
|
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. PDA. (a, b) Coronal oblique maximum intensity projection images show the PDA branch (arrow) and posterolateral left ventricular branches (arrowheads) of the RCA. (c, d) Axial oblique (c) and posterior coronal (d) volume-rendered images show the PDA (arrow) coursing in the posterior interventricular groove and the posterolateral left ventricular branch (arrowhead). RV = right ventricle.
|
|
View larger version (158K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4c. PDA. (a, b) Coronal oblique maximum intensity projection images show the PDA branch (arrow) and posterolateral left ventricular branches (arrowheads) of the RCA. (c, d) Axial oblique (c) and posterior coronal (d) volume-rendered images show the PDA (arrow) coursing in the posterior interventricular groove and the posterolateral left ventricular branch (arrowhead). RV = right ventricle.
|
|
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4d. PDA. (a, b) Coronal oblique maximum intensity projection images show the PDA branch (arrow) and posterolateral left ventricular branches (arrowheads) of the RCA. (c, d) Axial oblique (c) and posterior coronal (d) volume-rendered images show the PDA (arrow) coursing in the posterior interventricular groove and the posterolateral left ventricular branch (arrowhead). RV = right ventricle.
|
|
Left Main Coronary Artery
The left main artery arises from the left coronary sinus, is 5–10 mm long, and is constant in diameter (29). It passes leftward posterior to the pulmonary trunk and bifurcates into the LAD and LCX arteries (Fig 5). Occasionally, the left main artery trifurcates into the LAD artery, LCX artery, and ramus intermedius. The ramus intermedius has a course similar to that of the first diagonal branch of the LAD artery to the anterior left ventricle (30). Also, in 0.41% of cases, the left main artery is absent and the LAD and LCX arteries arise separately from the left coronary sinus (27,31).
View larger version (136K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5a. Left main coronary artery. (a) Axial CT image shows origin of the left main coronary artery (arrow) from the aorta. (b-d) Axial oblique (b) and coronal oblique (c, d) volume-rendered images show bifurcation of the left main coronary artery (black arrow). The LAD artery and diagonal branches (white arrows) and the LCX artery (arrowhead) are also seen. High-attenuation areas in the vessels represent calcifications.
|
|
View larger version (133K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5b. Left main coronary artery. (a) Axial CT image shows origin of the left main coronary artery (arrow) from the aorta. (b-d) Axial oblique (b) and coronal oblique (c, d) volume-rendered images show bifurcation of the left main coronary artery (black arrow). The LAD artery and diagonal branches (white arrows) and the LCX artery (arrowhead) are also seen. High-attenuation areas in the vessels represent calcifications.
|
|
View larger version (157K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5c. Left main coronary artery. (a) Axial CT image shows origin of the left main coronary artery (arrow) from the aorta. (b-d) Axial oblique (b) and coronal oblique (c, d) volume-rendered images show bifurcation of the left main coronary artery (black arrow). The LAD artery and diagonal branches (white arrows) and the LCX artery (arrowhead) are also seen. High-attenuation areas in the vessels represent calcifications.
|
|
View larger version (139K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5d. Left main coronary artery. (a) Axial CT image shows origin of the left main coronary artery (arrow) from the aorta. (b-d) Axial oblique (b) and coronal oblique (c, d) volume-rendered images show bifurcation of the left main coronary artery (black arrow). The LAD artery and diagonal branches (white arrows) and the LCX artery (arrowhead) are also seen. High-attenuation areas in the vessels represent calcifications.
|
|
LAD Artery
The LAD artery runs in the anterior interventricular groove and terminates near the apex of the heart (Fig 6). It gives off diagonal branches to the anterior free wall of the left ventricle and septal branches to the anterior interventricular septum (30). These branches are numbered as they arise from the LAD artery.
LCX Artery
The LCX artery runs in the left atrioventricular groove (Fig 7). It gives off obtuse marginal branches to the lateral left ventricle, which are numbered as they arise from the LCX artery (30). In a left dominant or codominant anatomy, the LCX artery gives rise to the PDA and/or posterior left ventricular branches.
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7a. LCX artery. (a) Axial CT image shows the LCX artery (straight solid arrow) in the left atrioventricular groove, the RCA (open arrow) in the right atrioventricular groove, and the LAD artery (curved arrow) coursing toward the apex of the heart. (b-d) Coronal oblique volume-rendered image (b), oblique volume-rendered image of the top of the heart (c), and axial oblique volume-rendered image (d) show the LCX artery (arrow) in the left atrioventricular groove and obtuse marginal branches (arrowhead) to the lateral left ventricle.
|
|
View larger version (149K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7b. LCX artery. (a) Axial CT image shows the LCX artery (straight solid arrow) in the left atrioventricular groove, the RCA (open arrow) in the right atrioventricular groove, and the LAD artery (curved arrow) coursing toward the apex of the heart. (b-d) Coronal oblique volume-rendered image (b), oblique volume-rendered image of the top of the heart (c), and axial oblique volume-rendered image (d) show the LCX artery (arrow) in the left atrioventricular groove and obtuse marginal branches (arrowhead) to the lateral left ventricle.
|
|
View larger version (160K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7c. LCX artery. (a) Axial CT image shows the LCX artery (straight solid arrow) in the left atrioventricular groove, the RCA (open arrow) in the right atrioventricular groove, and the LAD artery (curved arrow) coursing toward the apex of the heart. (b-d) Coronal oblique volume-rendered image (b), oblique volume-rendered image of the top of the heart (c), and axial oblique volume-rendered image (d) show the LCX artery (arrow) in the left atrioventricular groove and obtuse marginal branches (arrowhead) to the lateral left ventricle.
|
|
View larger version (159K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7d. LCX artery. (a) Axial CT image shows the LCX artery (straight solid arrow) in the left atrioventricular groove, the RCA (open arrow) in the right atrioventricular groove, and the LAD artery (curved arrow) coursing toward the apex of the heart. (b-d) Coronal oblique volume-rendered image (b), oblique volume-rendered image of the top of the heart (c), and axial oblique volume-rendered image (d) show the LCX artery (arrow) in the left atrioventricular groove and obtuse marginal branches (arrowhead) to the lateral left ventricle.
|
|
Anomalies
An anomalous origin and course of the coronary arteries can be benign or life threatening. The prevalence of coronary artery anomalies is 0.85%, and they can be responsible for sudden death in young adults (31). The anomalies are most commonly seen in individuals with congenital heart disease and bicuspid aortic valves. In a study of conventional angiograms of 4,250 patients, the most common anomaly was origin of the LCX artery from the RCA or right sinus of Valsalva, which was noted in 30 patients (32). The next most common anomaly was separate origins of the LCX and LAD arteries from the left sinus of Valsalva, which was present in eight of 4,250 patients. In most cases, the anomalous LCX artery has a retroaortic course to the left atrioventricular groove (31). This path is not life threatening as the LCX artery is not compressed between the aorta and main pulmonary artery.
A clinically significant anomaly is origin of the RCA from the left coronary sinus, which was seen in three of 4,250 patients undergoing angiography (31,32). The RCA often lies between the aorta and pulmonary artery as it courses to the right side and is susceptible to compression, particularly during exercise. Other anomalies are origin of the left main artery or LAD artery from the right coronary sinus or RCA, origin of the left main artery from the pulmonary artery, and fistulas between the coronary arteries and right-sided structures such as the right ventricle and coronary sinus (31).
|
Assessment of Coronary Arteries with Multi–Detector Row CT
|
|---|
Studies of multi–detector row CT of the coronary arteries have focused on the diagnostic visibility of vessels and detection of stenoses in comparison with conventional angiography. Most studies have been performed with four detector row scanners, and reasons for unassessable vessels include cardiac motion, arrhythmia, severe calcifications, vessel size less than 1.5 mm, breathing, presence of stents, and poor enhancement (7). "Blooming" of high-attenuation material such as wall calcification and stents secondary to soft-tissue reconstruction kernels obscures the vessel lumen (33). The number of unassessable vessels is lower with 16 detector row CT (7%) than with four detector row CT (27%) (4,7).
Sensitivities for stenoses are lower if both assessable and unassessable vessel segments are included but are higher if reported only for vessel segments that are adequately seen (Table 2). As expected, sensitivities are higher for proximal vessel disease compared with distal vessel disease (34). A high sensitivity of 95% and specificity of 86% were reported for detecting greater than 50% diameter stenoses in vessels greater than 2 mm in diameter with 16 detector row CT (4). In the New Age Pilot Trial, which compared multi–detector row CT with angiography and intravascular ultrasonography (US), the severity of 85% of plaques was correctly determined with multi–detector row CT and there was good correlation with US for vessel size (35). Additional studies are necessary to more completely evaluate the diagnostic accuracy of multi–detector row CT for coronary artery disease.
View this table:
[in this window]
[in a new window]
|
TABLE 2. Sensitivity and Specificity of Multi-Detector Row CT for Greater than 50% Diameter Stenosis in the Coronary Arteries
|
|
The results of multi–detector row CT compare favorably with those of electron-beam CT and MR coronary angiography. Approximately 20%–30% of vessel segments are unassessable with electron-beam CT and MR imaging, and sensitivity for stenosis ranges from 50% to 80% (8,9,36). At electron-beam CT, a compromise has to be made between z-axis resolution and scan coverage, so the distal coronary arteries may not be included in the acquisition. In addition, spatial resolution and signal-to-noise ratio are lower with electron-beam CT, whereas scanning time is higher (37). Therefore, evaluation with electron-beam CT and MR imaging is usually limited to the proximal and middle vessel segments (4). With multi–detector row CT, the entire heart can be covered, so both proximal and distal vessel segments can be visualized (34).
An attempt has also been made to visualize bypass grafts with multi–detector row CT. By using a four detector row scanner, graft occlusion was reliably diagnosed and sensitivity for stenosis in assessable grafts was 75%, with a specificity of 92% (38). Metal clip artifact was a common factor limiting satisfactory visualization of all grafts. Early work characterizing atherosclerotic plaques suggests that lipid plaques have an attenuation of less than 50 HU, intermediate plaques have an attenuation of 50–119 HU, and calcified plaques have an attenuation of greater than 120 HU (16,18). The effect of partial volume averaging on attenuation measurements is reduced with thin sections (39). Lipid plaques are not as stenotic but are prone to rupture, which can lead to acute occlusive thrombosis (40).
|
Role of Multi–Detector Row CT in Cardiac Imaging
|
|---|
Coronary artery disease is a common cause of death in adults in this country. Approximately 25% of all deaths in the United States are secondary to coronary artery disease; in 50% of fatalities, the individuals have no prior symptoms (10). Coronary artery disease is secondary to narrowing of the coronary vessels from atherosclerosis (10). Injury to the endothelium leads to an inflammatory reaction, and there is accumulation of inflammatory cells, smooth muscle cells, and fat deposits in the vessel wall. These lead to formation of an atherosclerotic plaque that narrows the lumen. Risk factors are well known and include age, male gender, lack of exercise, obesity, high blood pressure, elevated blood lipid levels, smoking, and diabetes.
Currently, imaging of coronary artery disease is performed with a variety of methods. Nonenhanced CT (electron-beam CT or multi–detector row CT) is used to detect calcified plaques, a nuclear medicine thallium stress test demonstrates areas of myocardial ischemia or infarction, and catheter coronary angiography allows definition of vascular stenoses. The noninvasive examinations such as nonenhanced CT and stress testing provide only indirect evidence or partial evaluation of coronary artery disease, and diagnosis of vascular narrowing currently requires invasive conventional angiography (34). Fluoroscopic angiography has a 0.15% mortality rate and 1.5% morbidity rate (19). A noninvasive test would be advantageous for detecting coronary artery disease in patients with unexplained symptoms but a low likelihood of disease, a history of recent myocardial infarction, or recurrent angina after percutaneous intervention (8).
Recent advances in the technology of mechanical CT scanners have shown that angiography of the coronary arteries is feasible. The coronary arteries are well seen, and the sensitivity for detecting stenoses is high in individuals with slow heart rates (4). An anomalous course of the coronary arteries is also easy to diagnose on axial or three-dimensional images. Finally, reconstruction of the images in systole and diastole provides information on ventricular function (13,22,41).
|
Conclusions
|
|---|
CT coronary angiography is becoming feasible with the availability of faster mechanical scanners with multisection imaging capabilities. Potential applications of multi–detector row CT angiography include noninvasive diagnosis of plaques and coronary artery stenoses. Three-dimensional volume-rendered and multiplanar images display arterial anatomy similarly to conventional angiography. Close attention to technique to maximize spatial and temporal resolution while decreasing motion artifacts is necessary for successful imaging.
|
Appendix A
|
|---|
The diastolic interval of the cardiac cycle is calculated as follows (1):
1. Diastolic time = R-R interval - electromechanical systole.
2. R-R interval = 60,000/HR, where R-R interval is in milliseconds and HR = heart rate per minute.
3. Electromechanical systole = 546 - (2.1 x HR) in males.
4. Electromechanical systole = 549 - (2.0 x HR) in females.
5. Diastolic percentage = 100 - (546 x HR/600) + (2.1 x HR x HR/600) in males.
6. Diastolic percentage = 100 - (549 x HR/600) + (2.0 x HR x HR/600) in females.
|
Appendix B: Temporal Resolution
|
|---|
The temporal resolution, which is limited to one-half the gantry rotation time for prospective electrocardiographic triggering, can be improved up to 1/(2N) times the rotation time by using scan data of N subsequent cardiac cycles for image reconstruction (17). With increased N, better temporal resolution is achieved but at the expense of reduced volume coverage within one breath-hold time or loss of transverse resolution. To maintain good transverse resolution and thin-section images, every z position of the heart has to be imaged by a detector section at every time during the N cardiac cycles. As a consequence, the larger the N and the lower the heart rate, the more the spiral pitch has to be reduced. If the pitch is too high, there will be z positions that are not covered by a detector section in the desired phase of the cardiac cycle. To obtain images at these z positions, far-reaching spiral interpolations have to be performed, which degrade section sensitivity profiles and reduce transverse resolution.
At higher heart rates, temporal resolution can be improved by dividing the partial scan data segment used for image reconstruction into N subsegments acquired in subsequent cardiac cycles. Each subsegment is generated by using data from one heart period only, and there are temporal gaps between the multisection data segments used for image reconstruction. Similar to the case N = 1 (discussed earlier), for each projection angle within subsegment j a linear interpolation is performed between the data of those two detector sections that are in closest proximity to the desired image plane. The result is N single-section partial scan subsegments located at the given image z position.
With this technique, the heart rate and the gantry rotation time of the scanner have to be properly desynchronized to allow improved temporal resolution. Two requirements have to be met: First, the start- and end-projection angles of the subsegments have to fit together to build up a full partial scan interval. As a consequence, the start projections of subsequent subsegments have to be shifted relative to each other. Second, all subsegments have to be acquired in the same relative phase of the patient’s cardiac cycle to reduce the total time interval contributing to an image. If the patient’s cardiac cycle and the rotation of the scanner are completely synchronous, the two requirements are contradictory. For instance, for a heart rate of 60 beats per minute an
Top
Abstract
LEARNING OBJECTIVES FOR TEST...
