Coronary CT Angiography

Updated: Nov 20, 2016
  • Author: Eugene C Lin, MD; Chief Editor: Eugene C Lin, MD  more...
  • Print
Overview

Practice Essentials

Coronary computed tomography angiography (CCTA) is a noninvasive method to image the coronary arteries. Applications include the following:

  • Diagnosis of coronary artery disease (CAD)
  • Diagnosis of in-stent restenosis
  • Evaluation of coronary bypass graft patency

Clinical application in CAD

Based on the combined efforts of 9 specialty societies, [1] the following indications were rated as appropriate for CCTA:

  • Detection of CAD in symptomatic patients without known heart disease, either nonacute or acute presentations
  • Detection of CAD in patients with new-onset or newly diagnosed clinical heart failure and no prior CAD
  • Preoperative coronary assessment prior to noncoronary cardiac surgery
  • Patients with prior electrocardiographic exercise testing - Normal test with continued symptoms or intermediate risk Duke treadmill score
  • Patients with prior stress imaging procedures - Discordant electrocardiographic exercise and imaging results or equivocal stress imaging results
  • Evaluation of new or worsening symptoms in the setting of a past normal stress imaging study
  • Risk assessment post-revascularization - Symptomatic if post-coronary artery bypass grafting or asymptomatic with prior left main coronary stent of 3 mm or greater
  • Evaluation of cardiac structure and function in adult congenital heart disease
  • Evaluation of cardiac structure and function - Ventricular morphology and systolic function
  • Evaluation of cardiac structure and function - Intracardiac and extracardiac structures

Recent technology

Dual-source CT-scanning

The primary advantage of dual-source CT scanning is greater temporal resolution, which allows CCTA to be performed at higher heart rates without the use of beta blockers.

256- and 320-slice CT scanning

The primary advantage of 256- and 320-slice CT is the increased craniocaudal coverage. In a comparison of prospectively gated 64- and 256-slice CT scanning, the 256-slice scan provided better and more stable image quality, at equivalent effective radiation dose. [2]

Radiation dose

The reported effective radiation doses for retrospectively gated, single-source, 64-slice CT scanning have ranged from 9.5-21.4 mSv. [3] However, various technologies and techniques have made it possible to lower the dose to less than 5 mSv, and doses of less than 1 mSv are possible in some patients.

Patient preparation

At the author’s institution, patients are instructed to avoid caffeine and smoking 12 hours prior to the procedure to avoid cardiac stimulation. They are also instructed to avoid eating solid food 4 hours before the study and to increase fluid intake prior to the exam. Standard precautions with regard to contrast allergy history and renal function are taken.

Beta blockers

Beta-blocker administration is often helpful in cardiac CT scanning to lower the heart rate and decrease motion artifact. The level to which the heart rate should be lowered depends on the temporal resolution of the scan.

However, heart rate variability may be a more important determinant of image quality than absolute heart rate. [4] Beta blockers are also helpful in patients with irregular heart rates, supraventricular tachycardias, and arrhythmias.

Several contraindications to beta-blocker therapy exist, including a heart rate below 60 bpm, a systolic blood pressure below 100 mm Hg, and decompensated cardiac failure, among others.

Nitroglycerin

The administration of sublingual nitroglycerin dilates the coronary arteries and increases side branch visualization. [5] Nitroglycerin is contraindicated in patients who are allergic to it and in patients who are taking phosphodiesterase inhibitors for erectile dysfunction. [6] Patients should not have taken a phosphodiesterase inhibitor for at least 48 hours before the exam.

Nitroglycerin can cause orthostatic hypotension; it should be used with caution in patients who have low systolic blood pressure (eg, < 90 mm Hg) and who are volume depleted from diuretic therapy. Angina caused by hypertrophic cardiomyopathy can also be aggravated.

Artifacts

These include the following:

  • Stairstep artifacts: Associated with heart rate variability [4] (see the image below)
    Stairstep artifact: Volume-rendered CT image demon Stairstep artifact: Volume-rendered CT image demonstrates "stairstep" artifact. This is a phase-misregistration artifact that is typically secondary to an irregular heartbeat.
  • Coronary artery motion artifacts: Result in image blurring (see the image below)
    Coronary artery motion: Axial CT images reconstruc Coronary artery motion: Axial CT images reconstructed at 90% (A) and 70% (B) of the R-R interval demonstrate the importance of imaging during the phase of least cardiac motion.
  • Respiratory motion artifacts
  • Streak artifacts: Can result from beam hardening secondary to metal clips or can be caused by dense contrast
  • Blooming artifacts: Can cause small, high-contrast structures such as stents and calcium to appear larger than they are [4] (see the image below)
    Blooming artifact: Multiplanar reconstruction (MPR Blooming artifact: Multiplanar reconstruction (MPR) CT image demonstrates blooming artifact from a left anterior descending (LAD) artery stent. The stent lumen is poorly visualized secondary to this artifact.

Image reconstruction

The coronary arteries are optimally imaged when there is the least cardiac motion. This occurs during so-called rest periods, which is typically in middiastole (diastasis). Coronary motion is also minimal during end-systole (isovolumic relaxation), but this is of shorter duration than diastolic diastasis at low heart rates.

A variety of postprocessing techniques are useful in CCTA. [7] Many interpreting physicians will start with the axial source images and then utilize multiplanar reconstructions in at least 2 planes.

Stenosis grading

There are many different methods to grade the degree of stenosis, including the following:

  • Visual assessment
  • Manually determined diameter or cross-sectional area on multiplanar reformats perpendicular to the median centerline of the vessel ("end-on" view)
  • Diameter on maximum intensity projection (MIP) images parallel to the long axis of the vessel
  • Software calculation of diameter or area [8]
Next:

Overview

Coronary computed tomography angiography (CCTA) has recently emerged as an effective noninvasive method to image the coronary arteries. The purpose of this article is to provide an overview of clinical applications, technology, anatomy, and interpretation, and it includes material of interest to radiologists, cardiologists, physicians in training, and referring physicians. Physicians interpreting CCTA studies should be aware of the American College of Cardiology [9] and American College of Radiology [10] guidelines for training. [11]

Previous
Next:

Clinical Applications: Coronary Artery Disease

Calcium score

Calcium scoring and coronary computed tomography (CT) angiography (CCTA) have different clinical indications. Calcium scoring is primarily used for risk stratification of asymptomatic patients, while CCTA is primarily used in patients with acute or chronic chest pain. One potential use of performing a nonenhanced calcium scoring study before a CCTA is to decide whether to proceed with CCTA in patients with extensive coronary calcium. There is no established calcium score cutoff value above which CCTA will not be diagnostic, but a score of 1000 is often used.

In the multicenter Assessment by Coronary Computed Tomographic Angiography of Individuals Undergoing Invasive Coronary Angiography (ACCURACY) trial, [12] the specificity of CCTA was significantly reduced (from 86% to 53%) in patients with calcium scores greater than 400. However, in a meta-analysis of 51 studies [13] published in 2012, the accuracy of CCTA for significant stenoses was high, even in cases of severe coronary calcification, as long as 64-slice or new CT systems were used.

In symptomatic patients with low-to-intermediate pretest probability of coronary artery disease (CAD) (the categories for which CCTA has been endorsed), a negative coronary CTA shows a very high negative predictive value, independent of the coronary calcium score. In addition, the calcium score does not influence the result of the CCTA when it is considered positive for obstructive CAD. [14]

Coronary CTA

See the image below.

CT angiography and catheterization: Multiplanar re CT angiography and catheterization: Multiplanar reconstruction (MPR) CT image (A) demonstrates a severe proximal left anterior descending (LAD) artery stenosis (arrow), also demonstrated on cardiac catheterization (B) (arrow). Injection of the LAD after angioplasty and stenting (C) demonstrates no residual stenosis.

Indications

A 2008 scientific statement from the American Heart Association (AHA) [15] indicates that the potential benefit of noninvasive coronary angiography is likely to be the greatest for symptomatic patients who are at intermediate risk for coronary artery disease (CAD) after initial risk stratification, including patients with equivocal stress tests. CCTA is recommended over coronary magnetic resonance angiography (MRA) because of superior diagnostic accuracy. [16]

Neither coronary CTA nor magnetic resonance angiography (MRA) is recommended to screen for CAD in patients who have no signs or symptoms suggestive of CAD. In 7590 individuals without chest pain syndrome or history of CAD in the CONFIRM registry, the additional risk-predictive advantage of CCTA was not clinically meaningful compared with a risk model based on coronary calcium score alone. [17]

Results from the multicenter CONFIRM registry support that CCTA can be used effectively as a gatekeeper to invasive coronary angiography. [18]

Appropriateness criteria were published in 2010 from the combined efforts of 9 specialty societies. [1] The following indications (see the article for specific indications within the broad categories below) were rated as appropriate for CCTA:

  • Detection of CAD in symptomatic patients without known heart disease, either nonacute or acute presentations
  • Detection of CAD in patients with new onset or newly diagnosed clinical heart failure and no prior CAD
  • Preoperative coronary assessment prior to noncoronary cardiac surgery
  • Patients with prior ECG exercise testing - Normal test with continued symptoms or intermediate risk Duke treadmill score
  • Patients with prior stress imaging procedures - Discordant electrocardiographic (ECG) exercise and imaging results or equivocal stress imaging results
  • Evaluation of new or worsening symptoms in the setting of a past normal stress imaging study
  • Risk assessment post revascularization - Symptomatic if post coronary artery bypass grafting or asymptomatic with prior left main coronary stent greater than or equal to 3 mm
  • Evaluation of cardiac structure and function in adult congenital heart disease
  • Evaluation of cardiac structure and function - Ventricular morphology and systolic function
  • Evaluation of cardiac structure and function - Intracardiac and extracardiac structures

Emergency department use

In a multicenter trial [19] of 1000 patients presenting to the emergency department (ED) with symptoms suggestive of acute coronary syndromes, incorporation of CCTA into the triage strategy improved the efficiency of clinical decision making (significantly higher rates of discharge from the ED and reduced mean length of hospital stay) compared with standard evaluation. However, incorporation of CCTA resulted in an increase in downstream testing and radiation exposure with no decrease in overall costs of care. In a meta-analysis of 4 randomized controlled trials, [20] the use of CCTA in the ED was associated with decreased ED cost and length of stay but a 2% increase in invasive coronary angiography and revascularization (whether the increased rate of invasive procedures leads to improved patient outcomes is unknown).

Cost-effectiveness

Data as to the cost-effectiveness of CCTA are limited. Two analyses [21, 22] and one multicenter trial [23] suggest that CCTA may be a more cost-effective alternative to myocardial perfusion scintigraphy. CCTA may be a cost-effective method of avoiding unnecessary conventional coronary angiography in patients with a pretest probability of disease 50% or lower. [24]

Accuracy

The majority of studies (with the exception of the Coronary Evaluation Using Multidetector Spiral Computed Tomography Angiography using 64 Detectors [CORE 64] study) indicate that a negative CCTA can effectively rule out obstructive coronary artery disease. In a 2008 meta-analysis, [21] 64-slice CCTA had a sensitivity of 99% and negative predictive value (NPV) of 100% for patient-based detection of significant CAD. However the specificity has been lower than the sensitivity in most studies, and false-positive results are possible, particularly in patients with high calcium scores.

In the Assessment by Coronary Computed Tomographic Angiography of Individuals Undergoing Invasive Coronary Angiography (ACCURACY) prospective multicenter trial of patients with chest pain without known CAD and intermediate disease prevalence, 64-slice CCTA had a patient-based sensitivity of 94% and a specificity of 83% in detecting stenosis of 70% or greater (comparable values were seen at a 50% stenosis level). Patients with high calcium scores were not excluded from the study. Calcium scores greater than 400 reduced specificity significantly. The NPV of CCTA was 99%. [12]

In the CORE trial, 64 prospective multicenter trial of patients with suspected symptomatic CAD referred for conventional coronary angiography, 64-slice CCTA had a patient–based sensitivity of 85% and specificity of 90% (excluding patients with a calcium score greater than 600) for detecting stenoses 50% or greater. However, the NPV of 83% in this study was lower than in other studies. [25]

In a 2008 meta-analysis, the sensitivity was highest in the left main artery and lowest (85%) in the circumflex artery. [21]

In a systematic review that evaluated the diagnostic accuracy of CCTA for detecting cardiac allograft vasculopathy (CAV) compared with conventional coronary angiography (CCAG) alone or with intravascular ultrasound (IVUS), Wever-Pinzon et al found that CCTA had high sensitivity, specificity, and NPV for the detection of any CAV and significant CAV. [26] When 64-slice was compared with 16-slice CCTA, a trend toward improved sensitivity and NPV for identifying significant CAV was noted.

The ACCURACY trial suggested that, compared with other noninvasive modalities such as stress echocardiography and stress nuclear testing, CCTA has comparable specificity but superior sensitivity and NPV. [12]

Comparison with functional testing

In the Prospective Multicenter Imaging Study for Evaluation of Chest Pain (PROMISE) trial, [27]  a strategy of initial CCTA showed no mprovement in clinical outcomes over a median follow-up of 2 years compared with functional testing. The investigators randomized 10,003 symptomatic patients to initial testing with eiher CCTA or functional testing (exercise electrocardiography, nuclear stress testing, or stress echocardiography). Although more patients in the CCTA group than the functional testing group underwent catheterization within 90 days, those in CCTA group also had with fewer catheterizations showing no obstructive CAD. [27]

Previous
Next:

Clinical Applications: Coronary Artery Stents

Indications

In appropriateness criteria published in 2010 from 9 specialty societies, [1] the use of coronary computed tomography (CT) angiography (CCTA) in patients with a history of percutaneous revascularization with stents of 3 mm or greater and chest pain syndrome was rated as uncertain, and it was rated as appropriate in asymptomatic patients. A 2010 consensus statement [28] states that in patients known to have larger stents and whose clinical presentation suggests low-to-intermediate probability for restenosis, 64-slice CCTA may be a reasonable alternative to invasive angiography to rule out significant in-stent restenosis. See the image below.

Coronary artery stent: Axial CT image demonstrates Coronary artery stent: Axial CT image demonstrates a left anterior descending (LAD) artery stent. The stent lumen is well-visualized in this case.

Accuracy

Visualization of the stent lumen by CCTA is variable.

In a meta-analysis, the sensitivity and specificity of 64-slice CCTA for the detection of coronary in-stent restenosis (>50%) were 90% and 91%, respectively, not including nonassessable stents; 89% of stents were assessable. [29]

Stent size and material can affect evaluability by CCTA. Stents under 3 mm are often nonassessable. [30] Magnesium is the most favorable stent material for imaging. [31] Stainless-steel and cobalt stents are also favorable. [32] Other factors that can potentially limit stent evaluability include overlapping positioning, strut thickness, and large patient size.

Imaging pearls

See the list below:

  • Contrast opacification of the artery distal to the stent cannot be used as an indicator of stent patency, as collateral vessels may feed the vessel distal to an occluded stent.
  • Edge-enhancing filters can increase spatial resolution and decrease blooming artifact, and they are often helpful in evaluating coronary stents. However, the use of these filters increases image noise. Optimizing intraluminal contrast enhancement can counterbalance increased image noise. [33]
  • It is difficult to evaluate stents in large patients because of increased image noise and decreased vascular opacification. The upper limit of the patient’s body mass index (BMI) should be 35-40 kg/m 2. [34]
  • Wide window settings are helpful in evaluating the in-stent lumen. Pugliese et al recommend a window width of 1500 Hounsfield units (HU) and a window center of 300 HU. [33]
  • The positive predictive value of diagnosing in-stent restenosis can be low, particularly in small stents, where false-positive results are often seen. [33]
  • Left main stents are often evaluable due to their relatively large size.
  • Stent fractures can be accurately detected by CCTA. [35]
Previous
Next:

Clinical Applications: Coronary Bypass Grafts

Indications

According to appropriateness criteria published in 2010 from 9 specialty societies, the use of coronary computed tomography (CT) angiography (CCTA) in patients to evaluate bypass graft patency in patients with chest pain syndrome was rated as appropriate, and it was rated as inappropriate in asymptomatic patients less than 5 years after surgery, and it was rated uncertain in asymptomatic patients more than 5 years after surgery. [1] See the images below.

Saphenous vein graft: Volume-rendered CT image of Saphenous vein graft: Volume-rendered CT image of a saphenous vein graft to the right coronary artery.
Left internal mammary artery (LIMA) graft: Volume- Left internal mammary artery (LIMA) graft: Volume-rendered CT image of a LIMA to left anterior descending (LAD) artery graft.

In a 2007 consensus statement from the Society of Cardiovascular Computed Tomography and the North American Society for Cardiac Imaging, stress nuclear testing or echocardiography was suggested as the first method of evaluation rather than CCTA, as the question of graft patency is not as important as the functional significance of the grafts and bypass vessels. CCTA was suggested for reoperation bypass mapping of the previous bypass grafts, the setting of aortic dissection, difficult catheterization, or high risk of catheterization (eg, Marfan syndrome). [34]

Accuracy

Bypass grafts are well visualized by CCTA. In a meta-analysis, [36] 16- and 64-slice CCTA had a sensitivity of 98% and specificity of 97% for coronary artery bypass graft stenosis or occlusion; 92% of bypass grafts were evaluable. However, the accuracy for evaluating the coronary artery bypass graft (CABG) anastomoses [37] is less than that for other CABG segments.

Imaging pearls

See the list below:

  • Artifact from surgical clips can limit graft evaluation by CCTA. The distal anastomosis in particular can be difficult to assess due to motion and/or clip artifacts.
  • The most challenging aspect of CCTA in post-CABG patients is often evaluation of the native vessels. In one study, 29% of native vessel segments could not be evaluated in CABG patients. [37]
  • Proximity of the graft(s) to the sternum should be described to avoid damage during a redo sternotomy. If the graft abuts the sternum, the position of the graft relative to the sternal notch should be reported. [38]
  • In patients with internal mammary grafts, the scan should extend superiorly to the level of the clavicle so that subclavian stenosis can be ruled out.
Previous
Next:

Recent Technology

A full discussion of multidetector computed tomography (CT) technology is beyond the scope of this article, but is well addressed in other literature. [39, 40, 41] The 2 major recent advancements in multidetector CT technology are dual-source 64-slice CT and single-source 256- and 320-slice CT. Both of these techniques offer the possibility of reduced radiation dose compared with single-source 64-slice CT. Dual-source CT allows coronary CT angiography (CCTA) to be performed at higher heart rates. Currently, there is little published literature on 256- or 320-slice CCTA.

Dual-source CT

The primary advantage of dual-source CT is greater temporal resolution.

A dual-source CT contains 2 tube/detector sets, arranged at 90º angles to each other. In CCTA, the data are typically reconstructed from a 180º rotation (partial scan reconstruction) to maximize temporal resolution. If the gantry rotation time is 330 msec, a single-source CT performing CCTA with partial scan reconstruction has a temporal resolution of 165 msec. With 2 tubes, only a quarter rotation is needed for data collection, and the temporal resolution is 83 msec.

The higher temporal resolution of dual-source CT allows CCTA to be performed at higher heart rates without the use of beta blockers.

Although the tube current is doubled with 2 tubes, the scan time is halved, and the tube current-time product (mAs) is unchanged as compared with single-source CT. [42] However, radiation dose can usually be lower than that with single-source CT. Because of higher temporal resolution, the pitch can be increased at higher heart rates, which will decrease dose (see Radiation Dose). In addition, other dose-reduction techniques, such as ECG-dependent tube current modulation and prospectively triggered sequential scanning (see Radiation Dose), can be optimally used with the increased temporal resolution of a dual-source scanner.

In addition, simultaneous data acquisition can be performed with the tubes operating at different voltages (80 kV and 140 kV). [42] This offers the possibility of improved tissue differentiation, but it is unclear what impact this will have on CCTA.

In a meta-analysis of 25 studies, [43] dual-source CT had a per-patient sensitivity of 99% and specificity of 89%. Accuracy remained high in a systematic review of the use of dual-source CT in difficult-to-image patient groups. [44]

In an in vitro evaluation of in-stent lumen visibility of 27 commonly used modern coronary stents, Gassenmaier et al reported that  use of third-generation dual-source CTA enables stent lumen visibility of up to 80% in metal stents and 100% in bioresorbable stents. CTA may be a valid alternative for detection of in-stent restenosis. [45]

256-slice CT and 320-slice CT

The primary advantage of 256- and 320-slice CT is the increased craniocaudal coverage.

The 256- and 320-slice CT scanners have a craniocaudal coverage of approximately 12 and 16 cm, respectively. This potentially allows the heart to be scanned in one tube rotation and one heartbeat, without table movement. This technique is ideal for the prospectively triggered sequential scan technique, [46] which substantially reduces radiation dose compared with retrospectively gated helical techniques (see Radiation Dose). If the prospectively triggered technique is used, heart rate control is useful, as a slower heart rate allows a narrower phase window to be used, [47] further decreasing radiation dose. Another advantage is that if the data can be acquired in one heartbeat, phase misregistration artifacts arising from irregular heartbeats are not an issue.

In a comparison of prospectively gated 64- and 256-slice CT, 256-slice CT provided significantly improved and more stable image quality compared with 64-slice CT, at equivalent effective radiation dose. [2]

Previous
Next:

Radiation Dose

Attention has been drawn to the risk of cancer from computed tomography. In these discussions, coronary computed tomography (CT) angiography (CCTA) is often cited for its high radiation dose. For example, in a review by Smith-Bindman et al, [48] it was estimated that 1 in 270 women, and 1 in 600 men, who underwent CCTA at age 40 would develop cancer from that scan. However, these figures do not take into account recent technological developments that can greatly decrease the radiation dose from CCTA.

There are several things that are important to note. First, the radiation dose from CCTA is highly variable and is largely dependent upon the specific equipment and techniques used. For example, one study suggests that use of a prospectively triggered technique reduced the risk of cancer by 87%, compared to a retrospectively gated technique. [49] Therefore, it is very important for the practitioner to understand ways of reducing the radiation dose to the patient. Other examinations that might be used instead of CCTA, such as cardiac nuclear medicine studies, have relatively high radiation doses. In addition, noninvasive examinations such as CCTA have the potential to reduce the use of coronary catheterization, which is invasive and involves relatively high radiation doses. In a study of 398,978 patients who underwent elective coronary catheterization, [50] only 37.6% had obstructive coronary disease.

Examinations that do not utilize ionizing radiation, such as stress echocardiography, could also be considered. A decision analytic model suggests that stress echocardiography followed by CCTA if needed is most appropriate for evaluation of patients with a pretest probability for coronary artery disease of less than 20%, while CCTA alone is more appropriate for intermediate-risk patients. [51]

Radiation doses for CCTA studies, if performed with retrospective gating in helical mode, are typically relatively high. Radiation doses are high because data are acquired throughout the cardiac cycle, and with the fast gantry rotation required for high temporal resolution in CCTA, a low pitch (table travel per gantry rotation/collimation) is required in order to avoid gaps in the data. If data are acquired throughout the cardiac cycle, the table should not move more than the beam width during one cardiac cycle. In particular, slow movement is required with fast gantry rotation; otherwise, all phases of the heart at a specific location will not be seen by the detector. As pitch is inversely related to radiation dose, a low pitch results in a high radiation dose. [40]

The reported radiation doses for CCTA vary depending on the specific technology and techniques employed. With retrospectively gated single-source 64-slice CT, the reported effective radiation doses have ranged from 9.5 to 21.4 mSv. [3] However, using many of the technologies and techniques discussed below, it is possible to lower the dose to less than 5 mSv, and doses less than 1 mSv are currently possible in some patients. For comparison, the average yearly background radiation dose is around 3 mSv, and a chest x-ray dose is 0.05 mSv. [15] Depending on the technique, CCTA may have a higher or lower effective dose than conventional coronary angiography (3.1-9.4 mSv). However, the dose from a cardiac single-photon emission computed tomography (SPECT) scan performed using technetium-99m is typically high (8 -17.5 mSv).

A variety of methods exist for decreasing radiation dose from cardiac CT. [52] In general, radiation dose from a CT scan can be reduced by reducing tube current, reducing tube voltage, or increasing pitch.

ECG-dependent tube current modulation

Anatomic-based tube current modulation can be performed where the tube current is adjusted for patient size and shape. However, a more effective method of dose reduction is electrocardiograph (ECG)-dependent tube current modulation. As the best image quality for CCTA is typically obtained at the specific phase of the R-R interval (usually mid- to end-diastole), the tube current (and thus dose) can be reduced in the phases where image quality is not optimal. [52] This is the most common method to reduce radiation dose. The primary disadvantage of this technique is that optimal functional imaging is not possible because of poor image quality in the portions of the cardiac cycle where lower tube current was used. The use of ECG-dependent dose modulation can result in a 20-50% decrease in radiation dose. [3] This technique can be optimally used with dual-source CT, as the time requiring the highest tube current is shorter.

Slow and steady heart rates are necessary for effective ECG-dependent dose modulation. At high heart rates, the period of reduced tube output (diastolic duration) becomes shorter relative to the cardiac cycle. With irregular heart rates, the optimal time point in the cardiac cycle to apply the full tube current is less predictable.

Society of Cardiovascular Computed Tomography (SCCT) guidelines [53] recommend ECG-based, tube-current modification in retrospective gated studies, except in patients with highly irregular heart rhythms.

Reduced tube voltage

Reducing the tube voltage (from 120 kV to 100 or 80 kV) will reduce the radiation dose. Another advantage is that opacification of the blood vessels may increase at lower kV because of an increase in the photoelectric effect. [3] The primary disadvantage is increased image noise. SCCT guidelines [53] recommend a tube potential of 100 kV for patients weighing no more than 90 kg or with a body mass index of no more than 30 kg/m2. A tube potential of 120 kV is indicated for larger patients, and even higher tube potentials may be indicated for severely obese patients.

Increased pitch

With dual-source scanners, the pitch can be increased at high heart rates to reduce radiation dose.

Increasing the pitch will decrease the radiation dose, as the patient is exposed to radiation for a shorter period of time. The pitch can be increased at higher heart rates, because the time necessary to collect data throughout the cardiac cycle is decreased when the R-R interval is shorter.

However, single-source scanners typically do not allow pitch to be increased at high heart rates. Single-source scanners usually need to utilize multisegment reconstruction to increase temporal resolution at high heart rates, and multisegment reconstruction requires a low pitch. In this technique, the data required for image reconstruction are selected from multiple sequential heart cycles. This technique requires retrospective gating and a regular heart rate. [39] For data from several cardiac cycles to be used for image reconstruction, the same position has to be covered by the detector during consecutive cardiac cycles. Thus, the pitch must be lowered, which will increase radiation dose. Multisegment reconstruction is effective in improving temporal resolution only at specific heart rates (the heart rate and gantry rotation time need to be desynchronized).

A dual-source scanner has greater temporal resolution, and multisegment reconstruction is not necessary at high heart rates. This allows the pitch to be increased, and dose decreased, at higher heart rates.

Prospective ECG triggering and sequential scanning

Coronary CT angiography is usually performed with retrospective ECG gating, where scanning occurs throughout the cardiac cycle and simultaneously acquired ECG data are used retrospectively during image reconstruction. The acquisition of data throughout the cardiac cycle increases radiation dose. In addition, the scan is performed helically with a low pitch, resulting in substantial tissue overlap during scanning, as well as increasing radiation dose. [39]

With prospective ECG triggering, the data are acquired at a specific point in the R-R interval. The scanner acquires data sequentially ("step and shoot") rather than in helical mode. Radiation dose is decreased, as data are not acquired throughout the cardiac cycle, and there is minimal tissue overlap with a sequential scan technique. This technique is standard for coronary artery calcium scoring but can be used to reduce radiation dose substantially during CCTA. Using a prospectively triggered sequential scan technique, Earls et al achieved an 83% reduction in dose as compared to the retrospective gated technique. [3]

The primary disadvantage of this technique is the lack of functional data. In addition, as data are only available from predefined phases of the R-R interval, reconstructions from additional phases to improve image quality are not possible.

This technique is optimal in patients with a low and stable heart rate. [54, 55] High heart rates are not optimal for this technique, as reconstructions at multiple phases during the R-R cycle are sometimes needed. With irregular heart rates, the acquisitions may be triggered at different points in the R-R interval.

Prospective ECG triggering is optimal for 256- or 320-slice CT, where the entire heart could potentially be scanned in one tube rotation and one heartbeat. This obviates the issue of phase misregistration in patients with irregular heart rates. In one study, the median effective radiation dose of 320-slice CT was 4.2 mSv, [56] which was lower than an 8.5 mSv median dose from catheter angiography performed in the same patients. Prospective ECG triggering is also well suited for use with dual-source CT, as the increased temporal resolution may allow the technique to be used at a higher heart rate threshold. [54]

In a meta-analysis of 20 studies [57] in patients with coronary artery disease (CAD) and without tachyarrhythmia, prospectively triggered CCTA provided image quality and diagnostic accuracy comparable to retrospectively gated CTA, but at a much lower radiation dose (3.5 mSv average compared with 12.3 mSv).

SCCT guidelines [53] state that prospective ECG triggering should be used in patients who have stable sinus rhythm and low heart rates (typically < 60-65 beats per minute). The width of the data acquisition window should be kept to a minimum. Retrospective gating is recommended for patients who do not qualify for prospective scanning due to irregular rhythms or high heart rates.

Iterative reconstruction

Iterative reconstruction is a new CT reconstruction technique that reduces image noise, which then allows radiation dose to be decreased. Interactive reconstruction has been shown to both reduce image noise and improve image quality. [58]

How low can radiation dose be?

In one study, 320-slice CT, prospective gating, iterative reconstruction and automated exposure control were used in conjunction with lower radiation dose in 107 consecutive patients. [59] The dosage was less than 4 mSv for 96% of the patients and less than 1 mSv for 54% of the patients. For comparison, the average yearly background radiation dose is around 3 mSv.

Previous
Next:

Patient Preparation

At our institution, patients are instructed to avoid caffeine and smoking 12 hours prior to the procedure to avoid cardiac stimulation. They are also instructed to avoid eating solid food 4 hours before the study and to increase fluid intake prior to the study. Standard precautions with regard to contrast allergy history and renal function are taken.

Beta blockers

Beta-blocker administration is often helpful in cardiac computed tomography (CT) to lower the heart rate and decrease motion artifact. The level to which the heart rate should be lowered depends on the temporal resolution of the scan. With single-source CT scanners, it is usually helpful to lower heart rate below 65 beats per minute (bpm), and ideally below 60 bpm. Dual-source CT scanners have higher temporal resolution and can be performed at heart rates of up to 90 bpm, obviating the need for beta blockers in many cases. Cardiac MRI has higher temporal resolution than CT and can be performed without beta blockers.

However, heart rate variability may be a more important determinant of image quality than absolute heart rate. [4] Beta blockers are also helpful in patients with irregular heart rates, supraventricular tachycardias, and arrhythmias. For example, in atrial fibrillation, the negative chronotropic and dromotropic effects can lengthen diastole. [60]

Possible contraindications to beta-blocker administration include the following [6, 60] :

  • Heart rate slower than 60 bpm
  • Systolic blood pressure lower than 100 mm Hg
  • Asthma or chronic obstructive pulmonary disease (COPD) on beta 2 -agonist inhaler
  • Active bronchospasm
  • Second- or third-degree atrioventricular block
  • Sick sinus syndrome
  • Decompensated cardiac failure
  • Pheochromocytoma (can be given in combination with an alpha blocker if the alpha blocker has been initiated beforehand)
  • Allergy to beta blockers

Beta blockers should be used with caution in patients with severe peripheral vascular disease and in patients taking calcium channel blockers.

Metoprolol is a frequently used beta blocker for coronary CT angiography (CCTA). The effects of an oral dose are seen within 1 hour after administration, with peak plasma concentration at 90 minutes. [60] IV-push metoprolol has a peak effect 5-10 minutes after administration.

There are many different protocols for metoprolol administration. An oral dose of 50-100 mg can be administered 60-90 minutes before the study. [60, 61] If this does not lower the heart rate to the desired level, 5-mg doses of IV metoprolol can be administered at 3- to 5-minute intervals, up to a total dose of 15-30 mg. [60, 61]

Atenolol [61] and esmolol [62] have also been successfully used.

Diltiazem or verapamil can also be used in patients in whom beta blockade is contraindicated, although these are less effective and result in more hypotension. [61] These drugs can also be used in combination with a very low dose of metoprolol.

Nitroglycerin

The administration of sublingual nitroglycerin dilates the coronary arteries and increases side branch visualization. [5] Nitroglycerin is contraindicated in patients who are allergic to it and in patients who are taking phosphodiesterase inhibitors for erectile dysfunction. [6] Patients should not have taken a phosphodiesterase inhibitor for at least 48 hours before the exam. The concomitant use of phosphodiesterase inhibitors can cause severe hypotension. Nitroglycerin can cause orthostatic hypotension and should be used with caution in patients who have low systolic blood pressure (eg, <90 mm Hg) and who are volume depleted from diuretic therapy. Angina caused by hypertrophic cardiomyopathy can also be aggravated.

Previous
Next:

Artifacts

Stairstep artifacts

Stairstep artifacts are associated with heart rate variability. [4] With irregular heart rates, phase misregistration can occur when data from different cardiac phases are used for reconstruction. A stairstep appearance results from the data reconstructed from different cardiac phases.

Stairstep artifact: Volume-rendered CT image demon Stairstep artifact: Volume-rendered CT image demonstrates "stairstep" artifact. This is a phase-misregistration artifact that is typically secondary to an irregular heartbeat.

Beta blockers are helpful in reducing heart rate variability and avoiding stairstep artifacts. Manual electrocardiographic (ECG) editing can also be helpful. With 256- and 320-slice computed tomography (CT) scanning, stairstep artifacts should not be seen if the heart is scanned in one heartbeat.

Coronary artery motion artifacts

Artifacts from motion of the coronary arteries result in image blurring. The right coronary artery is often most affected by motion artifact.

Coronary artery motion: Axial CT images reconstruc Coronary artery motion: Axial CT images reconstructed at 90% (A) and 70% (B) of the R-R interval demonstrate the importance of imaging during the phase of least cardiac motion.

General strategies to decrease motion artifact are to increase the time during the cardiac cycle where there is the least motion and to image as quickly as possible (increase temporal resolution).

Motion can be minimized by reconstructing the data during a phase where there is minimal motion. Choosing the optimal phase of the R-R cycle to reconstruct the data is discussed in the image reconstruction section, below. Decreasing the heart rate with beta blockers has the advantages of decreasing the motion velocity of the coronary arteries [4] and increasing the relative and absolute duration of the diastolic rest period in the cardiac cycle.

Temporal resolution can be increased in 2 ways. Dual-source CT scanners have substantially higher temporal resolution. With a single-source scanner, one way to increase temporal resolution, typically used in patients with higher heart rates, is to use a multiple-segment reconstruction technique. In this technique, the data required for image reconstruction are selected from multiple sequential heart cycles. This technique requires retrospective gating and a regular heart rate. [4] For data from several cardiac cycles to be used for image reconstruction, the same position has to be covered by the detector during consecutive cardiac cycles. Thus, the pitch must be lowered, which will increase radiation dose. Multi-segment reconstruction is only effective in improving temporal resolution at specific heart rates (the heart rate and gantry rotation time need to be desynchronized).

Arrhythmias

Arrhythmias present a challenge for CCTA because of both high and irregular heart rates, and both stairstep and motion artifacts can be seen. Atrial fibrillation has sometimes been considered a relative contraindication to the performance of CCTA. However, in recent studies, CCTA has been successfully performed in patients with atrial fibrillation by using dual-source CT and end-systolic reconstruction and by using single-source 64-slice CT with ECG-editing [63] and middiastolic reconstruction. [64]

Respiratory motion artifacts

Most patients can breath-hold for the time necessary to complete a CCTA study. A Valsalva maneuver should be avoided, as this can decrease inflow into the right atrium and decrease contrast enhancement. [4] Respiratory motion artifact can be recognized on the lung windows and is most prominent on coronal and sagittal images.

Streak artifacts

Streak artifacts from beam hardening can be seen secondary to metal clips. Streak artifact in the superior vena cava and right atrium from dense contrast can limit evaluation of the right coronary artery. This can be mitigated by the use of a saline bolus chaser. However, a saline bolus chaser can result in poor contrast opacification of the right heart lumen, which may limit morphologic and functional evaluation. Protocols that utilize an admixture of saline and contrast are helpful in maintaining right heart opacification without streak artifact.

Blooming artifacts

Blooming artifacts can cause small high-contrast structures such as stents and calcium to appear larger than they are. [4] Edge-enhancing kernel filters can decrease blooming artifacts and may be helpful for evaluating a stent lumen, although image noise is increased.

Blooming artifact: Multiplanar reconstruction (MPR Blooming artifact: Multiplanar reconstruction (MPR) CT image demonstrates blooming artifact from a left anterior descending (LAD) artery stent. The stent lumen is poorly visualized secondary to this artifact.
Previous
Next:

Image Reconstruction

Reconstruction Window

The coronary arteries are optimally imaged when there is the least cardiac motion. This occurs during so-called rest periods, which is typically in middiastole (diastasis). Coronary motion is also minimal during end-systole (isovolumic relaxation), but this is of shorter duration than diastolic diastasis at low heart rates.

With dual-source computed tomography (CT) scanning at low heart rates, the optimal reconstruction window is often 70–75% of the R-R interval (diastolic) for all of the coronary arteries. A 30-35% systolic window may occasionally be helpful for the right coronary artery. [65] As heart rate increases, diastole shortens relative to systole, and diastasis shortens dramatically. The optimal reconstruction window transitions from diastole to systole around 75-85 bpm. [4, 66] At high heart rates, the optimal reconstruction windows are 85-90% (diastole) and 40-45% (systole). [66] End-systolic reconstruction windows [63] may be helpful in patients with atrial fibrillation, as the systolic rest period will be less variable than the diastolic rest period.

Use of a fixed reconstruction window, [63] such as 50 msec rather than a percentage of the R-R interval, is helpful in patients with atrial fibrillation and variable R-R intervals.

Postprocessing Techniques

A variety of postprocessing techniques are useful in coronary CT angiography (CCTA). [7] Many interpreting physicians will start with the axial source images and then utilize multiplanar reconstructions in at least 2 planes.

Axial source images

Axial source images are often the initial images used to review the coronary arteries and are used to evaluate the extracardiac structures.

Multiplanar reconstruction (MPR)

MPRs can be performed at oblique planes to the body or the coronary arteries. For coronary artery imaging, a curved MPR technique is usually used where the reconstruction plane is locked onto the target vessel. This requires a manual or automatic centerline to be drawn along the vessel. Curved MPR images can often be difficult to obtain if the centerline is difficult to trace, for reasons such as motion artifact, poor contrast opacification, or dense calcifications. The curved MPR can be unfolded so that the vessel appears to be straight ("ribbon view"). Note that a single curved MPR may not adequately display eccentric lesions, and correlation with orthogonal MPR views, such as an end-on view perpendicular to the vessel, is necessary.

Curved MPR: Curved multiplanar reconstruction (MPR Curved MPR: Curved multiplanar reconstruction (MPR) image of a left anterior descending (LAD) artery stenosis (arrow).
"Ribbon" multiplanar reconstruction (MPR): "Ribbon "Ribbon" multiplanar reconstruction (MPR): "Ribbon" MPR image of a left anterior descending (LAD) artery stenosis (arrow).

Maximum intensity projection (MIP)

With the MIP technique, the highest voxel attenuation values from a volume of CT data are used to reconstruct the image. The MIP technique can be used to create "angiographic" images. However, as voxels with lower attenuation values are suppressed, noncalcified plaques can be masked by luminal contrast, and calcified plaque can mask less dense luminal contrast. [7] The MIP technique tends to overestimate stenosis.

Maximum intensity projection (MIP): "Angiographic" Maximum intensity projection (MIP): "Angiographic" MIP image of the coronary arteries.

3D volume rendering

Volume-rendered images are visually appealing, but they usually play little role in primary interpretation. They are helpful for visualizing anomalous vessels and bypass grafts. Generation of volume-rendered images is computationally intensive and often requires manual editing.

Volume-rendering: Volume-rendered image of the hea Volume-rendering: Volume-rendered image of the heart demonstrates both the right and left coronary arteries as well as the cardiac chambers.
Previous
Next:

Anatomy

The left anterior descending artery (LAD) and posterior descending artery (PDA) run in the interventricular groove, while the circumflex and right coronary arteries (RCA) run in the atrioventricular groove. [67, 68, 69]

In general, hypoplasia of one vessel will be accompanied by prominence of another vessel. For example, if the LAD is small and does not extend to the apex, the PDA is typically prominent and extends to the apex. If the circumflex is small, there are typically prominent posterolateral branches arising from the RCA. If there is a large ramus intermedius, the diagonals may be small. [67, 68, 69]

Dominance

Right dominant (80-85%): Both the PDA and the posterolateral (also called posterior ventricular) branches arise from the RCA.

Right coronary artery (RCA): Volume-rendered CT im Right coronary artery (RCA): Volume-rendered CT image of the RCA demonstrates a right-dominant RCA with acute marginal, posterior descending artery, and posterior ventricular branches.

Left dominant (15-20%): Both the PDA and the posterolateral branches arise from the circumflex.

Left-dominant system: Volume-rendered CT image dem Left-dominant system: Volume-rendered CT image demonstrates a left-dominant system with the posterior descending artery (PDA) (arrowhead) and a posterolateral branch (white arrow) arising from the circumflex artery (black arrow).

Codominant (5%): The PDA arises from the RCA. The posterolateral branches arise from the circumflex.

Codominant system: Volume-rendered CT image demons Codominant system: Volume-rendered CT image demonstrates a codominant system with the posterior descending artery (arrow) arising from the right coronary artery (RCA) and a posterior ventricular branch (arrowhead) arising from the circumflex.

Left main coronary artery

The left main coronary artery is variable in length, 11 mm ± 5 mm. [70] If intervention in the left coronary artery system is a possibility, it may be helpful to report the length of the left main artery. The left main coronary artery will usually bifurcate into the LAD and the left circumflex artery. In approximately 30% of cases, the left main artery will trifurcate, with a ramus intermedius artery between the LAD and circumflex. The ramus intermedius artery will supply a lateral wall territory between the first diagonal and the first obtuse marginal branch territories.

Rarely, the left main coronary artery will be absent and the LAD and circumflex artery will arise directly from the aorta.

Left coronary arteries: Volume-rendered (A) and mu Left coronary arteries: Volume-rendered (A) and multiplanar reconstruction (MPR) (B) CT images in 2 patients demonstrate the left anterior descending (LAD) artery (1), first septal perforator (2), first diagonal (3), ramus intermedius (4), circumflex artery (5), and first obtuse marginal (6).

Left anterior descending artery

The LAD is variable in length and can terminate before the apex, supply the apex, or supply the distal inferior wall.

The LAD gives rise to diagonal branches supplying the anterolateral wall and to septal perforators supplying the interventricular septum. Compared with the diagonals, the septal perforators usually are less implicated in ischemia and are less often targets of intervention.

In 1% of cases, there can be dual LADs. [71] In these cases, a short LAD terminates high in the interventricular groove. A long LAD originates as an early branch from the LAD proper or, less likely, from the RCA. The proximal LAD courses outside the interventricular groove, with the distal portion returning to the groove. This should be distinguished from a diagonal branch; the diagonal will not enter the interventricular groove distally.

Left circumflex artery

The left circumflex artery gives off obtuse marginal branches that supply the posterolateral wall. In many cases, particularly in a right-dominant system, the first obtuse marginal branch will be larger than the circumflex. The circumflex artery is the vessel that remains in the atrioventricular groove.

The left circumflex artery usually terminates in the atrioventricular groove. In a left dominant system, the PDA and posterolateral branches arise from the circumflex. In the codominant system, the posterolateral branches arise from the circumflex.

Right coronary artery

The branches from the RCA are called acute marginal or right ventricular marginal arteries. However, the acute marginal branch is also used to refer to the largest marginal artery, which arises at the inferior aspect of the right border of the heart, coursing towards the apex. The marginal arteries usually do not cause significant ischemia and are rarely targets for intervention.

At the inferior aspect of the right atrioventricular groove, the RCA, if dominant, bifurcates into the PDA and posterolateral left ventricular branch. The PDA runs along the undersurface of the heart in the posterior interventricular groove and gives rise to septal perforators supplying the posterior third of the interventricular septum. The posterolateral branch has a hairpin curve (first cephalad, then caudad), and supplies the posterior and lateral left ventricular walls.

There are 3 "named" arteries that can arise from the RCA, as follows:

  • The first branch off the RCA is often the conus artery. In 50% of cases, the conus artery arises from the RCA; in the other 50%, it arises from the aorta. The conus artery heads anteriorly toward the conus (right ventricular outflow tract). (See the image below.)
    Conus artery: Axial maximum intensity projection ( Conus artery: Axial maximum intensity projection (MIP) image demonstrates the conus artery (arrow) arising from the right coronary artery (RCA).
  • The second branch off the RCA is often the sinoatrial (SA) nodal artery. In 55% of cases, the SA nodal artery arises from the RCA; in the other 45%, it arises from the circumflex. The SA nodal artery heads posteriorly toward the sinoatrial node. The sinoatrial node is located in the superior aspect of crista terminalis of the right atrium (near where the superior vena cava joins the right atrium). The crista terminalis is a vestigial remnant located between the right atrial appendage and the sinus venosus. The crista terminalis can be seen as a right atrial "pseudomass" on computed tomography (CT) scanning, magnetic resonance imaging (MRI), or echocardiography. (See the images below.)
    Sinoatrial node artery: Axial maximum intensity pr Sinoatrial node artery: Axial maximum intensity projection (MIP) image demonstrates the sinoatrial node artery (arrow) arising from the right coronary artery (RCA).
    Conus and sinoatrial node arteries: Volume-rendere Conus and sinoatrial node arteries: Volume-rendered CT image of the right coronary artery demonstrates the conus artery (arrow) heading anteriorly and the sinoatrial node artery (arrowhead) heading posteriorly.
    Crista terminalis: Axial CT image demonstrates the Crista terminalis: Axial CT image demonstrates the crista terminalis (arrow) in the superior right atrium.
  • The last named branch off the RCA is the atrioventricular (AV) nodal artery, which arises from the RCA in 80-87% of cases. It can also arise from the circumflex or from both the RCA and the circumflex. If it originates from the RCA, it typically arises from the proximal posterolateral branch. It heads superiorly through the septum to the AV node, which is in the inferior aspect of the interatrial septum.

Collateral pathways

Collateral pathways are typically better visualized on invasive coronary angiography than on CCTA. There is a large number of potential collateral pathways.( [72] )

Two "named" collateral pathways are the following:

  • Kugel's artery [73] : Connects an anterior artery (typically the circumflex) with a posterior artery (typically the RCA) at the crux (junction of the posterior interventricular and AV grooves) of the heart. This may connect with the AV nodal artery and supply the AV node.
  • Arc of Vieussens'/Vieussens' ring [74] : Connects the RCA (typically via the conus artery) with the LAD.

Anomalies

Coronary artery anomalies can be broadly classified as anomalies of origin, anomalies of course, and anomalies of termination. [75]

In anomalous cases, the coronary arteries should be identified by their location rather than by their origin or specific branches. The right coronary artery lies in the right atrioventricular groove and supplies the right ventricular free wall. The LAD lies in the anterior interventricular groove and supplies the anterior interventricular septum (the LAD need not give rise to diagonal branches). The left circumflex artery lies in the left atrioventricular groove and supplies the left ventricular free wall.

Most anomalies are incidental findings but may be important during surgical planning to avoid accidental vascular injury. Only a few anomalies are potentially malignant, with the potential to result in ischemia, infarction, or sudden death. These include origin of the artery from the opposite coronary sinus with interarterial course, pulmonary artery origin, and coronary artery fistulae. Patients with anomalous coronary artery origin from the pulmonary arteries show symptoms in infancy and early childhood.

The left and right coronary arteries can arise from the noncoronary sinus or the opposite sinus. In these cases, the arteries can take 4 courses: retroaortic, prepulmonic, septal (beneath the right ventricular outflow tract), or interarterial (between the aorta and pulmonary artery). Patients with an interarterial course, particularly of the left coronary artery, are at risk for ischemia, infarction, and sudden cardiac death, particularly during exercise. (See the images below.)

Anomalous right coronary artery (RCA): Axial CT im Anomalous right coronary artery (RCA): Axial CT image of an anomalous RCA arising from the left coronary sinus, with an interarterial course between the aorta and pulmonary artery.
Anomalous left circumflex: Axial CT image of an an Anomalous left circumflex: Axial CT image of an anomalous circumflex artery (arrow) passing posterior to the aorta.

Myocardial bridging, [76] also called tunneled artery, is a congenital anomaly where myocardium encases a segment of coronary artery. It is most common in the mid-LAD. The artery may be compressed in the systolic phase. Although it is usually a benign anomaly, it has been associated with myocardial ischemia. Myocardial bridging is well demonstrated by CCTA. As most diagnostic images are obtained in diastole, it is important to also review systolic images, if available, to evaluate for systolic compression. Atherosclerotic changes are more common proximal to the tunneled artery. (See the image below.)

Myocardial bridging: Maximum intensity projection Myocardial bridging: Maximum intensity projection (MIP) image demonstrates a myocardial bridge (arrow) involving the mid-left anterior descending artery (LAD).

Myocardial loops refer to muscle bundles from the atrial myocardium surrounding three quarters of the circumference of an artery. These are of no clinical significance.

Coronary artery fistulas are usually congenital and can be symptomatic if large. They are well visualized by CCTA. Coronary artery fistulas originate from the RCA in two thirds of cases and the left coronary system in a quarter of cases. More than 90% drain into the right atrium, coronary sinus, or right ventricle. [77] On CCTA, contrast opacification of the receiving chamber/vessel (shunt sign) [77] is useful for determining the exact site of entry of the fistula. However, this finding will be obscured if there is a significant amount of preexisting contrast in the receiving chamber/vessel. (See the image below.)

Coronary artery fistula: Volume-rendered CT image Coronary artery fistula: Volume-rendered CT image (A) demonstrates a fistula from the left anterior descending (LAD) artery (arrow) to the right ventricular outflow tract (RVOT) (arrowhead). Coronal CT image (B) demonstrates a small jet of contrast in the region of the shunt entrance (arrow) into the RVOT.

Other anomalies are not hemodynamically significant but important to describe in detail if intervention is a possibility. For example, a dual LAD can result in diagnostic error during cardiac catheterization or in technical difficulty during revascularization.

Previous
Next:

Stenosis Grading

There are many different methods to grade the degree of stenosis, including visual assessment; manually determined diameter or cross-sectional area on multiplanar reformats perpendicular to the median centerline of the vessel ("end-on" view); diameter on maximum intensity projection (MIP) images parallel to the long axis of the vessel; and software calculation [8] of diameter or area. Dodd et al found that the cross-sectional area technique had the highest correlation with quantitative coronary angiography, and MIP technique had the smallest interobserver variability. [78] Grading is less accurate in calcified plaques and in distal coronary vessels. In one report the most common etiologies of diagreement between computed tomography angiography (CTA) and catheter angiography were motion-related degradation of image quality, calcification, smaller vessel diameter, left anterior descending (LAD) artery location, and bifurcation location. [79]

Maximum intensity projection (MIP): MIP image of a Maximum intensity projection (MIP): MIP image of a left anterior descending (LAD) artery stenosis secondary to calcified plaque. MIP images parallel to the long axis of the vessel can be used to assess the degree of stenosis.
"End-on" multiplanar reformat (MPR): "End-on" MPR "End-on" multiplanar reformat (MPR): "End-on" MPR view (perpendicular to the median centerline of the vessel) of a left anterior descending (LAD) artery stenosis secondary to calcified plaque. The degree of stenosis could be assessed on this view visually, or from manually or software-calculated diameter or cross-sectional area.

Cross-sectional images at the level of the most severe narrowing can be compared to a reference minimal lumen diameter averaging the segments proximal and distal to the stenosis. The diameter should be measured lumen to lumen rather than wall to wall. The distal reference vessel should not be distal to a bifurcation.

Because the spatial resolution is inadequate for precise grading, coronary stenoses are often graded with semiquantitative descriptors such as normal, mild (< 50%), moderate (50–70% stenosis), severe (>70% stenosis), and occluded.

Stenosis is typically overestimated in areas where heavily calcified plaques are present. In one study, when calcification was greater than 50% of the luminal diameter, there was a signficant decrease in specificity, positive predictive value, and accuracy of CTA for evaluating coronary stenosis. [80] In the presence of extensive calcification, reconstruction of a additional dataset using a sharper convolution kernel (as used for stents) and use of bone window setting can reduce blooming artifacts from calcification. [81] Zhang et al offer the following suggestions to better assess the degree of stenosis when calcified plaques are present [82] :

  • A significant luminal stenosis is unlikely if the plaque thickness measures 50% or less of the diameter of a nearby normal segment and if it is eccentrically positioned on a cross-sectional multiplanar reconstruction (MPR) view or there is visible lumen adjacent to the plaque on a long-axis MPR view, .
  • A significant stenosis is likely if calcified plaque fills the entire central portion of the lumen on a cross-sectional MPR image.
  • A significant stenosis can be suggested if calcified plaque is 50% or greater than the diameter of a nearby normal segment on cross-sectional MPR images but does not completely fill the lumen; however, the interpreter might add that coronary computed tomography (CT) angiography (CCTA) may overestimate the degree of stenosis in this situation.

Pitfalls  [81, 83]

Several areas may be difficult to evaluate due to curvature, and additional review of these regions using thick-slap MIP may be helpful, as follows:

  • Distal segment of the RCA and origin of the PDA
  • Origin of the first diagonal branch
  • Distal circumflex near the origin of the obtuse marginal

A stenosis should always have an associated visible plaque, calcified or noncalcified. This is helpful in differentiating stenosis from artifactual apparent narrowing.

Previous
Next:

Reporting

A consensus report from 2 cardiac Computed tomography (CT) specialty societies suggests a reporting template for coronary (CT) angiography (CCTA). [34] A structured approach to reporting can also be used. [38]

Reporting template

The reporting template includes the following:

Indication for examination

Imaging technique:

  • Contrast agent administered
  • CT technique
  • Vasodilator or beta blocker
  • Workstation methods for image reconstruction
  • Complications

Description of findings:

  • Overall description of image quality/diagnostic confidence
  • Source of limitations, such as calcification and motion

Coronary anatomy/anomalies:

  • Anomalies of coronary origin
  • Right or left dominant system
  • Location and size of coronary artery aneurysm/dilatation

Coronary artery atherosclerosis:

  • Calcium score (if performed)
  • Description of atherosclerotic narrowing for vessels 2 mm or less in diameter
  • Location of atherosclerotic narrowing by anatomic landmarks, or a 15-segment model used for conventional angiography
  • Diffuse or focal disease description
  • Description of plaque, such as noncalcified, mixed, or calcified

Ventricular size and function, when requested and available

Extracardiac findings

Summary/impression and recommendation

Previous