Article

Protocol Fundamentals for Coronary Computed Tomography Angiography

Abstract

Citation:US Cardiology 2005;2(1):119-21

Open access:

The copyright in this work belongs to Radcliffe Medical Media. Only articles clearly marked with the CC BY-NC logo are published with the Creative Commons by Attribution Licence. The CC BY-NC option was not available for Radcliffe journals before 1 January 2019. Articles marked ‘Open Access’ but not marked ‘CC BY-NC’ are made freely accessible at the time of publication but are subject to standard copyright law regarding reproduction and distribution. Permission is required for reuse of this content.

Most advances in cardiac computed tomography (CT), particularly coronary CT angiography (CTA), have come from the development of protocols consistent with rapid incremental improvements in CT technology. The evolution of cardiac CT from early technology has been paralleled with evolving protocols that have extended cardiac CT beyond the imaging of coronary arteries alone: current applications include the assessment of coronary bypass grafts, cardiac valve analysis, cardiac function, and chest pain imaging. While the details of these more advanced protocols are beyond the scope of this article, all advanced cardiac CT protocols are founded on basic CTA. This article introduces coronary CTA by breaking down the examination into its core components.

Building a CT Protocol

Cardiac motion separates cardiac and coronary CT from the CT assessment of other body parts. Ultimately, successful coronary imaging by any modality relies on the ability of the hardware (for CT, the scanner) to produce motion-free images, or to scan faster than the heart beats. Thus, coronary CT relies on faster imaging or slowing cardiac motion. The imaging speed is measured by the temporal resolution and is determined by the CT gantry rotation time, defined as the time required for the CT gantry to make one full revolution. The temporal resolution is half the gantry rotation time, owing to the fact that image reconstruction requires CT data acquired from approximately one-half of a gantry rotation. Among commercially available scanners, gantry rotation times are now as low as 330 milliseconds, yielding a temporal resolution as low as 165 milliseconds. (Note that spatial resolution, described below, refers to the image voxel size and is independent of temporal resolution.) Manufacturers have steadily developed faster CT gantries, but even with these advances, imaging during 165 milliseconds of the cardiac cycle yields motion artifact for a significant fraction of patients. Thus, using current technology, heart rate control remains a critical component of the examination.

Beta-blockade for Heart Rate Control

In patients who do not routinely take beta-blockers, administration of metoprolol at the time of scanning is essential. One rule of thumb for the target heart rate is 'the first number is a five', i.e. an ideal heart rate between 50 and 59 beats per minute. While this goal is not achieved in every patient, it provides a useful reference frame. With cardiac monitoring, intravenous (IV) metoprolol is routinely and safely administered by both cardiologists and radiologists—5-mg increments given every five minutes to a total dose between 15mg and 25mg. Routine IV delivery has supplanted oral administration, which has the disadvantages of a longer serum half life and the fact that premedication requires patient compliance before reaching the examination site.

ECG Gating

Despite the importance of improved temporal resolution and heart rate control, it is ECG gating that enables coronary CTA. ECG gating refers to the simultaneous acquisition of both the patient's ECG tracing and the CT data (see Figure 1). By acquiring both pieces of information, CT images can be reconstructed using only a short temporal segment periodically located in the same location of the R-wave to R-wave (R-R) interval over multiple cardiac cycles. Each temporal segment of the R-R interval is named by its 'phase' in the cardiac cycle; the typical nomenclature is to name the percentage of a specific phase with respect to its position in the R-R interval. The number of phases is manufacturer-dependent. For example, if a manufacturer enables reconstruction of 20 (equally spaced) phases, they would typically be named 0%, 5%, 10%, up to 95%, beginning with one R-wave and ending with the following R-wave. The period in which the heart has the least motion is usually (but not always) in diastole, near a phase between 55% and 75%.Thus, under the assumption that the position of the heart remains consistent over the R-R intervals during which CT data is acquired, cardiac motion is minimized by producing images from the same phase over multiple cardiac cycles. This explains why ECG gating typically fails to freeze cardiac motion patients with an irregular rhythm such as atrial fibrillation (AF). Consequently, AF patients rarely have CTA examinations that are diagnostic over all coronary segments.

If only static coronary images are desired, reconstruction can be performed over a small number of phases in which motion is minimized. Motion of the left and right coronary arterial system differs in most patients, with right coronary artery (RCA) motion being greater. Thus, the phase of the cardiac cycle that proves best for diagnosis of the left main and left anterior descending (LAD) is frequently different to the phase that proves most diagnostic for the RCA. It is often necessary to view more than one phase to best assess the full extent of an individual artery and its branches (e.g. the LAD and the extent of its diagonal branches). Moreover, in some clinical settings (e.g the evaluation for coronary anomalies), only static images are required. In these clinical settings, image reconstruction will include roughly five phases during diastole, e.g. 50%, 55%, 60%, 65%, and 70%. Data from these phases will provide adequate visualization of the coronary ostea. However, it is important to re-emphasize that image data is acquired throughout the cardiac cycle. Thus, the CT data (and the radiation used to acquire that data) in the remaining phases (in this example 0% to 45% and 75% to 95%) is wasted.

Current modulation represents one strategy to lower the overall patient radiation by modulating the tube current (expressed as the mA) over the course of the cardiac cycle (see Figure 1c) so that the desired diagnostic tube current is delivered in diastole while the current is reduced for the remainder of the cardiac cycle. While featured on newer generations of CT scanners, there are potentially significant drawbacks. First, once current modulation is used, images subsequently reconstructed during phases with low tube current will be noisy. This is important because there is a subset of patients for whom the most diagnostic images, particularly for evaluation of the right coronary artery, is in systole. Second, current modulation eliminates the potential to reconstruct cine imaging, that is, reconstruction of all phases in the cardiac cycle displayed in a continuous loop. While cardiac CT cine loops are technically inferior to similar magnetic resonance (MR) acquisitions, global function can be routinely computed, and regional wall motion abnormalities seen. This can be particularly important in patients with a contraindication to MR, such as patients with a pacemaker.

ECG gating requires image data 'oversampling' because, for the reconstruction of each interval, only a small portion of the cardiac cycle is used. The CT pitch (a unitless parameter) is most accurately characterized as the distance that the patient moves through the scanner in a single gantry rotation divided by the width of the X-ray beam used. Because such a small part of the R-R interval is used to reconstruct an entire image, significant overlap along the craniocaudal extent of the patient is required, translating into a pitch between 0.2 and 0.35, or an oversampling rate of between 5:1 and roughly 3:1.The only other application for which ECG gating is routinely required is CTA of the thoracic aorta, where motion extending from the aortic valves to the arch can limit diagnostic accuracy. The fundamental difference between thoracic CT aortography and coronary CTA is the need for superior spatial resolution, as described below.

The practical consequence of oversampling is that the cardiac scan time (craniocaudal imaging over approximately 15cm) is far greater than non-gated scanning of the same Z-axis region of any other body part. For this reason, one great benefit of cardiac scanners equipped with a larger number of detectors is the ability to cover a larger craniocaudal territory per rotation. For example, while a four-slice coronary CTA may require a 35-second breath-hold, a 64-slice acquisition (performed at the same gantry rotation time or temporal resolution) on the same patient may require only 15 seconds. This is a significant advance, as breathing motion dramatically degrades image quality and, more often than not, eliminates the opportunity to capitalize on the high negative-predictive value of coronary CTA to determine that the patient has no significant coronary artery stenosis.

Scanning Parameters

As with ECG gating and oversampling, coronary imaging pushes the limits of CT technology with respect to the parameters required to achieve diagnostic images. Typical values of mAs are 550-700mAs with 120kV.The spatial resolution along the Z-axis (the craniocaudal direction) is determined by the image slice thickness and can be as low as 0.4mm.

Although the details of image interpretation are beyond the scope of this article, it is important to point out that one major advantage of coronary CTA in comparison with catheter angiography is the ability to perform multi-planar reconstructed images. The quality of the reconstructed images is inversely proportional to the image slice thickness, and it is beneficial to perform reconstructed images with so called 'isotropic data'; that is, CT data sets where the spatial resolution is equal in the X,Y, and Z directions. At present, the best isotropic resolution commercially available is 0.4 x 0.4 x 0.4mm.Thus, with perfect ECG gating and no respiratory motion, a 3-mm coronary artery spans seven or eight high-quality pixels (3/0.4) in any direction. This explains why properly performed CTA has a high negative-predictive value but can be limited in characterization of stenoses.

Image Field of View

For imaging the native coronaries alone, the superior border of the field of view should be set at the top of the carina, and the inferior border should include the entire inferior wall of the heart. Ideally, the planned field of view should include several slices of the liver to account for cardiac displacement during breath-holding. Because the CT acquisition is in the craniocaudal direction, obtaining a small amount of CT data inferior to the heart does not affect image quality. As hinted in the introduction, cardiac CT can be extended beyond imaging the coronaries alone. In several cases, the greatest protocol change is the field of view. For example, in the assessment of coronary artery bypass grafts, imaging includes the internal mammary arteries, so the superior aspect of the field of view is extended to the apices of the lungs to ensure that the subclavian arteries are included.

For some cross-sectional evaluations (e.g. cardiac MR imaging (MRI),CT myelography), image reconstruction is performed with a limited field of view. In some cases, such as cardiac MRI, limiting the field of view can be beneficial because the imaging time and potential wrap artifact can be minimized. However, coronary CTA data includes complete imaging of the thorax over the entire field of view, and a full field-of-view reconstruction and interpretation is required to evaluate for findings outside the coronary arteries.

Contrast

Dual injection with iodinated contrast followed by saline at rates of at least 5cc per second are now standard. The purpose of the saline is to avoid dense opacification of the right heart and subsequent artifacts that can limit interpretation of the RCA. The volume of contrast material is determined by the rate of contrast injection and the scan time for the prescribed craniocaudal field of view. For example, given a 12-second scan of the native coronaries alone using an injection rate of 5cc per second, an adequate volume of contrast media would be 60cc (12 seconds x 5cc/sec). Because the timing of the contrast bolus may be imperfect, a slightly larger volume of contrast (e.g. an additional 10cc) may be administered without introducing artifacts. Typically, 50cc of saline following the contrast is adequate to eliminate artifacts that obscure analysis of the RCA. One consequence of the more widespread use of cardiac CT scanners with better temporal resolution (faster rotation time) and more rows (more Z-axis coverage per rotation) is an overall decreased volume of contrast material used.

Conclusion

Because it must overcome cardiac motion, coronary CTA is the most sophisticated CT examination to date. Study requires cardiac gating, high spatial and temporal resolution, and imaging to push the limit of CT technology. However, understanding and careful adherence to CT protocols can assure that clinically useful images are routinely obtained in the vast majority of patients, enabling those patients to benefit from the diagnostic power of CT.