April 1, 2019 — Here is a checklist of dose-sparing practices for coronary computed tomography angiography (CCTA) imaging. This list was included in a new 2018 consensus document to guide the optimal use of ionizing radiation in cardiovascular imaging.
The consensus document was issued in May 2018 jointly by the American College of Cardiology (ACC), Heart Rhythm Society (HRS), North American Society for Cardiovascular Imaging (NASCI), Society for Cardiovascular Angiography and Interventions (SCAI), Society of Nuclear Medicine and Molecular Imaging (SNMMI) and the Society of Cardiovascular Computed Tomography (SCCT). It includes input from experts from a total of nine cardiology societies and includes best practices for safety and effectiveness when using computed tomography (CT), nuclear imaging and angiographic/fluoroscopic imaging.
The guidelines offer a checklist to help lower CT dose in cardiovascular CT:
Patient Case Selection:
• Consider patient age, co-morbidities, natural life expectancy
• Consider appropriateness and utility of nonradiation-based imaging techniques
• Use acquisition detector doses as low as compatible with diagnostic image quality
• Select lowest-dose acquisition protocol compatible with study goals. Retrospective gating should be selected when feasible.
• Use ECG-gated variable tube output if retrospective gating is used
• Use the lowest X-ray tube voltage compatible with adequate diagnostic quality image acquisition
• Use the lowest X-ray tube current compatible with diagnostic quality image acquisition. Use topogram-based tube current modulation
• Use the largest scan pitch compatible with adequate diagnostic quality image acquisition
• Minimize patient heart rate
• Confine scanned body area to the area relevant to the study’s diagnostic purpose
Variables That Affect Patient CT Dose
The consensus document states radiation dose to a patient is determined by a combination of the patient’s physical characteristics and scanner protocol selection. Patient X-ray exposure will need to increase with patient size and body mass index (BMI) to enable diagnostic-quality images. Depending on the specific acquisition parameters, the increased exposure does not need to increase dose to radiosensitive tissues. Patient size is not a variable that determines exam appropriateness as long as the patient’s size does not preclude obtaining diagnostic-quality images, the guidelines state
CT systems may either use a constant X-ray tube output or, in some acquisition protocols, ECG-gated variable output. The operator needs to select the acquisition protocol based on patient characteristics and the study purpose with the intent to deliver a sufficient exposure to permit an acceptable degree of noise in the reconstructed images.
The consensus document also states the CT system operator is responsible for selecting the scanning protocol that optimizes the examination’s diagnostic yield while minimizing dose. Essential considerations in this process include:
Scan Length: Scan length, defined as the distance imaged along the cranio-caudal axis, should be kept to a minimum to encompass only the anatomy of interest and not expose structures that are not relevant to the examination’s purpose. Care needs to be taken to ensure that the diaphragm position seen on the topogram is the same as during the scanning. This requires similar breath hold instructions.
X-ray Beam Intensity: X-ray beam intensity is determined by both the X-ray tube potential (in units of kV) and the X-ray tube current (in units of mA). Modern CT scanners modulate the tube current dynamically throughout the CT acquisition to minimize radiation exposure.
Tube Potential: Studies of radiation dose reduction have shown the most important factor in controlling dose is adjustment of X-ray tube voltage.[2-4] The authors of the consensus document said increasing tube voltage increases the X-ray beam’s mean photon energy level and increases radiation dose roughly proportionally to the square of the voltage. This means with a constant tube current, a decrease of tube voltage from 120 to 100 kilovolts (kV) reduces the radiation dose by nearly 40 percent. The tube voltage in most scanners can be adjusted between 70-140 kV. The voltage is selected by the operator based on the patient’s weight or BMI. The consensus document cites an example of a commonly used adjustment scale to provide diagnostic-quality images with most scanners. For example, 120 kV is used for patients with BMI of 30 kg/m2 or more, 100 kV for BMI of 21-29 kg/m2, and 80 kV for BMI of less than 21 kg/m2. Image noise decreases as potential increases, so in extreme cases (BMI of 40 kg/m2 or larger) the maximum tube potential of 150 kV may be necessary to produce diagnostic quality images, the authors stated. Consequently, selecting the tube potential involves a trade-off between image noise and dose.
Tube Current: The X-ray tube current (in mA) is defined as the number of electrons accelerated across the tube per unit of time and is proportional to the number of X-ray photons produced per unit time, the consensus document states. The radiation dose is linearly proportional to the tube current. Image noise is inversely proportional to the square root of the tube current. So, decreasing tube current at a given tube potential decreases the dose at the expense of increased image noise. The tube current may be modified based upon patient size assessed by visual inspection, measurement of body weight or body mass index, thoracic circumference or diameter, or noise measurement from a cross-sectional scout scan or topogram. Most modern scanners offer tube current modulation based upon the thickness of the body estimated from the topogram, the experts said. Modulation may be applied longitudinally as well as circumferentially. This approach can reduce radiation exposure of thoracic CT examinations by 20 percent without increasing image noise. A specific form of tube current modulation that is applicable to retrospective ECG gating modulates the tube current during each heart beat relative to position within the R-R interval.
Rotation time: Defined as the time required for the gantry to perform one rotation. Exposure increases linearly with rotation time.
X-ray beam filtration: Filters placed beneath the X-ray tube are used to selectively attenuate low-energy X-rays that do not significantly contribute to the image, but do contribute to radiation dose. The net effect is to increase the mean energy of the X-rays while not altering the maximum energy. The authors of the consensus document said these filters may be small, medium or large, and either flat or bowtie. The choice of filter depends on the size of the patient and the acquisition field of view.
Scan Acquisition Mode: This is a major determinant of radiation dose, the consensus document states. Different acquisition modes can deliver substantially different doses while producing similar images. There are three principal CT scan modes — axial or “conventional” scanning, helical scanning, and fixed table or single-station scanning.
Axial scanning may or may not be ECG triggered. It images a portion of the anatomy during a single gantry rotation while the table is stationary, then the table advances to the next position to continue imaging. The imaging segments are based on the width of the detector array and another scan ensues. The process is repeated until the full area of interest has been imaged.
Helical scanning combines continuous gantry rotation with continuous table advancement to trace a contiguous helical, or spiral, path of the scan. The ratio of the width of the detector array to the distance that the table advances per complete gantry rotation is the scan pitch. Radiation exposure for helical scanning at a pitch of one is comparable to axial scanning, the consensus document authors state. When the pitch is less than one, the radiation exposure is greater. When the pitch is more than one, radiation exposure is decreased. A specialized form of helical scanning called high pitch scanning has been developed for use in dual-source CT scanners. Dual-source scanners with two X-ray tube/detector systems can interleave two sets of projection data acquired simultaneously. However, the two beams are separated in-plane by approximately 90 percent, allowing the pitch to increase to more than three. When applied to cardiac imaging, the reduced overlap between gantry rotations in this scanning mode reduces radiation dose more than any of the other scan modes, to values of less than 1 mSv.[8,9] However, high-pitch scanning at its current stage of development is vulnerable to image artifacts, the authors wrote, explains it is suitable for coronary artery imaging only in patients with slow, very regular heart rates. They also stated it has value in circumstances where minor imaging artifacts can be accepted, as in pulmonary vein mapping.
Fixed table scanning uses a specialized form of axial scans performed when using a CT with a detector array that has a width equal, or exceeding, the length of the anatomy of interest, the documents states. In this instance, the table remains stationary during a single or multiple gantry rotations. Radiation exposure approximates axial scanning when a single gantry rotation is applied and increases linearly with increased gantry rotation time.
Cardiac Motion Compensation: Compensation for cardiac motion is rarely applied outside of direct cardiac and aortic root imaging. So, the majority of cardiovascular imaging does not employ ECG gating or triggering. In contrast, when imaging the heart or aortic root, cardiac motion compensation is critical to avoiding motion-related artifacts that degrade image quality. Depending upon the scan mode, there are two cardiac compensation methods used.
Prospective ECG triggering involves the operator prospectively selecting an imaging window within the cardiac cycle prior to the scan. This window may be defined as a percentage from one R-wave to the next or an absolute time delay after each R-wave.. Scans are then triggered to coincide with the selected scan window. Prospective triggering may be applied to each of the three scan modes. In the case of axial scanning, ECG triggering is used to trigger the acquisition at each table position. Scanning occurs during every other heartbeat and the table is incremented during intervening heart beats, the authors explained. The principle is the same with fixed table scanning, but a single scan is acquired, initiated by the ECG trigger.In the case of high-pitch scanning, a single sub-second helical acquisition can be triggered based upon a prospectively acquired ECG. For the specific application of coronary CTA, prospective triggering has been associated with the lowest-dose scans, but effective prospective triggering requires a regular, slow heart rate (typically 50-65 beats per minute for most scanners). The data acquisition window may be widened with what is called “padding” to allow for retrospective adjustments of the acquisition window at the expense of increased radiation, the document states. A disadvantage of prospective gating is the potential that, if the image quality proves to be unsatisfactory, the entire scan must be repeated because no projection data are available from other portions of the cardiac cycle.
Retrospective gating is applicable to both helical and fixed table scan modes. With helical scanning, acquisition is performed using a low pitch of about 0.2. The slow acquisition images the entire cardiac anatomy across the entire cardiac cycle, providing a 4-D dataset that allows each spatial location within the heart to be reconstructed at any time-point across the cardiac period, the consensus document authors wrote. Data are continuously acquired along with the ECG signal while covering the anatomy of interest. The data are subsequently rebinned at each slice location for image reconstruction, according to the time of the cardiac cycle from the ECG signal. The selection of a specific time-point for reconstruction is determined after the scan is completed, or retrospectively.
Retrospective gating is the only method that allows the assessment of dynamic cardiac structures such as native and prosthetic valves, myocardium and chamber dimensions, the experts explained. It also can reveal intracardiac shunts and dehiscent graft anastomoses, owing to variations in iodine enhancement across the cardiac cycle. When compared with prospective triggering, coronary CTA performed with retrospective gating offers diagnostic image quality of the coronary arteries in patients with higher basal heart rates and a greater degree of beat-to-beat variability and allows assessment of regional myocardial function. However, retrospective gating is associated with a higher radiation dose. Typically, image reconstruction for static coronary CTA uses only the time period during the cardiac cycle when cardiac motion is minimal (diastole), and the helically arranged projection data from other periods of the cardiac cycle are ignored. This scanning mode, while radiation-inefficient, has a number of advantages. In particular, the entire image dataset is available for image reconstruction, enabling post-processing selection of the best quality images. In addition, because image data are available from the entire cardiac cycle, images from different portions of the cardiac cycle can be combined to construct cine loops to show global and regional left ventricular function.
ECG-triggered Tube Current Modulation: ECG-triggered tube current modulation is used to reduce radiation dose during systole when there is the greatest cardiac motion and can reduce the radiation exposure significantly, the authors wrote. In this circumstance, tube current is at nominal value only during the portion of the cardiac cycle likely to be used for reconstruction (typically end diastole). During the remainder of the cardiac cycle, the tube current is reduced to reduce radiation output. The authors said recent refinements of this technique have allowed reduction of the window during which tube current is nominal and reduction of tube current during the undesired portions of the cardiac cycle by 20 percent, and to as little as 3-5 percent of the nominal value.[11-13] One disadvantage of this technique is that images reconstructed from projection data acquired with low tube current might be too noisy to be diagnostic for coronary anatomy, thee authors stated. Retrospectively ECG-triggered tube current modulation works best in patients with stable sinus rhythm and low heart rates.
Image Reconstruction: Filtered back-projection has historically been used to reconstruct CT images, but the development of greater computing power has made an alternative statistical method of iterative reconstruction practical for CT, the consensus document states. This method predicts projection data based upon an initial assumption about the attenuation in each voxel and compares that data to measured projection data. The voxel attenuation values are modified iteratively until an acceptable level of error between the predicted and measured data is obtained. The resulting reconstructed images have lower noise values compared with those obtained with filtered back projection. This permits reducing tube voltage and/or current to obtain images with comparable noise and lower radiation dose.[14,15]
The consensus authors point out that an important characteristic of iterative reconstruction is that excessively low-dose images do not appear grainy, as is the case with filtered back projection. Instead, structures become blurred and can develop a blotchy or "plastic" appearance, which might undermine their diagnostic value.
Image Post-processing Filters: These also can be applied to acquired images to reduce image noise while preserving image contrast and edges. The feasibility of using these filters for radiation dose reduction has been recently demonstrated, the authors stated.
Related CT Content:
VIDEO: The Role of Advanced Imaging in Structural Heart Interventions — Interview with Robert Quaife, M.D.
1. John W. Hirshfeld Jr., Victor A. Ferrari, Frank M. Bengel, et al. 2018 ACC/HRS/NASCI/SCAI/SCCT Expert Consensus Document on Optimal Use of Ionizing Radiation in Cardiovascular Imaging: Best Practices for Safety and Effectiveness: A Report of the American College of Cardiology Task Force on Expert Consensus Decision Pathways. Journal of the American College of Cardiology. May 2018. DOI: 10.1016/j.jacc.2018.02.016.
4. Sigal-Cinqualbre A.B., Hennequin R., Abada H.T., et al. (2004) Low-kilovoltage multi-detector row chest CT in adults: feasibility and effect on image quality and iodine dose. Radiology 231:169–174.
6. Mulkens T.H., Bellinck P., Baeyaert M., et al. (2005) Use of an automatic exposure control mechanism for dose optimization in multi-detector row CT examinations: clinical evaluation. Radiology 237:213–223.
8. Achenbach S., Goroll T., Seltmann M., et al. (2011) Detection of coronary artery stenoses by low-dose, prospectively ECG-triggered, high-pitch spiral coronary CT angiography. J Am Coll Cardiol Img 4:328–337.
9. Alkadhi H., Stolzmann P., Desbiolles L., et al. (2010) Low-dose, 128-slice, dual-source CT coronary angiography: accuracy and radiation dose of the high-pitch and the step-and-shoot mode. Heart 96:933–938.
11. Abada H.T., Larchez C., Daoud B., et al. (2006) MDCT of the coronary arteries: feasibility of low-dose CT with ECG-pulsed tube current modulation to reduce radiation dose. AJR Am J Roentgenol 186:S387–S390.
12 Buechel R.R., Husmann L., Herzog B.A., et al. (2011) Low-dose computed tomography coronary angiography with prospective electrocardiogram triggering: feasibility in a large population. J Am Coll Cardiol 57:332–336.
13. Wang R., Yu W., Wang Y., et al. (2011) Incremental value of dual-energy CT to coronary CT angiography for the detection of significant coronary stenosis: comparison with quantitative coronary angiography and single photon emission computed tomography. Int J Cardiovasc Imaging 27:647–656.
14. Leipsic J., LaBounty T.M., Heilbron B., et al. (2010) Estimated radiation dose reduction using adaptive statistical iterative reconstruction in coronary CT angiography: the ERASIR study. AJR Am J Roentgenol 195:655–660.
15. Leipsic J., LaBounty T.M., Heilbron B., et al. (2010) Adaptive statistical iterative reconstruction: assessment of image noise and image quality in coronary CT angiography. AJR Am J Roentgenol 195:649–654.