Dose/quality balance dominates in cardiac CT

August 19, 2008

Interest in cardiac imaging with multislice CT is growing, as evidenced by the large number of studies that have been published on this topic. Advances in cardiac MSCT have also been aided by the introduction of extremely fast, user-friendly scanners.

Interest in cardiac imaging with multislice CT is growing, as evidenced by the large number of studies that have been published on this topic. Advances in cardiac MSCT have also been aided by the introduction of extremely fast, user-friendly scanners.

The radiation dose associated with ECG-gated cardiac MSCT, however, has generated considerable debate.1 The International Commission on Radiological Protection recommends that "all exposures [are kept] as low as reasonably achievable."2 This goal is known as the ALARA principle, and many strategies have been introduced to achieve it. These have been based on x-ray emissions and/or scanning parameters such as mAs, kV, pitch, and collimation and on the individual patient's characteristics (i.e., automatic exposure control systems and ECG-pulsing techniques for ECG-gated acquisitions).

The adjustment of scanning parameters can lead to an increase in image noise, which may, in turn, affect the diagnostic acceptability of images. This has been a major concern. Scan parameters should be tailored to provide diagnostic images, but without exposing patients to unnecessary radiation.

FOCUS ON DOSE

The fundamental parameter for measures of radiation dose is the CT dose index. The CTDI, the radiation dose absorbed by a dose phantom, is measured in either grays (Gy) or rads (1 Gy = 100 rad). It represents the integrated dose along the z-axis from a single axial CT scan (one rotation of the x-ray tube).

The weighted CTDI (CTDIw) accounts for differences in absorbed dose within the scanned region. Absorption is approximately twice as high at the surface as at the center of the field-of-view. When considering volumetric scan protocols, it is also essential to consider any gaps or overlaps between radiation dose profiles from consecutive rotations of the x-ray tube. The volume CTDI (CTDIvol) can be used to express the average dose delivered to the scan volume for a specific examination. This value can be read directly from the scanner, allowing a direct and rapid estimate of the amount of radiation delivered to patients.

The absorbed dose does not account for differences in the sensitivity of target organs to radiation damage.3 This assessment is achieved with a second parameter: the equivalent dose, which is calculated by multiplying the absorbed dose to a specific tissue with a radiation weighting factor. The weighting factor for x-rays is approximately unity, so the equivalent dose has the same numerical value as absorbed dose. Equivalent dose is measured in sieverts (Sv) or rems (1 rem = 10 mSv).

Another important parameter is the effective dose, which is useful when assessing the potential biological risk from specific x-ray examinations. It is calculated by summing the absorbed doses in individual organs, weighted by their radiation sensitivity. This provides an estimate of the whole-body dose required to produce the same risk as a partial-body dose delivered by a localized radiological procedure. The effective dose allows comparisons among different x-ray examinations. Doses associated with medical imaging examinations can also be compared with those received from other sources; for example, natural background radiation.

CT image noise generally depends on the number of x-ray photons interacting with the detector array (quantum noise), the electronic noise of the detector system, and the reconstruction kernel (sharper kernels give noisier images). The level of noise can be quantified easily from images by positioning a standard region of interest in a structure of known density and measuring the standard deviation of the Hounsfield units values.

Noise is also dependent on the patient's individual attenuation, convolution filters, slice thickness, pixel dimension, and the radiation dose. The relationship between image noise and individual dose is such that the dose must be quadrupled to halve the noise.

Operator-selected scanning parameters-tube current (mAs), tube potential (kV), and scan length-affect significantly the radiation dose received by patients. Dose is also dependent on scanner-specific design factors, though to a lesser extent. Multislice scanners that allow higher mAs values, longer scan lengths, and multiphase studies have the potential to increase the amount of radiation a patient gets.

Another indirect but significant factor that can affect dose is image slice width. The smaller the slice width, the greater the level of image noise. Because MSCT scanning is usually performed with narrower slices than single-slice CT, higher mAs values are required to keep the level of noise constant.

MSCT image quality and radiation dose are also affected by helical pitch; that is, the table feed divided by the x-ray beam collimation width. The radiation dose decreases proportionally with increasing pitch in CT systems with a single detector row, so long as the tube voltage and current are kept constant. For MSCT scanners, the relationship between pitch and radiation dose is nonlinear in ECG-gated acquisition cases and linear in noncardiac mode.

CARDIAC CONCERNS

The introduction of advanced MSCT systems with gantry rotation times of up to 330 msec has been a significant step forward for cardiac CT studies. Images can be acquired using multisegment reconstructions with an isotropic resolution up to 0.4 mm and a temporal resolution up to ~85 msec, depending on the patient's heart rate.

The radiation dose associated with ECG-gated MSCT is relatively high because data are acquired at a low pitch with considerable overlap (up to 0.18 for a 0.33-sec rotation) to ensure sufficient data sampling over the entire cardiac volume with no interslice gap.4-6 Examinations are also being performed using at least half reconstruction algorithms (180° + x-ray beam fan angle) to optimize temporal resolution and reduce motion artifacts. A higher mAs will be needed for the same slice thickness to get signal-to-noise levels comparable with noncardiac studies (pitch 0.8 to 1), and so the dose to patients will be raised.7

ECG-gated cardiac studies performed with 64-slice CT using fixed tube current techniques have produced estimates of effective dose ranging from 13 mSv to 21.4 mSv.

8,9

This reading indicates that the dose delivered to patients by these latest generation scanners will be significantly higher than if four- or 16-slice systems were used. The patient dose associated with these earlier technologies for ECG-gated cardiac CT is 7.1 to 11.9 mSv and 8.8 mSv, respectively.

3,10

In comparison, the dose delivered to patients undergoing conventional coronary angiography ranges from 2 mSv to 9 mSv (Figure 1).

3,11,12

A simple way of reducing radiation dose is to lower the x-ray tube current and tube voltage. This reduces the number and mean energy of photons interacting with the detector array.3,11 Studies have shown that a drop in x-ray tube voltage from 120 kVp to 80 kVp can produce a 70% reduction in radiation for a given tube current.13,14 Doing this will inevitably increase image noise and decrease image quality. Low-dose CT image quality can be improved, however, by using a postprocessing method based on noise reduction filters.

A more sophisticated way of reducing radiation dose is to adjust the tube current according to the patient's individual characteristics with the use of automatic exposure control.

15

This functionality has been an integral part of conventional x-ray units for many years. Automatic exposure control systems have now been developed for CT as well and introduced into clinical practice.

15,16

A set of modulation techniques allows the tube current to be adjusted according to the patient's overall size and variations in attenuation along the z-axis (z-axis modulation), as well as during the course of each rotation (x-y or angular modulation). Systems may adjust the tube current according to one of these factors or use a combination (Figure 2).

Dose reductions of 20% to 68% have been reported when automatic exposure control was used instead of fixed tube current techniques for body CT applications.15 We found that for patients with coronary artery disease who underwent retrospective ECG-gated cardiac CT on a 64-slice system, automatic exposure control reduced the effective dose by 21.7% to 29.4% compared with nonadapted techniques.17

The dose associated with ECG-gated cardiac CT can also be reduced by the use of ECG pulsing techniques. These involve a progressive online reduction of tube output during the systolic phases of each cardiac cycle. Tube current is maintained at normal values during the mid-end diastolic phase but drops to a lower level (e.g., 20% of its nominal value) for the remainder of the cardiac cycle. The rationale behind this technique is that only data acquired in the mid-to-late diastolic phase, when there is least cardiac motion, are used for image reconstruction (Figure 3). Dose savings of approximately 50% have been reported with this approach.

18

LOOKING AHEAD

The introduction of dual-source CT (DSCT) systems has improved cardiac CT image quality while also lowering dose significantly.19-22 The use of two orthogonal x-ray tubes has halved exposure time and cut temporal resolution to well below 100 msec.19,20

The main advantage of DSCT technology with ECG-gated examinations is the possibility of performing single-segment reconstructions and applying pitch values as high as 0.5 for patients with heart rates above 100 beats per minute (Figure 4).

21,22

Multisegment reconstruction methods that must be used when such patients are examined on single-source CT systems, because of the severity of motion artifacts, combine projection data derived from several adjacent R-R intervals to improve effective temporal resolution. For scanners with sub-0.4-sec rotation times, multisegment reconstruction requires a low-speed table, corresponding to pitch values of approximately 0.2, which leads to an obvious increase in radiation dose.

21

One group has reported an effective dose of 7.8 mSv to 8.8 mSv for coronary artery assessment, however, when DSCT was combined with a range of different ECG-pulsing algorithms.

22

Another proposed technical solution uses a prospectively triggered axial step-shoot acquisition technique (SnapShot Pulse, GE Healthcare). The tube current is applied during a predefined cardiac phase and then turned off completely for the remainder of the cycle.

23

A statistical model based on the length and variability of previous heart cycles is used to time the next axial scan, thus minimizing motion artifacts related to possible heart rate variations during data acquisition (Figure 5).

An initial study using prospective triggering step-and-shoot with a single-source CT scanner produced an average effective dose of 5.6 mSv (range 0.8 to 8 mSv) for coronary artery assessment and 9.8 mSv (range 4 to 13 mSv) for aorta and/or bypass examinations. The image quality was no different from that achieved with retrospective ECG gating.23

Given the level of patient exposure and the growing availability of faster scanners, it is essential for radiologists to understand the estimates of radiation dose associated with cardiac MSCT. Radiologists should also become familiar with strategies for optimizing image quality at lower doses. Careful selection of scan parameters is imperative if the important ALARA principle is to be followed.

Dose parameters should be defined precisely and reported in a standardized format, allowing direct comparison between different CT protocols. Ongoing technical developments should help practitioners reduce radiation doses without affecting image quality in cardiac CT.

DR. FRANCONE, PROF. CATALANO, and DR. VASSELLI are radiologists in the department of radiological sciences at University Sapienza of Rome. Assisting in the preparation of this manuscript were Dr. Luca Francone from the division of radiology, Casa di Cura "nuova Villa Claudia," and Dr. Raffaele Lezoche from the department of radiological sciences (chairman: Prof. Roberto Passariello), University of Sapienza, Rome.

References

1. Kalender WA. Considerations for multi-slice spiral CT. In: Kalender WA, Computed tomography: fundamentals, system technology, image quality, applications. Munich: Publicis MCD Verlag, 2000:76-78.
2. International Commission on Radiological Protection. 1990 Recommendations of the ICRP: ICRP Publication 60. Ann ICRP 1991;21(1-3).
3. Morin RL, Gerber TC, McCollough CH. Radiation dose in computed tomography of the heart. Circulation 2003;107(6):917-922.
4. Schoepf UJ, Becker CR, Ohnesorge BM, Yucel EK, et al. CT of coronary artery disease. Radiology 2004;232(1):18-37.5. Ohnesorge B, Flohr T, Becker C, et al. Cardiac imaging by means of electrocardiographically gated multisection spiral CT: initial experience. Radiology 2000;217(2):564-571.
6. Wintersperger BJ, Nikolaou K. Basics of cardiac MDCT: techniques and contrast application. Europ Radiol 2005;15(Suppl 2):B2-B9.
7. Flohr TG, Stierstorfef K, Ulzheimer S, et al. Image reconstruction and image quality evaluation for a 64-slice CT scanner with z-flying focal spot. Med Phys 2005;32(8):2536-2547.
8. Mollet NR, Cademartiri F, van Mieghem CA, et al. High-resolution spiral computed tomography coronary angiography in patients referred for diagnostic conventional coronary angiography. Circulation 2005;112(15):2318-2323.
9. Raff GL, Gallagher MJ, O'Neill WW, et al. Diagnostic accuracy of noninvasive coronary angiography using 64-slice spiral computed tomography. J Am College Cardiol 2005;46(3):552-557.
10. Hunold P, Vogt FM, Schmermund A, et al. Radiation exposure during cardiac CT: effective doses at multi-detector row CT and electron-beam CT. Radiology 2003;226(1):145-152.
11. Bae KT, Hong C, Whiting BR. Radiation dose in multidetector row computer tomography cardiac imaging. J Mag Reson Imag 2004;19(6):859-863.
12. Coles DR, Smail MA, Negus IS et al. Comparison of radiation doses from multislice computed tomography coronary angiography and conventional diagnostic angiography. J Am College Cardiol 2006;47(9):1840-1845.
13. Jakobs TF, Wintersperger BJ, Herzog P, et al. Ultra-low-dose coronary artery calcium screening using multislice CT with retrospective ECG gating. Europ Radiol 2003;13(8):1923-1930.
14. Hohl C, Mühlenbruch G, Wildberger JE, et al. Estimation of radiation exposure in low-dose multislice computed tomography of the heart and comparison with a calculation program. Europ Radiol 2006;16(8):1841-1846.
15. Kalra MK, Maher MM, Toth TL, et al. Techniques and applications of automatic tube current modulation for CT Radiology 2005;233(3):649-657.
16. Francone M, Napoli A, Carbone I, et al. Noninvasive imaging of the coronary arteries using a 64-row multidetector CT scanner: initial clinical experience and radiation dose concerns. Radiol Med (Torino) 2007;112(1):31-46.
17. Francone M, Castro E Di, Napoli A, et al. Dose reduction and image quality assessment in 64-detector row computed tomography of the coronary arteries using an automatic exposure control system. J Comput Assist Tomogr. In press.
18. Jakobs TF, Becker CR, Ohnesorge B, et al. Multislice helical CT of the heart with retrospective ECG gating: reduction of radiation exposure by ECG-controlled tube current modulation. Europ Radiol 2002;12(5):1081-1086.
19. Flohr TG, McCollough CH, Bruder H, et al. First performance evaluation of a dual-source CT (DSCT) system. Europ Radiol 2006;16(2):256-268.
20. Johnson TR, Nikolaou K, Wintersperger BJ, et al. Dual-source CT cardiac imaging: initial experience. Europ Radiol 2006;16(7):1409-1415.
21. McCollough CH, Primak AN, Saba O, et al. Dose performance of a 64-channel dual-source CT scanner. Radiology 2007;243(3):775-784.
22. Stolzmann P, Scheffel H, Schertler T, et al. Radiation dose estimates in dual-source computed tomography coronary angiography. Europ Radiol 2008;18(3)592-599.
23. Sablayrolles JL, Treutenaere JM, Feignoux J, et al. Cardiac CT exam at 5 mSv average without compromising the image quality with the SnapShot Pulse mode on LightSpeed VCT XT. Presented at annual meeting of European Society of Cardiac Radiology, Rome; October 2007:2714.