| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Diagnostic Radiology, Graduate School of Medical Sciences,
Kumamoto University, 1-1-1 Honjo, Kumamoto 860-8556, Japan.
2 Department of Radiological Technology, School of Health Sciences, Kumamoto
University, Kumamoto, Japan.
Received March 17, 2005;
accepted after revision September 18, 2005.
Address correspondence to Y. Nakayama
(yoshiharu156{at}lily.ocn.ne.jp).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. We evaluated 74 patients for aortic disease on MDCT angiography (collimation, 16 x 1.5 mm; beam pitch, 0.9). In 36 patients, we used the standard tube voltage (120 kVp) and a contrast dose of 100 mL (300 mg I/mL) (protocol 1), and in the remaining 38 patients we applied a reduced tube voltage (90 kVp) and a contrast dose of 40 mL (300 mg I/mL) (protocol 2). The patients' weights, CT attenuation of the aorta, visualization of the celiac axis and renal artery, and graininess and streak artifacts on transverse CT scans were evaluated and recorded for each data set. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were also measured. For statistical analysis, we used the two-tailed Student's t test and logistic regression; agreement between measurements recorded independently by two blinded reviewers was assessed using Cohen kappa statistics.
RESULTS. In both protocols a negative correlation was seen between
patient weight and CT attenuation. In three protocol 1 patients weighing more
than 70 kg, CT attenuation was less than 200 H. No difference was seen between
the two protocols with respect to mean attenuation of the aorta (p =
0.13) or visualization of the celiac axis and renal artery (p = 0.35
and 0.60, respectively). Although the SNR and CNR were significantly higher in
protocol 1 than in protocol 2, qualitative evaluation of graininess and streak
artifacts showed no statistically significant difference (p = 0.15
and 0.48, respectively). Interobserver agreement for quality assessments was
within an acceptable range (
= 0.42-0.80).
CONCLUSION. Low-contrast and low-voltage scans are appropriate for lighter patients (< 70 kg in body weight) with aortic disease. Moreover, this method is particularly valuable for follow-up studies of heavier patients (> 70 kg) with renal dysfunction.
Keywords: aortography • arteriography • contrast media • CT • radiation dose • tube voltage
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
As the use of MDCT has become routine in clinical practice, concerns have been raised regarding radiation exposure [5]. According to the current literature, the theoretic risk to patients for radiation-induced cancer from CT examinations is not negligible [6-9]. Several techniques for achieving a reduction in the radiation dose during CT examinations have therefore been reported [10-14]. Although low tube voltage is one option to achieve a radiation dose reduction in patients undergoing CT [5, 14, 15], few published scanning protocols for low-voltage CT angiography with reduced contrast doses are available.
Huda et al. [16], who studied the relationship between tube voltage and the CT attenuation value of iodine, confirmed that the attenuation value of iodinated enhancement is increased at a lower tube voltage, resulting in higher enhancement. Our previous studies revealed that, keeping all other parameters the same, dynamic CT of the abdominal organs and aorta provides much more intensive enhancement at 90 kVp than the standard voltage of 120 kVp [17, 18]. Wintersperger et al. [19] reported that low-voltage CT angiography yields higher enhancement and a simultaneous reduction in radiation dose compared with standard-voltage CT angiography. These studies suggest that in CT angiography, the low-tube-voltage technique might deliver diagnostic images at a reduced contrast dose. Consequently, it might be possible to reduce both the contrast dose and the radiation exposure without sacrificing image quality. To our knowledge, no published studies have evaluated the effect of low-voltage CT on images produced at reduced contrast doses. Therefore, we assessed the diagnostic efficacy of aortic MDCT angiography using a low tube voltage and a reduced contrast dose.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Patients
Between January and December 2004, we evaluated 74 patients (40 men, 34
women; mean age, 68.1 years; range, 42-86 years) with a history of
arteriosclerosis or aortic aneurysm. The patients were referred for follow-up
aortic CT angiography for the assessment of a known infrarenal abdominal
(n = 25) or thoracic (n = 22) aortic aneurysm,
arteriosclerosis (n = 14), or follow-up after aneurysm repair or
stent-graft placement (n = 13). Patients with aortic dissection were
excluded because their slower aortic flow might have made data acquisition
more difficult in studies involving the use of reduced contrast doses. The 74
patients were randomly assigned to CT aortography protocol 1 (120-kVp tube
voltage, 100 mL of contrast material; n = 36) or protocol 2 (90-kVp
tube voltage, 40 mL of contrast material; n = 38). The protocol 1
group consisted of 20 men and 16 women ranging in age from 51 to 86 years
(mean, 69.4 years), and the protocol 2 group, of 20 men and 18 women ranging
in age from 42 to 76 years (mean, 67.3 years). The patients' demographic data,
including their age, sex, and body weight, are recorded in
Table 1. No statistically
significant differences existed between protocol 1 and protocol 2 patients
with respect to age (two-tailed Student's t test, p = 0.71),
sex (chi-square test, p = 0.80), or body weight (56.0 ± 8.6 kg
vs 58.2 ± 9.5 kg, respectively). Patient distribution in different
weight classes was equal in both protocols.
|
CT and Contrast Injection Protocols
In all CT examinations we used an MDCT scanner (IDT16, Philips Medical
Systems) capable of acquiring 16 sections per gantry rotation. In the protocol
1 group, we applied a tube current of 260-300 mAs according to patient body
weight (Table 2); 260 mAs was
used when the patient's weight was less than 50 kg, and 300 mAs when it was
greater than 60 kg. Patients weighing between 50 and 60 kg were scanned at 280
mAs. We used the nominal CT dose index (nominal CTDIvol) of each tube current
setting based on the manufacturer's data for estimation of radiation exposure.
The nominal CTDIvol values were 12.9 mGy at 260 mAs, 14.6 mGy at 280 mAs, and
15.0 mGy at 300 mAs at a tube potential of 120 kVp.
|
Because images scanned at a 30% dose reduction were previously found to be of diagnostic quality without compromising low-contrast detectability [18], we chose tube current settings for the protocol 2 scans that were theoretically adjusted to approximately 70% of the radiation doses of 120-kVp scans (30% reduction in radiation dose). Patients weighing more than 60 kg were scanned at 485 mAs (nominal CTDIvol: 10.2 mGy); those weighing less than 50 kg, at 405 mAs (8.6 mGy); and those weighing between 50 and 60 kg, at 435 mAs (10.9 mGy).
The remaining parameters were identical in protocols 1 and 2—that is, detector configuration, 1.5 x 16 mm; table feed, 28.8 mm/s; beam pitch, 0.9; and tube rotation time, 0.75 seconds. All scans were obtained in a single breath-hold. Each examination began with the acquisition of a series of unenhanced conventional CT images from the upper level of the clavicle to the level of the iliac bifurcation. After review of these images, contrast-enhanced scans usually ranged from the level of the aortic arch to at or above the aortic bifurcations. Contiguous slices reconstructed for image analysis were set at 5 mm in both protocols.
In the protocol 1 group, 100 mL of low-osmolar contrast material (Iopamiron 300 [iopamidol], Nihon Schering) was administered at 3.0 mL/s and no saline chaser was delivered. In the protocol 2 group, 40 mL of the same contrast material was administered at 2.0 mL/s followed by a 20-mL saline flush (2.0 mL/s) using a twin injector (Dual Shot, Nemoto-Kyorindo). In each patient, a low-dose automatic-timing bolus protocol (120 kVp, 20 mAs) (Bolus Pro Ultra, Philips Medical Systems) was used to optimize the delay time from the start of injection to the start of scanning at the level of the descending aorta. Scanning was started 4 seconds after the attenuation of the ascending aorta increased to 100 H.
Image Reconstruction
Slices (2.0 mm thick) for 3D visualization of the aorta were contiguously
reconstructed every 1.0 mm to achieve a 50% space overlap and to minimize
longitudinal resolution for qualitative evaluation of the 3D aortograms.
Images were sent to an online workstation (M900QUADRA, Amin), and CT
angiograms were produced using volume-rendering and
maximum-intensity-projection (MIP) algorithms.
Quantitative Assessment
Contrast-enhanced attenuation measurements were obtained by placing a
manually defined 1- to 2-cm2 region of interest (ROI) in the lumen
of the abdominal aorta. The ROI measured was as large as possible. All
measurements were performed by a board-certified radiologist with 10 years'
experience interpreting abdominal CT scans. Initial, middle, and final
attenuation measurements were derived from the first image obtained at or near
the level of the aortic arch or descending aorta, the middle image of the
aorta at or above the level of the diaphragm, and the final aortic image at or
above the aortic bifurcation, respectively. In addition, the SD, which serves
as a quantitative marker of the image noise of the surrounding air, and the
attenuation value of the psoas muscle were measured in both protocols. Mean
aortic attenuation for each patient was calculated by averaging the recorded
initial, middle, and final attenuation values. The signal-to-noise ratio (SNR)
and contrast-to-noise ratio (CNR) of the aorta were calculated using the
following formulas:
 |
 |
The relationship between the SNR and CNR values and patient body weight was correlated for protocols 1 and 2.
Qualitative Assessment
Two board-certified radiologists with 8 and 3 years of experience in 3D CT
image interpretation, who were blinded to the technical scanning parameters,
performed qualitative evaluations of the 74 randomized 3D aortograms on
volume-rendering and MIP-reconstructed images. These images were presented on
a digital PACS using a diagnostic workstation (Image VINS Pro version 3.01,
Yokogawa Electric). Visualization of the major aortic branches was evaluated
on the celiac axis and the renal artery using a modification of a previously
proposed CT angiography rating system
[20,
21]. Sharpness of the arterial
boundary of the celiac axis and the renal artery was also assessed using a
3-point scale in which images with a score of 1 were regarded as unacceptable
(unsatisfactory delineation of the boundary); 2, as acceptable (delineation of
the boundary equivocal but within an acceptable range); and 3, as excellent
(boundary clearly delineated, resolution satisfactory).
The images were also scored for image noise (graininess), streak artifacts, and sharpness of the arterial boundary between the celiac axis and renal artery using a 3-point scale in which a score of 1 was regarded as unacceptable (interference by graininess or streaking with visualization of anatomic structures); 2, as acceptable (some graininess or streaking, satisfactory visualization of small anatomic structures); and 3, as excellent (minimal graininess or streaking). When the reviewers' assessments were different, the scans were reevaluated for consensus.
Measurement of Radiation Exposure
Because calculation of the effective dose to each patient could not be made
under our routine clinical conditions, we used the CTDIvol for estimation
[22,
23]. For both protocols, we
determined the measured CTDIvol values with a cylindric polymethyl
methacrylate CT dose measurement phantom (model 20CT14, RadCal). The phantom
was 32.0 cm in diameter and featured a central cavity and four peripheral
cavities, each measuring 13.0 mm in diameter. A pencil beam ion chamber (model
3CT, RadCal) with a 10-cm-long active chamber and a 3 cm3 active
volume was inserted into each cavity to measure radiation exposure at 90 and
120 kVp. We then calculated the measured CTDIvol for both protocols using the
following formula:
 |
Statistical Analysis
To evaluate differences between protocols 1 and 2 in scanning delays,
acquisition times, scanning ranges, and mean attenuations of the aorta, we
used a two-tailed Student's t test. We also examined and graphically
depicted the relationship between patient weight and mean attenuation in both
protocols. SNR and CNR values between the two protocols were also examined
using a two-tailed Student's t test.
For statistical analysis of qualitative data, we used the Mann-Whitney U test. To determine interobserver agreement in the assessment of image quality, visualization of the celiac and renal arteries, graininess, and streak artifacts, we calculated Cohen kappa statistics. All kappa values were interpreted as proposed in the literature [25]. Kappa values less than 0.20 indicated poor agreement; 0.21-0.40, fair agreement; 0.41-0.60, moderate agreement; 0.61-0.80, good agreement; and 0.81-1.00, excellent agreement. We used a statistical software package (SPSS for Windows [Microsoft], release 10.05, SPSS) for all statistical analysis, and p values less than 0.05 were considered to indicate statistically significant differences.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
All patients successfully underwent CT angiography of the aorta. A significant difference was seen in the scanning delay between groups (protocol 1 vs protocol 2: 20.2 ± 2.7 seconds vs 25.9 ± 4.1 seconds; p < 0.001); however, neither the acquisition time nor the scanning range was significantly different between protocols 1 and 2 (Table 3).
|
No significant difference was seen between protocols 1 and 2 with respect to average aortic CT attenuation (protocol 1 vs protocol 2: 289.5 ± 53.4 H vs 270 ± 53.9 H; p = 0.132). The contrast-enhanced attenuations for both protocols at the mean initial, middle, and final levels of the aorta are presented in Figure 1. Although attenuation at the final level of the aorta was lower in protocol 2 than in protocol 1 (protocol 1 vs protocol 2: 292.6 ± 59.7 H vs 246.0 ± 70.2 H), it exceeded 200 H, the value reportedly necessary for diagnostic efficacy of CT angiographs [26] (Figs. 2A, 2B, 2C, and 2D). In four (10.5%) of the 38 patients in protocol 2, attenuation at the final level of the aorta was below 180 H (Figs. 3A, 3B, and 3C). In both groups, we observed a generalized decrease in mean attenuation with increasing patient weight (Fig. 4). All three patients in protocol 2 whose images had a mean attenuation of less than 200 H weighed more than 70 kg. The mean image noise, psoas muscle attenuation value, SNR, and CNR are presented in Table 4. The mean image noise of surrounding air was lower in protocol 1 than in protocol 2 (protocol 1: 6.3 ± 1.2; protocol 2: 8.2 ± 2.1). The mean SNR was 45.2 for protocol 1 and 34.2 for protocol 2, and the difference was statistically significant (p < 0.05). In both protocols, there was an inverse correlation between body weight and SNR values (protocol 1: r = 0.56, p < 0.05; protocol 2: r = 0.51, p < 0.05) (Fig. 5). The mean CNR values were 36.4 for protocol 1 and 27.5 for protocol 2 (p < 0.05). The relationship between CNR and body weight showed an inverse correlation for both protocols (protocol 1: r = 0.51, p < 0.05; protocol 2: r = 0.60, p < 0.05) (Fig. 6).
|
|
|
|
|
|
|
|
|
|
|
|
The mean scores for visualization of the celiac axis and renal artery were 2.7 and 2.6 (protocol 1) and 2.5 and 2.5 (protocol 2), respectively. No significant difference was seen between protocols 1 and 2 with respect to these scores (p = 0.35 and 0.60, respectively), graininess, or streak artifacts (p = 0.15 and 0.48, respectively) (Table 5). The Cohen kappa values for interobserver agreement ranged from 0.42 to 0.80.
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Previous studies have reported threshold attenuation values for obtaining adequate CT angiographic images. Sheiman et al. [26] suggested a value of 160 H as the minimum vessel attenuation needed for CT angiography of the abdomen. More recently, Macari et al. [27] defined a mean attenuation value of 200 H as the minimum threshold for adequate vessel contrast enhancement. Of our 38 patients in protocol 2, 34 (89.5%) had a mean CT attenuation value greater than 200 H after the injection of 40 mL of contrast material.
A lower radiation dose is associated with higher image mottle and might therefore result in the degradation of image quality. In the image noise analysis, the mean image noise was higher in protocol 2 than in protocol 1. However, qualitative analysis showed that good visualization of the celiac axis and renal artery was retained without a significant increase in graininess and streak artifacts. Although SNR and CNR values were significantly lower in protocol 2 than in protocol 1, in 35 (92.1%) of 38 patients in protocol 2 the SNR exceeded 20.
Our current results show that low-tube voltage CT angiography results in a substantial reduction in contrast dose. Many patients undergoing CT angiography present with renal dysfunction due to arteriosclerotic changes in the renal arteries. Reduced-contrast-dose protocols provide protection of renal function. Utsunomiya et al. [28] reported that in CT aortography, the delivery of 50 mL of contrast material followed by a saline flush yielded acceptable contrast enhancement in patients weighing less than 70 kg. Because their study used the standard tube voltage of 120 kVp, it revealed only the effect of lower contrast doses on iodine enhancement. In the present study, we simultaneously examined the effect of lowered contrast doses and tube voltage. In low-voltage scans, the contrast is enhanced because the iodine-based material provides greater X-ray attenuation as the result of an increase in the relative atomic number of iodine and because the X-ray energy is reduced. Furthermore, the k-edge of iodine is closer to the reduced voltage, beam attenuation is increased, and higher attenuation readings are obtained [29].
We found that attenuation values were decreased in protocol 2 when patients weighing more than 60 kg were scanned. To maintain image quality, patients weighing more than 70 kg should be scanned according to the standard (120 kVp, 100 mL) protocol. However, in patients heavier than 70 kg who have renal dysfunction, scanning with a low-voltage, reduced-contrast-dose protocol might help reduce renal overload. Therefore, we recommend the use of this protocol only in patients weighing 70 kg or less or in those with renal dysfunction.
Because the use of a lower dose of contrast material might interfere with the optimal scanning duration, we performed 20-second scans. Because a 2.0 mL/s flow rate is perhaps too low, especially in heavier patients, we followed the bolus contrast injection with a saline flush. Several reports have suggested the importance of a saline chaser in reducing the contrast medium dose in contrast-enhanced CT and CT angiography [30, 31]. In our reduced-contrast-medium technique, a saline chaser is indispensable. In some cases, the contrast material passed through the final level of the aorta before data acquisition was complete. Because we used 16-MDCT, about 20 seconds was required to obtain scans of a large area at a slice thickness of 1.5 mm. With newer instruments such as 32-, 40-, or 64-MDCT scanners, shorter scanning times and higher flow rates can be achieved.
In the current study, we focused on the image quality of vascular enhancement and did not evaluate the image quality of nonvascular structures. Images of the abdomen, pelvis, and chest obtained for vascular indications often yield findings not related to the vascular system. Protocols that involve reduced contrast material and reduced voltage levels might therefore be suitable for follow-up studies of patients with vascular lesions.
Our study has some limitations. First, we excluded patients with aortic dissection because their aortic flow rate might have been too slow for obtaining useful scans with the low-voltage protocol. However, a low-voltage, reduced-contrast-dose protocol is possibly useful in evaluating the patency of aortic dissection or in searching for endoleaks after stent placement, adding a delayed scan of the aorta. Second, we based our evaluation of dose reduction on CTDIvol values rather than estimations of the effective dose because calculations of the effective dose cannot be made under routine clinical conditions. This issue requires further study because the effective radiation dose depends not only on the scanning parameters but also on the volume and radiosensitivity of the target tissues. Finally, protocols that apply a low contrast dose might encounter instances of inhomogeneous aortic opacification. However, this problem can be overcome with the newest CT scanners by using a much higher flow rate and compact contrast medium injection.
In conclusion, we evaluated the quality of images obtained with a low-tube-voltage, reduced-contrast-dose protocol in patients undergoing aortic MDCT angiography. By changing the tube voltage from 120 to 90 kVp, we were able to reduce the radiation doses by more than 20% and the use of contrast medium by 60%. We suggest that low-dose scans are appropriate for lighter patients (< 70 kg) with aortic disease, and that this method is particularly valuable for follow-up studies of heavier patients (> 70 kg) with renal dysfunction. Our findings might contribute to the development of new technologies that permit reductions in radiation and contrast doses during CT angiography by varying the applied tube voltage.
Acknowledgments
We thank our radiologic technologist, Masahiro Hatemura, for his support
during the CT scanning.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A. H. de Vries, H. W. Venema, J. Florie, C. Y. Nio, and J. Stoker Influence of Tagged Fecal Material on Detectability of Colorectal Polyps at CT: Phantom Study Am. J. Roentgenol., October 1, 2008; 191(4): W181 - W189. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. T. Boll, E. M. Merkle, E. K. Paulson, R. A. Mirza, and T. R. Fleiter Calcified Vascular Plaque Specimens: Assessment with Cardiac Dual-Energy Multidetector CT in Anthropomorphically Moving Heart Phantom Radiology, October 1, 2008; 249(1): 119 - 126. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-Y. Yang, Y.-F. Chen, C.-W. Lee, A. Huang, Y. Shen, C. Wei, and H.-M. Liu Multiphase CT Angiography versus Single-Phase CT Angiography: Comparison of Image Quality and Radiation Dose AJNR Am. J. Neuroradiol., August 1, 2008; 29(7): 1288 - 1295. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. T. Boll, E. M. Merkle, E. K. Paulson, and T. R. Fleiter Coronary Stent Patency: Dual-Energy Multidetector CT Assessment in a Pilot Study with Anthropomorphic Phantom Radiology, June 1, 2008; 247(3): 687 - 695. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and 90
kVp (•) groups are -0.71 and -0.75, respectively. A correlation
coefficient of 1.0 indicates a perfect linear relationship.
