DOI: 10.1148/radiol.2461070307
(Radiology 2008;246:125-132.)
© RSNA, 2007
Hypervascular Liver Tumors: Low Tube Voltage, High Tube Current Multi–Detector Row CT for Enhanced Detection—Phantom Study1
Sebastian T. Schindera, MD,
Rendon C. Nelson, MD,
Srinivasan Mukundan, Jr, MD, PhD,
Erik K. Paulson, MD,
Tracy A. Jaffe, MD,
Chad M. Miller, MD,
David M. DeLong, PhD,
Keigo Kawaji, BS,
Terry T. Yoshizumi, PhD, and
Ehsan Samei, PhD
1 From the Department of Radiology (S.T.S., R.C.N., S.M., E.K.P., T.A.J., C.M.M., D.M.D., K.K., T.T.Y., E.S.) and Division of Radiation Safety (T.T.Y.), Duke University Medical Center, Box 3808, Erwin Rd, Durham, NC 27710; and Department of Biomedical Engineering, Duke University, Durham, NC (S.M., K.K., E.S.). Received February 14, 2007; revision requested April 24; revision received May 15; final version accepted June 25.
Address correspondence to R.C.N. (e-mail: rendon.nelson{at}duke.edu).
 |
ABSTRACT
|
|---|
Purpose: To prospectively evaluate, for the depiction of simulated hypervascular liver lesions in a phantom, the effect of a low tube voltage, high tube current computed tomographic (CT) technique on image noise, contrast-to-noise ratio (CNR), lesion conspicuity, and radiation dose.
Materials and Methods: A custom liver phantom containing 16 cylindric cavities (four cavities each of 3, 5, 8, and 15 mm in diameter) filled with various iodinated solutions to simulate hypervascular liver lesions was scanned with a 64-section multi–detector row CT scanner at 140, 120, 100, and 80 kVp, with corresponding tube current–time product settings at 225, 275, 420, and 675 mAs, respectively. The CNRs for six simulated lesions filled with different iodinated solutions were calculated. A figure of merit (FOM) for each lesion was computed as the ratio of CNR2 to effective dose (ED). Three radiologists independently graded the conspicuity of 16 simulated lesions. An anthropomorphic phantom was scanned to evaluate the ED. Statistical analysis included one-way analysis of variance.
Results: Image noise increased by 45% with the 80-kVp protocol compared with the 140-kVp protocol (P < .001). However, the lowest ED and the highest CNR were achieved with the 80-kVp protocol. The FOM results indicated that at a constant ED, a reduction of tube voltage from 140 to 120, 100, and 80 kVp increased the CNR by factors of at least 1.6, 2.4, and 3.6, respectively (P < .001). At a constant CNR, corresponding reductions in ED were by a factor of 2.5, 5.5, and 12.7, respectively (P < .001). The highest lesion conspicuity was achieved with the 80-kVp protocol.
Conclusion: The CNR of simulated hypervascular liver lesions can be substantially increased and the radiation dose reduced by using an 80-kVp, high tube current CT technique.
© RSNA, 2007
|
INTRODUCTION
|
|---|
Detection of hypervascular liver tumors at contrast material–enhanced computed tomography (CT) is directly related to tumor-to-liver contrast (1–3). The highest tumor-to-liver contrast is usually seen during the so-called late arterial phase, when tumor neovascularity maximally enhances while there is minimal enhancement of the surrounding hepatic parenchyma (4). In addition to scanning during the peak enhancement of hypervascular tumors, it is important to choose the most effective contrast medium technique to improve depiction. As demonstrated by the results of numerous investigations, the administration of contrast material with a high iodine concentration (370–400 milligrams of iodine per milliliter [mg I/mL]), a fast bolus injection rate, or both can improve substantially the tumor-to-liver contrast of hypervascular liver tumors during the late arterial phase (3,5–8). Despite these advances, contrast-enhanced CT is still considered to be relatively insensitive in depiction of small hepatocellular carcinomas (HCCs) in the setting of cirrhosis (9–11).
Iodinated contrast material demonstrates higher attenuation levels at lower x-ray tube voltages owing to a greater photoelectric effect and decreased Compton scattering (12,13). To take advantage of the inherent attenuation property of iodinated contrast material, the mean photon energy of the tube voltage has to be adjusted closer to the k edge of iodine (33.2 keV). Because the lower tube voltage CT technique comes at the cost of greater image noise, the tube current–time product has to be increased to counteract the higher image noise. With the recent advent of high-output x-ray tubes, peak tube currents of up to 800 mA are now feasible. On the basis of the increased attenuation values for iodinated contrast material at lower tube voltages and the capability of using high tube current settings, we hypothesized that the contrast-to-noise ratio (CNR) of hypervascular liver tumors scanned during the late arterial phase with multi–detector row CT can be substantially increased by using a low tube voltage and high tube current technique.
Thus, the purpose of our study was to prospectively evaluate, for the depiction of simulated hypervascular liver lesions in a phantom, the effect of a low tube voltage, high tube current CT technique on image noise, CNR, lesion conspicuity, and radiation dose.
|
MATERIALS AND METHODS
|
|---|
GE Healthcare (Milwaukee, Wis) provided financial support for purchasing a custom liver phantom. One of the authors (R.C.N.) is a consultant for GE Healthcare, the manufacturer of the CT scanner used in this study. However, the authors of this manuscript who are not consultants for GE Healthcare had complete control of any data and information that might present a conflict of interest to the author who is a consultant.
Liver Phantom
A custom liver phantom (QRM, Möhrendorf, Germany) was manufactured to mimic the liver parenchyma with parenchymal attenuation values at various tube voltages during the late arterial phase (Fig 1). Recently published attenuation values for the liver parenchyma that were acquired during the late arterial phase at 120 and 90 kVp (+92 and +103 HU, respectively) were used to design the custom attenuation specification of the phantom (13). The phantom's background measured +85.7 HU ± 1.1 (standard error of the mean) at 140 kVp, +90.8 HU ± 1.3 at 120 kVp, +105.6 HU ± 1.4 at 100 kVp, and +119.6 HU ± 1.5 at 80 kVp.
View larger version (146K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1: The custom liver phantom, fabricated of polyurethane, was placed in a plastic container filled with water. The diameter of the liver phantom was 15.0 cm. The container had the geometry of an elliptic cylinder (height, 17.0 cm; semimajor axis, 25.0 cm; semiminor axis, 22.0 cm). The phantom contained 16 cylindric cavities that were filled with iodinated solutions through the filling screws (arrow).
|
|
The cylindric liver phantom consisted of 16 cylindric cavities (Fig 1). The 16 cavities had four diameters (3, 5, 8, and 15 mm), with four cavities at each diameter. To simulate hypervascular liver tumors with various degrees of enhancement, the cavities were filled with varying concentrations of iodinated solution.
CT Scanning
The liver phantom was scanned with a 64-section multi–detector row CT scanner (LightSpeed VCT; GE Healthcare) by using four protocols (Table 1). The phantom was scanned once for each protocol. Protocol A, the baseline protocol, included the standard parameters used in our institution for abdominopelvic CT scans (140 kVp, 225 mAs). For protocols B, C, and D, the tube voltages were lowered (to 120, 100, and 80 kVp, respectively) while the tube current–time product was simultaneously increased (to 275, 420, and 675, respectively). By increasing the tube current–time product for the 120- and 100-kVp protocols, we attempted to offset the added image noise. The increases in tube current–time product were determined from published phantom data on image noise as a function of tube current–time product at different tube voltages (14). At 80 kVp, a tube current of 675 mA (the maximum allowed by the CT scanner), together with a gantry rotation time of 1.0 second, was used to minimize the increased image noise. Gantry rotation time was not lengthened more than 1.0 second because of the associated risk of missing the late arterial phase in the clinical setting. Scanner configuration (64 x 0.625 mm) and beam pitch (1.375) were kept constant for the four protocols.
For CT scanning, the liver phantom was placed within a water-filled plastic container that had the geometry of an elliptic cylinder (semiminor axis, 22.0 cm; semimajor axis, 25.0 cm) (Fig 1). The size of the container was selected to match the abdominal cross-sectional dimensions of the anthropomorphic phantom used for the radiation dose measurements in a previous study (15). The water-liver phantom was positioned within the isocenter of the scanner with its cross section perpendicular to the scanner's z-axis.
So that we could compare the overall attenuation of the water-liver phantom with the overall attenuation of a human body at the level of the upper abdomen, we assessed the square root of the attenuation projection area (SQRTPA) from the phantom's scout scan (16). The SQRTPA provides a single metric for describing the overall attenuation of a patient. The calculated SQRTPA for our water-liver phantom was 41.9. The value was compared with published population data based on 549 adult abdominal scout scans (mean SQRTPA for women, 49.7; mean SQRTPA for men, 50.3) (16). Seven abdominal scout scans with a SQRTPA of 42.0 had a mean anteroposterior diameter of 23.9 cm and a lateral diameter of 29.2 cm. Thus, the overall attenuation of our water-liver phantom was representative of that of a thin adult patient with a small abdominal girth.
Quantitative Image Assessment and Statistical Analysis
Quantitative assessment included measurements of the CNR for four simulated lesions (lesions 1–4) and the noise on images acquired with the four CT protocols. Only the four 15-mm cavities were used for CNR measurements so that there would be enough space for adequate sampling of regions of interest (ROIs). One author (S.T.S.) measured the attenuation values of the lesions and their surrounding background by placing circular ROIs. The iodinated solutions were prepared by using iodinated contrast material with a concentration of 300 mg I/mL (Isovue 300; Bracco Diagnostics, Princeton, NJ) and distilled water. The concentration of the four simulated hypervascular lesions ranged from 4.8 to 5.4 mg I/mL in increments of 0.2 mg I/mL. A preliminary CT investigation performed by our group demonstrated that a lesion filled with an iodinated solution of 4.8 mg I/mL had a mean attenuation value of +103.4 HU ± 0.9 (standard error of the mean) at 140 kVp. Because the phantom's background measured +85.7 HU ± 1.1 at the baseline tube voltage of 140 kVp, iodinated solutions with concentrations of 4.8 mg I/mL or greater simulated hypervascular lesions with a minimum attenuation difference of 17 HU.
Quantitative image assessment was performed with reconstructed 5-mm-thick transverse images at a separate workstation (Advantage Windows 4.2; GE Healthcare). The ROIs measured approximately 50 mm2 for the simulated lesions and 550 mm2 for the phantom's background. For each of the four simulated lesions and the background, nine attenuation measurements were obtained. Each of the nine measurements was obtained on a different transverse CT image. The CNR of the simulated lesions was calculated as CNR = (ROIHL – ROIB)/NoiseB, where ROIHL is the mean attenuation value of the simulated hypervascular lesion, ROIB is the mean attenuation value of the background, and NoiseB is the mean image noise. Image noise was defined as the standard deviation of the attenuation value measured in the phantom's background. In addition, a figure of merit (FOM) was calculated as the ratio of the CNR2 to effective dose for each lesion at each of the four tube voltages. This quantity was necessitated by the fact that different tube voltages imparted different effective dose, and thus the image quality could not be compared independent of dose. The FOM quantity enabled the assessment of CNR change independent of the tube current–time product and the effective dose (17).
For differences in attenuation values between the different tube voltages, a two-way analysis of variance (ANOVA) model was calculated, with contrast material concentration and tube voltage forming the two ANOVA factors. Differences in image noise were analyzed with a one-way ANOVA model, with groups defined by tube voltages. For CNR differences, the following were calculated: (a) approximate standard error of the means of the CNR, (b) propagation of error by means of an approximate linearization of the expression for CNR (delta method, which uses a Taylor-Maclaurin series of the CNR around the sample means) (18), and (c) estimates for the standard errors and covariance of the three variables. By using these standard errors, a t statistic as the ratio of the CNR difference to the estimated standard error of the difference was calculated, and associated P values were determined. For the FOM differences, the mean FOM values of the six lesions were compared by applying a mixed linear model, which allowed within-lesion correlation, to the logarithmic transformation of the original measurements. The estimates from this model were compared in a pairwise fashion with a Bonferroni adjustment for multiple comparisons. A P value of less than .05 was considered to indicate a statistically significant difference. All statistical analyses were performed with software (SAS, version 9.1.3; SAS, Cary, NC).
Assessment of Lesion Conspicuity
For the assessment of lesion conspicuity, all 16 cavities of the phantom were filled with iodinated solutions. The cavities, with diameters of 3, 5, and 8 mm, were filled with an iodinated solution of 4.0 mg I/mL. An iodinated solution of 4.0 mg I/mL yielded a mean attenuation value of +86.8 HU ± 1.8 (standard error of the mean) at 140 kVp, compared with +85.7 HU ± 1.1 for the phantom's background. Choosing an iodinated solution of 4.0 mg I/mL for the 3-, 5-, and 8-mm cavities created a worst-case lesion-to-background scenario at 140 kVp. To establish a clear contrast to the worst-case lesion-to-background scenario, the concentration for the four 15-mm cavities ranged from 4.8 to 5.4 mg I/mL, with increments of 0.2 mg I/mL.
Three radiologists (E.K.P., T.A.J., C.M.M., with 15, 4, and 2 years of experience, respectively, with post-subspecialty training in abdominal radiology) independently assessed lesion conspicuity on four transverse CT images representing the four CT protocols. The reading was performed at a workstation (Advantage Windows 4.2). The readers were blinded to the CT parameters and to the number and location of simulated lesions. Before the start of the assessment, each reader was instructed in the criteria of image grading. Readers were asked to grade conspicuity for the 16 simulated hypervascular lesions by using a three-point subjective scale (where 1 = low, 2 = medium, and 3 = high conspicuity). The readers could alternate between a soft-tissue window (width, 400 HU; level, +40 HU) and a dedicated liver window (width, 160 HU; level, +60 HU). Images were reviewed in the following order: (a) images acquired with protocol A, (b) images acquired with protocol B, (c) images acquired with protocol C, and (d) images acquired with protocol D. By starting with the image that represented the worst-case lesion-to-background scenario, we attempted to minimize memory bias for assessment of the simulated lesions.
Radiation Dose Measurement
An adult-sized female anthropomorphic phantom (Model 702; CIRS, Norfolk, Va) with 20 imbedded high-sensitivity metal-oxide-semiconductor field-effect transistors, or MOSFETs (Model TN-1002RD; Thomson-Nielsen, Ottawa, Ontario, Canada), was used to determine the effective doses for each of the four CT protocols. The phantom was scanned three times with each CT protocol. The cephalocaudal scan coverage was 254 mm. Effective doses for these protocols were computed according to Publication 60 of the International Commission on Radiological Protection (19). Effective doses and their estimated standard errors were used to estimate differences in effective dose, standard errors of the difference, t statistics, and associated P values. A P value of less than .05 was considered to indicate a statistically significant difference.
|
RESULTS
|
|---|
Quantitative Image Assessment
The mean attenuation values of the four simulated lesions increased as the tube voltage decreased from 140 to 80 kVp. At comparison of the mean attenuation values of the four lesions at 140 kVp with those at 120, 100, and 80 kVp, the percentage increases for the lesions were 17% to 20%, 51% to 56%, and 95% to 103%, respectively (P < .001). The interaction term between tube voltage and concentration in the model for attenuation was not statistically significant (P > .05). At the same time, the percentage increases in image noise between the baseline protocol A and protocols B, C, and D were 8%, 17%, and 45%, respectively (P < .05) (Table 2).
The highest CNR values for all four lesions were seen with protocol D, with 80 kVp and 675 mAs (Table 2). There was a statistically significant increase in CNR for the lesions scanned with protocol D compared with the lesions scanned with protocol A, B, or C (P < .001). The percentage increase in CNR between the baseline protocol A and protocol D ranged from 136% to 208%. The smaller the CNR value of the simulated lesion with the baseline protocol A, the higher the percentage increase in CNR with protocols B, C, and D.
The FOM for each lesion increased as the tube voltage decreased from 140 to 80 kVp, with the highest values seen at 80 kVp (P < .001) (Fig 2). At the most conservative FOM value, compared with 140 kVp, CNR increased by 1.6, 2.4, and 3.6 times at 120, 100, and 80 kVp, respectively, while the effective dose was kept constant (P < .001). Similarly, while keeping the CNR of a simulated hypervascular lesion constant, the effective dose could be reduced by 2.5, 5.5, and 12.7 times at 120, 100, and 80 kVp, respectively (P < .001).
View larger version (11K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2: Graph shows FOMs (ratio of CNR2 to effective dose) for the four simulated lesions at 80, 100, 120, and 140 kVp. The highest FOM for each lesion is seen at 80 kVp, and the lowest is seen at 140 kVp. Error bars indicate standard errors of the mean.
|
|
Assessment of Lesion Conspicuity
The highest conspicuity for all lesions was found on images acquired with protocol D (3.0 ± 0 [standard error of the mean]) (Table 3) (Fig 3). On images acquired with protocol A, none of the 12 simulated lesions with a minimal lesion-to-background contrast were detected by the three readers (Fig 3). However, these lesions were delineated with medium or high conspicuity on the images acquired with protocols B, C, and D (Fig 3). Although the 3-, 5-, and 8-mm lesions contained the same concentration of iodinated solution, the average conspicuity for the 8-mm lesions (3.0 ± 0) on the image acquired with protocol B was greater than that for the 5- and 3-mm lesions (2.7 ± 0.35 and 2.0 ± 0.64, respectively).
View larger version (111K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a: Four transverse CT images of liver phantom surrounded by water obtained with (a) protocol A (140 kVp and 225 mAs), (b) protocol B (120 kVp and 275 mAs), (c) protocol C (100 kVp and 420 mAs), and (d) protocol D (80 kVp and 675 mAs). On a, only four of the 16 simulated lesions were detected by the three readers, whereas on d, all 16 lesions could be clearly delineated.
|
|
View larger version (115K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b: Four transverse CT images of liver phantom surrounded by water obtained with (a) protocol A (140 kVp and 225 mAs), (b) protocol B (120 kVp and 275 mAs), (c) protocol C (100 kVp and 420 mAs), and (d) protocol D (80 kVp and 675 mAs). On a, only four of the 16 simulated lesions were detected by the three readers, whereas on d, all 16 lesions could be clearly delineated.
|
|
View larger version (119K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3c: Four transverse CT images of liver phantom surrounded by water obtained with (a) protocol A (140 kVp and 225 mAs), (b) protocol B (120 kVp and 275 mAs), (c) protocol C (100 kVp and 420 mAs), and (d) protocol D (80 kVp and 675 mAs). On a, only four of the 16 simulated lesions were detected by the three readers, whereas on d, all 16 lesions could be clearly delineated.
|
|
View larger version (123K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3d: Four transverse CT images of liver phantom surrounded by water obtained with (a) protocol A (140 kVp and 225 mAs), (b) protocol B (120 kVp and 275 mAs), (c) protocol C (100 kVp and 420 mAs), and (d) protocol D (80 kVp and 675 mAs). On a, only four of the 16 simulated lesions were detected by the three readers, whereas on d, all 16 lesions could be clearly delineated.
|
|
Radiation Dose Measurement
The highest effective dose was measured for protocol A (11.1 mSv ± 0.42), and the lowest was measured for protocol D (4.8 mSv ± 0.21) (Table 2). The effective dose decreased between protocols A and D by 57% (P < .001).
|
DISCUSSION
|
|---|
In our phantom study, we attempted to increase the attenuation difference between simulated hypervascular lesions and background by decreasing the tube voltage from 140 to 80 kVp. The attenuation of the simulated lesions filled with iodinated contrast material increased as the mean photon energy of the tube approximated the k edge of iodine (33.2 keV). The reported mean photon energy for a tube voltage of 80 kVp is 43.7 keV, whereas the energy levels of the tube voltages commonly used for abdominal CT (120 or 140 kVp) range between 56.8 and 61.5 keV (20). In our phantom study, the decrease in tube voltage from 140 to 80 kVp yielded a significant increase—up to 103%—in attenuation for the four simulated lesions, which is well in accordance with previously published data (21). Similar percentage increases in the attenuation of hypervascular liver tumors scanned during the late arterial phase can be expected in vivo.
The main drawback of the low tube voltage technique is the greater image noise caused by the reduced photon flux. Thus, to realize an overall CNR improvement for hypervascular liver lesions at lower tube voltages, the greater image noise has to be diminished. With the latest tube technology, sufficiently high tube currents can be applied at low tube voltages to offset the increased image noise. In our investigation, we increased the tube current for the 80-kVp protocol to 675 mA, the maximum setting allowed by the scanner at 80 kVp. In addition, the gantry rotation time was increased from 0.5 to 1.0 second. As the tube current–time product was increased threefold (protocol A vs D), and the tube voltage decreased from 140 to 80 kVp, the overall increase in image noise was limited to 45%. A further reduction in image noise at 80 kVp could be achieved by additional lengthening of the gantry rotation time. However, a drawback of a gantry rotation time longer than 1.0 second is the increased risk of missing the temporal window of the late arterial phase. A gantry rotation time of 1.0 second yields an acquisition time of 5.5 seconds for 300 mm of longitudinal coverage, which we believe is sufficiently fast to capture the entire liver during the late arterial phase.
Our FOM results clearly indicated that the CNR of simulated hypervascular liver lesions can be significantly enhanced at lower tube voltages without increasing the radiation dose to the patient. The CNR increase of the simulated lesions at 80 kVp, together with a high tube current–time product, might improve the detection of hypervascular tumors, which would be particularly advantageous for detecting HCC in the cirrhotic liver. Screening of high-risk patients with cirrhosis for HCC is challenging because of the underlying fibrosis, distorted anatomy, and inhomogeneous parenchymal enhancement characteristic of this disease. Moreover, CT is still considered to be relatively insensitive for the detection of small HCCs in patients with cirrhosis (9–11). Indeed, the detection rate of HCCs smaller than 2 cm with dual-phase helical CT scanning ranges from 50% to 61%, whereas the detection rate for lesions larger than 2 cm ranges from 82% to 98% (9–11). Our phantom study results indicated increased conspicuity for small lesions as the tube voltage decreased. Furthermore, lesions with a minimal CNR with the 140-kVp protocol that could not be delineated by the human eye could be clearly delineated with the 80-kVp protocol. On the basis of our study results, we suggest that a late arterial phase CT scan at 80 kVp with a high tube current–time product would be a promising tool for detecting even small and subtly enhancing HCCs in patients with cirrhosis. As the sensitivity for detection of those lesions increases, there is an increased risk of false-positive findings, which might decrease lesion specificity. Clearly, early diagnosis of HCCs when lesions are still subtle and small is crucial for surgical therapy, which provides the best long-term outcome.
In addition to yielding a potential increase in CNR for hypervascular tumors, a low tube voltage technique can also result in decreased radiation exposure to the patient. The FOM results demonstrated that significant dose reduction can be achieved by reducing the tube voltage without any negative impact on CNR. Young patients suspected of having hepatic neoplasms who undergo multiple contrast-enhanced CT examinations that include multiple enhancement phases would particularly benefit from the reduced radiation dose.
There were limitations to our study. First, our liver phantom was simplified to mimic the hepatic enhancement condition existing momentarily during the late arterial phase. Variation in lesion size and enhancement was simulated by filling cylindric cavities of different diameters with varying concentrations of iodinated solution. However, our liver phantom did not model heterogeneous parenchymal enhancement or distorted hepatic anatomy, both of which are often found in the setting of cirrhosis. Because parenchymal inhomogeneity may influence the overall increase of the CNR of hypervascular liver tumors at 80 kVp, it is important to confirm the results of our phantom study in vivo. At our institution, a clinical study is under way to determine whether this low tube voltage, high tube current CT technique for detection of hypervascular HCCs in patients with cirrhosis proves beneficial.
Second, our investigation did not take into account differences in body sizes. Our phantom simulated a thin adult patient with an abdominal cross-sectional diameter of approximately 24 x 29 cm, which is smaller than the abdominal cross-sectional diameter of an average adult patient (25 x 33 cm) (22). Owing to increased image noise and beam hardening, absolute CNR values are expected to be smaller in patients with an abdominal cross-sectional diameter of greater than 24 x 29 cm. Although our phantom study results suggest the use of low tube voltage for the detection of hypervascular liver tumors in thin patients, it is not clear if our technique is also applicable in larger patients. There might exist a patient size cutoff point at which the use of the low tube voltage technique is no longer appropriate. Additional research is needed to understand the effect of patient size on this practice.
Third, the grading of lesion conspicuity was restricted for protocols C and D because of the readers' knowledge of the number and location of the simulated lesions from the previous image (protocol B). Because the conspicuity of the simulated lesions for protocols C and D was so striking, we do not believe that this limitation had a major impact on the qualitative results.
In conclusion, our phantom study data suggest the implementation of a low tube voltage CT technique at 80 kVp with a high tube current–time product for the detection of hypervascular liver tumors during the late arterial phase. This technique might increase the CNR and the conspicuity of hypervascular tumors while at the same time decreasing radiation dose.
Practical application: The phantom data indicate the feasibility of a low tube voltage CT technique at 80 kVp coupled with high tube current–time products for the detection of hypervascular liver tumors during the late arterial phase. However, it remains uncertain if the technique yields similar results in vivo. Further studies need to be conducted to evaluate the potential risk of false-positive findings of hypervascular tumors at 80 kVp, a setting that may decrease lesion specificity.
|
ADVANCES IN KNOWLEDGE
|
|---|
- In a liver phantom, low tube voltage multi–detector row CT at 80 kVp coupled with a high tube current–time product yields a significant (P < .001) increase in the contrast-to-noise ratio of simulated hypervascular lesions and their conspicuity.
- As an added benefit, the low tube voltage CT technique significantly (P < .001) reduces radiation dose.
|
IMPLICATION FOR PATIENT CARE
|
|---|
- On the basis of our phantom study results, applying a low tube voltage, high tube current multi–detector row CT technique for depiction of hypervascular liver tumors during the late arterial phase might lead to increased and earlier detection of hypervascular neoplasms.
|
ACKNOWLEDGMENTS
|
|---|
The authors thank Richard Youngblood, MA, for editing the manuscript and Greta Toncheva, MS, and Giao Nguyen, MS, for performing the radiation dose measurements.
|
FOOTNOTES
|
|---|
Abbreviations: CNR = contrast-to-noise ratio • FOM = figure of merit • HCC = hepatocellular carcinoma • ROI = region of interest
Guarantors of integrity of entire study, S.T.S., R.C.N., E.S.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, S.T.S.; experimental studies, S.T.S., S.M., K.K., T.T.Y.; statistical analysis, D.M.D.; and manuscript editing, all authors
|
References
|
|---|
- Oliver JH 3rd, Baron RL. Helical biphasic contrast-enhanced CT of the liver: technique, indications, interpretation, and pitfalls. Radiology 1996;201:1–14. [Abstract/Free Full Text]
- Funama Y, Awai K, Nakayama Y, et al. Radiation dose reduction without degradation of low-contrast detectability at abdominal multisection CT with a low-tube voltage technique: phantom study. Radiology 2005;237:905–910. [Abstract/Free Full Text]
- Awai K, Takada K, Onishi H, Hori S. Aortic and hepatic enhancement and tumor-to-liver contrast: analysis of the effect of different concentrations of contrast material at multi-detector row helical CT. Radiology 2002;224:757–763. [Abstract/Free Full Text]
- Foley WD, Mallisee TA, Hohenwalter MD, Wilson CR, Quiroz FA, Taylor AJ. Multiphase hepatic CT with a multirow detector CT scanner. AJR Am J Roentgenol 2000;175:679–685. [Abstract/Free Full Text]
- Sultana S, Morishita S, Awai K, et al. Evaluation of hypervascular hepatocellular carcinoma in cirrhotic liver by means of helical CT: comparison of different contrast medium concentrations within the same patient. Radiat Med 2003;21:239–245. [Medline]
- Marchiano A, Spreafico C, Lanocita R, et al. Does iodine concentration affect the diagnostic efficacy of biphasic spiral CT in patients with hepatocellular carcinoma? Abdom Imaging 2005;30:274–280.
- Itoh S, Ikeda M, Achiwa M, Satake H, Ota T, Ishigaki T. Multiphase contrast-enhanced CT of the liver with a multislice CT scanner: effects of iodine concentration and delivery rate. Radiat Med 2005;23:61–69. [Medline]
- Awai K, Inoue M, Yagyu Y, et al. Moderate versus high concentration of contrast material for aortic and hepatic enhancement and tumor-to-liver contrast at multi-detector row CT. Radiology 2004;233:682–688. [Abstract/Free Full Text]
- Valls C, Cos M, Figueras J, et al. Pretransplantation diagnosis and staging of hepatocellular carcinoma in patients with cirrhosis: value of dual-phase helical CT. AJR Am J Roentgenol 2004;182:1011–1017. [Abstract/Free Full Text]
- Noguchi Y, Murakami T, Kim T, et al. Detection of hepatocellular carcinoma: comparison of dynamic MR imaging with dynamic double arterial phase helical CT. AJR Am J Roentgenol 2003;180:455–460. [Abstract/Free Full Text]
- Lim JH, Kim CK, Lee WJ, et al. Detection of hepatocellular carcinomas and dysplastic nodules in cirrhotic livers: accuracy of helical CT in transplant patients. AJR Am J Roentgenol 2000;175:693–698. [Abstract/Free Full Text]
- Bahner ML, Bengel A, Brix G, Zuna I, Kauczor HU, Delorme S. Improved vascular opacification in cerebral computed tomography angiography with 80 kVp. Invest Radiol 2005;40:229–234. [CrossRef][Medline]
- Nakayama Y, Awai K, Funama Y, et al. Abdominal CT with low tube voltage: preliminary observations about radiation dose, contrast enhancement, image quality, and noise. Radiology 2005;237:945–951. [Abstract/Free Full Text]
- Boone JM, Geraghty EM, Seibert JA, Wootton-Gorges SL. Dose reduction in pediatric CT: a rational approach. Radiology 2003;228:352–360. [Abstract/Free Full Text]
- Hurwitz LM, Yoshizumi T, Reiman RE, et al. Radiation dose to the fetus from body MDCT during early gestation. AJR Am J Roentgenol 2006;186:871–876. [Abstract/Free Full Text]
- Toth T, Ge Z, Daly M. The influence of bowtie filter selection, patient size and patient centering on CT dose and image noise. Med Phys 2006;33:2006.
- Samei E, Dobbins JT 3rd, Lo JY, Tornai MP. A framework for optimising the radiographic technique in digital X-ray imaging. Radiat Prot Dosimetry 2005;114:220–229. [Abstract/Free Full Text]
- Rao CR. Linear interference and its applications. New York, NY: Wiley, 1972; 387–398.
- International Commission on Radiological Protection. Recommendations of the International Commission on Radiological Protection. ICRP Publication 60. Oxford, England: Pergamon, 1991.
- Huda W, Scalzetti EM, Levin G. Technique factors and image quality as functions of patient weight at abdominal CT. Radiology 2000;217:430–435. [Abstract/Free Full Text]
- Kalva SP, Sahani DV, Hahn PF, Saini S. Using the K-edge to improve contrast conspicuity and to lower radiation dose with a 16-MDCT: a phantom and human study. J Comput Assist Tomogr 2006;30:391–397. [CrossRef][Medline]
- Kalra MK, Maher MM, Kamath RS, et al. Sixteen-detector row CT of abdomen and pelvis: study for optimization of Z-axis modulation technique performed in 153 patients. Radiology 2004;233:241–249. [Abstract/Free Full Text]
TOP
ABSTRACT
INTRODUCTION
