| 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 Sciences, School of Health Sciences, Kumamoto
University, Kumamoto, Japan.
Received September 9, 2007;
accepted after revision October 17, 2007.
Address correspondence to Y. Yanaga.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. The study included 135 patients with known or suspected HCC who underwent dynamic CT on a 64-detector scanner and 47 were found to have 71 hypervascular HCCs. The patients were randomly assigned to one of three protocols: A contrast dose of 450, 525, or 600 mg I/kg of body weight was delivered over 30 seconds in protocols A, B, and C, respectively. We measured the tumor–liver contrast (TLC) during HAP in the three groups and compared the results. Two radiologists qualitatively evaluated tumor conspicuity during HAP using a 3-point scale; their results were compared.
RESULTS. The TLC in protocols A, B, and C was 26.5, 38.4, and 52.3 H, respectively; the difference was significant between protocols A and B (p = 0.05), A and C (p < 0.01), and B and C (p = 0.02). In our qualitative analysis of tumor conspicuity, the mean score for protocols A, B, and C was 1.6, 2.3, and 2.7, respectively; there was a significant difference between protocols A and B and A and C, but not between protocols B and C.
CONCLUSION. The administration of a total iodine dose of 525 mg or more per kilogram of body weight is desirable for the good or excellent depiction of hypervascular HCC, although the administration of 450 mg I/kg of body weight can depict hypervascular HCC.
Keywords: dynamic CT • hepatic hemodynamics • hepatocellular carcinoma • liver disease • MDCT
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Because the major role of hepatic dynamic CT that includes the HAP is the detection of hypervascular hepatic tumors, the conspicuity of liver tumors during the HAP, which can be expressed by the attenuation difference between the tumor and the hepatic parenchyma [23], is important. Therefore, we investigated the contrast dose necessary for optimal tumor conspicuity on hepatic dynamic CT scans. The purpose of our study was to examine the effect of contrast dose on the conspicuity of hypervascular HCC tumors during HAP at hepatic dynamic CT using a 64-detector CT scanner.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Patients and Tumors
Between July and December 2006, we enrolled 145 patients who met our
inclusion criteria. These criteria were as follows: a diagnosis of type B,
type C, or alcoholic hepatitis; confirmed HCC untreated during the 3 months
preceding the current CT study, the presence of a suspected space-occupying
hepatic lesion on the basis of sonographic study, or elevated levels of tumor
markers (-fetoprotein, protein induced by the absence of vitamin K, or
antagonist-II); and the absence of both renal failure (serum creatinine <
1.5 mg/dL) and a contraindication for iodinated contrast material.
Ten of the 145 patients were subsequently excluded, three because of tumor thrombi in the central portal veins; four because numerous tumors involving the entire liver may have changed hepatic hemodynamics; and three because they harbored tumors larger than 5.0 cm, which may also have altered hepatic hemodynamics. Thus, the final study population consisted of 135 patients, 97 men and 38 women who ranged in age from 35 to 85 years (mean, 66.1 years).
Among the men, the age range was 35–85 years (mean, 65.4 years); among the women, it was 39–82 years (mean, 67.9 years). There was no significant age difference between the male and female patients (p = 0.17, two-tailed Student's t test). The mean body weight for all patients was 58.7 ± 9.2 kg (SD) and the range was 34–85 kg.
Because the mean attenuation value of normal liver parenchyma is approximately 60 H on unenhanced images and approximately 80 H during the HAP [11], we defined hypervascular tumors as those with an attenuation value during HAP that was 25 H greater than on unenhanced scans. Of the 135 patients, 47 (39 men, eight women; age range, 48–85 years; mean age, 70.2 years) had solitary or multiple hypervascular HCC nodules (n = 75). The definitive diagnosis of hypervascular HCC was based on the following findings: histopathologic evidence after hepatic surgery (n = 13); needle biopsy results (n = 10); or substantially increased levels of -fetoprotein, protein induced by vitamin K absence, or antagonist-II with follow-up dynamic CT showing an increase in tumor size within 3 months (n = 24). All patients who had histologic evidence and 20 of 24 (83%) without histologically confirmed hypervascular HCC patients who did not have histologic evidence underwent CT during hepatic arteriography. Two board-certified radiologists with 21 and 9 years of experience with liver CT, respectively, consensually confirmed the presence of hypervascular tumors on CT during hepatic arteriography in those patients. At our institution, we routinely subject all patients with HCCs to CT during hepatic arteriography and CT portography before starting treatments.
CT Scanning and Contrast Material Infusion Protocols
All patients were scanned with a 64-MDCT scanner (Brilliance-64, Philips
Medical Systems) at the following settings: rotation time, 0.5 second; beam
collimation, 64 x 0.625 mm; section thickness and intervals, 5.0 mm;
helical pitch (beam pitch), 0.798; table movement, 63.8 mm/s, scanning field
of view (FOV), 40 cm; voltage, 120 kV; and tube current, 250–300 mAs.
Image reconstruction was performed in a 25- to 35-cm display FOV depending on
the patient's physique. All helical studies were started at the top of the
liver and proceeded in a cephalocaudal direction; unenhanced and three-phase
contrast-enhanced helical scans of the entire liver were obtained. Patients
were instructed to hold their breath with tidal inspiration during
scanning.
Three-phase contrast-enhanced CT of the liver was performed during the HAP, PVP, and equilibrium phases. An automatic bolus-tracking program (Bolus Pro Ultra, Philips Medical Systems) was used to time the start of scanning for each phase after contrast injection. The attenuation value was monitored by three radiology technologists with 3, 12, and 22 years of experi ence, respectively, with abdominal CT. Monitoring was at the L1 vertebral body level; the region-of-interest (ROI) cursor (0.8–2.0 cm2) was placed in the abdominal aorta. Real-time low-dose (120 kVp, 15 mAs) serial monitoring studies began 8 seconds after the start of contrast injection. The trigger threshold level was set at 150 H. During the real-time low-dose serial monitoring studies, patients were instructed to take shallow, regular breaths. The mean (± SD) trigger time in protocols A, B, and C was 19.8 ± 4.2 seconds (range, 13–29 seconds), 18.8 ± 2.5 seconds (range, 14–26 seconds), and 18.2 ± 3.0 seconds (range, 13–26 seconds), respectively.
HAP, PVP, and equilibrium phase scanning was started 18, 40, and 160 seconds, respectively, after triggering. We based our selection of the scanning delay for HAP on a recent article that described the optimal scanning delay for hypervascular HCCs during HAP at hepatic dynamic CT [10]. Consequently, the mean scanning times for HAP and PVP were 37.8 seconds (range, 31–47 seconds) and 59.8 seconds (range, 53–69 seconds) in protocol A; 36.8 seconds (range, 32–44 seconds) and 58.8 seconds (range, 54–66 seconds) in protocol B; and 36.2 seconds (range, 31–44 seconds) and 58.2 seconds (range, 53–66 seconds) in protocol C.
Patients were randomly assigned to one of three contrast injection protocols; the contrast dose was 450 mg I/kg of patient body weight in protocol A, 525 mg I/kg in protocol B, and 600 mg I/kg in protocol C. Yamashita et al. [22] administered 1.5, 2.0, or 2.5 mL/kg or a fixed dose (100 mL) of iopamidol 300 to patients who underwent hepatic dynamic CT and performed quantitative and qualitative evaluations of the enhancement of the aorta, portal vein, liver, and pancreas. They found that arterial enhancement did not differ between the 2.0 and 2.5 mL/kg groups. Contrast doses of 1.5 and 2.0 mL/kg correspond to 450 and 600 mg I/kg of patient body weight, respectively.
The contrast doses used in this study were based on the results of Yamashita et al. [22]. The injection duration was 30 seconds in all protocols. Consequently, the contrast injection rate was different in each patient. The mean injection rate in protocols A, B, and C was 3.0 mL/s (range, 2.1–3.9 mL/s), 3.4 mL/s (range, 2.5–4.4 mL/s), and 3.9 mL/s (range, 2.3–5.3 mL/s), respectively. The mean injection volume in protocols A, B, and C was 88.8 mL (range, 63–117 mL), 101.9 mL (range, 75.3–131.3 mL), and 117.4 mL (range, 68–160 mL), respectively.
Among the three patient groups, there were no significant differences with respect to patient age or weight (p = 0.45 and 0.89, respectively, by one-way analysis of variance [ANOVA]), sex dis tribution, the number of patients with HCCs (p = 0.46 and 0.99, respectively, by the chi-square test), or the size of the HCC nodules (p = 0.21, by one-way ANOVA) (Table 1). Iohexol (Omnipague, Daiichi Pharmaceutical) with an iodine concentration of 300 mg/mL was administered using a power injector (Dual Shot, Nemoto Kyorindo) and a 20-gauge IV catheter inserted into an antecubital vein. The delivery of contrast material was followed by flushing with 30 mL of physiologic saline administered at the same injection rate.
|
Quantitative Analysis
One radiologist with 10 years of experience with liver CT who was blinded
to which protocol had been used measured the mean attenuation values of the
abdominal aorta and hepatic parenchyma with a circular ROI cursor. Attempts
were made to maintain a constant ROI area of approximately 1.0 cm2; the range
of the ROI areas was 0.5–1.3 cm2. Aortic atten uation values
were determined on three consecutive images at the level of the main portal
vein during unenhanced and HAP scanning in all 135 patients. The measured
attenuation values obtained for each phase were averaged. Contrast enhancement
in the ab dominal aorta during HAP was calculated as the absolute difference
in aortic attenuation (in Houns field units) between unenhanced and HAP
scans.
|
The conspicuity of a hepatic tumor can be expressed by the attenuation difference between it and the hepatic parenchyma, the so-called "tumor–liver contrast" (TLC) [23]. We defined the TLC as the result obtained by subtracting the attenuation value of the parenchyma from the attenuation value of the hepatic tumor. In the 47 patients with hypervascular HCC, the same radiologist also determined the TLC for each phase of contrast-enhanced scanning. Tumor attenuation was assessed in the most enhanced portion of the tumor. An attempt was made to maintain an ROI area of approximately 0.5 cm2; the range of the ROI areas was 0.3–0.5 cm2. To obtain the parenchymal attenuation value used for TLC calculations, we measured the normal hepatic parenchyma at least 1 cm from the edge of the tumor to nullify the risk of encountering fibrosis [24]. An attempt was made to maintain a constant ROI area of approximately 2 cm2; the range of ROI areas was 0.8–2.0 cm2. In patients with fewer than three tumors, we calculated the mean TLC for all tumors; in patients with three or more tumors, we used the averaged TLC of the three largest tumors.
|
|
Statistical Analysis
Contrast enhancement values in the aorta and hepatic parenchyma and the TLC
values are reported as means ± SD.
We used one-way ANOVA to investigate intergroup differences among the three protocols with respect to aortic, hepatic, and portal vein enhancement and TLC values. When the overall differences were statistically significant, a post-hoc analysis was performed using the Tukey-Kramer test for multiple comparisons among the three protocols.
The Kruskal-Wallis test was applied to examine intergroup differences for tumor conspicuity among the three protocols. When the overall differences were statistically significant, a post-hoc analysis was performed using the Steel-Dwass test for multiple comparisons among the three protocols.
To estimate the 95% CI of the population mean for TLC during HAP for each visual grade, we applied t statistics. Interobserver variability between the two radiologists in their visual analysis of tumor conspicuity was assessed by the test of concordance to measure their degree of agreement. The scale for the kappa coefficients for interobserver agreement was as follows: less than 0.20, poor; 0.21–0.40, fair; 0.41–0.60, moderate; 0.61–0.80, substantial; and 0.81–1.00, almost perfect [25].
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The mean values for hepatic enhancement were 15.9 ± 8.1, 17.1 ± 8.5, and 17.6 ± 9.3 H, respectively, for protocols A, B, and C, and there was no overall difference among the three protocols (p = 0.63).
The mean values for TLC in protocols A, B, and C were 26.5 ± 8.3, 38.4 ± 8.6, and 52.3 ± 20.3 H, respectively, and the overall difference among the three groups was statistically significant (p < 0.01). We found a significant difference between protocols A and B (p = 0.05), A and C (p < 0.01), and B and C (p = 0.02) (Fig. 2).
Hepatic Enhancement and TLC During PVP
The mean values for hepatic enhancement were 43.2 ± 8.7, 48.7
± 9.1, and 53.9 ± 10.2 H, respectively, for protocols A, B, and
C, and the overall difference among the three groups was statistically
significant (p < 0.01). There were statistically significant differences
between protocols A and B (p = 0.02), A and C (p < 0.01), and B and C (p =
0.02) (Fig. 3).
The mean values for TLC in protocols A, B, and C were –1.5 ± 7.2, –2.3 ± 12.9, and –3.9 ± 14.0 H, respectively; the overall difference among the three groups was not statistically significant (p = 0.85).
Results of Visual Analysis of Tumor Conspicuity During HAP
Based on our quantitative analysis of tumor conspicuity during HAP, of the
47 hypervascular tumors, 10 were categorized as grade 1, 17 as grade 2, and 20
as grade 3. The mean TLC was 21.3 ± 5.3, 35.6 ± 2.7, and 53.2
± 16.7 H for grades 1, 2, and 3, respectively
(Table 2). The mean score for
tumor conspicuity in protocols A, B, and C was 1.6, 2.3, and 2.7,
respectively, and the overall difference among the three groups was
statistically significant (p < 0.01). Although there was no statistically
significant difference between protocols B and C (p = 0.22), there was a
significant difference between protocols A and B (p = 0.02) and A and C (p
< 0.01) (Fig. 4).
|
An HCC conspicuity score of 2 or 3 was assigned to eight of 16 HCCs (50.0%)
in protocol A, 13 of 15 HCCs (86.7%) in protocol B, and all 16 HCCs (100%) in
protocol C. Lastly, there was good interobserver agreement for the degree of
tumor conspicuity (
= 0.73).
|
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In previously reported hepatic dynamic CT studies, the contrast injection protocols involved fixed doses [5, 8, 27–30] or doses adjusted to patient weight [3, 6, 7, 9, 14–19, 31]. However, which type of contrast injection protocol is best for hepatic dynamic CT remains undetermined. Because hypervascular HCCs are usually fed by the hepatic artery, one of the branches from the abdominal aorta, theoretically, their enhancement is correlated with aortic enhancement during HAP. In fact, Yanaga et al. [32] recently reported that the TLC of hypervascular HCCs is almost proportional to aortic enhancement. On the other hand, in phantom experiments, Awai et al. [33] observed that at a given injection duration, aortic enhancement increased linearly with the administered contrast dose. Thus, we con cluded that contrast dose should be determined according to patient weight at hepatic dynamic CT for visualization of hypervascular HCCs. Yamashita et al. [22], who performed a prospective randomized study to evaluate the optimal contrast dose to achieve adequate aortic and pancreatic enhancement at ab dominal dynamic CT, also suggested that contrast dose be tailored to patient weight.
PVP images are important for the detection of hypovascular HCCs because these lesions may be best detected during that phase [1]. Hepatic enhancement of 50 H requires 521 mg I/kg of body weight [34]; this iodine dose corresponds to 1.73 mL/kg of contrast material with an iodine concentration of 300 mg I/mL and is very similar to the contrast dose in protocol B (1.75 mL/kg). The mean hepatic enhancement in protocols B and C was 48.7 and 53.9 H, respectively. Because this level of hepatic enhancement sufficed for diagnosing hypovascular hepatic tumors, the contrast dose in protocols B and C might be adequate for the diagnosis of these tumors during PVP, although we did not validate whether a contrast dose of 525 mg or more is appropriate for detection of hypovascular tumors in this study. However, the PVP is not useful for visualization of hypervascular HCC because TLC is minimized during that phase [2, 3]. In fact, the attenuation values of the tumors and of the liver parenchyma during PVP differed by less than 5 H in all the protocols examined in the current study.
In general, aortic enhancement increases as the injection rate increases at given contrast doses [12]. In our study, we determined contrast dose according to the patient weight and adopted a fixed injection duration (30 seconds). Therefore, the injection rate varied even among individuals treated under the same contrast injection protocol. Awai et al. [11] reported that variations in aortic peak times and aortic peak enhancement values were reduced with injection protocols that used fixed duration times and adjusted the contrast dose to the patient's weight. Because hypervascular HCCs are usually fed by branches from the aorta, theoretically, their enhancement may be almost constant in contrast injection protocols with fixed duration and contrast dose adjusted to patient weight. We believe that contrast injection protocols with a fixed duration and contrast dose adjusted to patient weight, which we adopted in our study, are effective for evaluation of the effect of contrast dose on the depiction of hypervascular HCCs.
The results presented here apply only to hepatic dynamic CT with an injection duration of 30 seconds. If the contrast dose is constant, aortic enhancement increases as the injection duration becomes shorter—in other words, the injection rate increases [12]. Therefore, lower contrast doses delivered at shorter injection duration may yield aortic enhancement and tumor conspicuity similar to those in the current study. On the other hand, the rate of increase of hepatic peak enhancement is very small at injection rates greater than 2.0 mL/s, although hepatic peak enhancement increases with the injection rate at a given contrast dose [12]. Consequently, the delivery of lower contrast doses at higher injection rates—that is, shorter injection duration—may not yield sufficient peak hepatic enhancement during PVP. According to Heiken et al. [34], peak hepatic enhancement increases with the contrast dose at a given injection rate and patient body weight. Therefore, satisfactory aortic and hepatic enhancement may require a sufficient contrast dose and an optimized injection duration. We suggest that our protocols B and C can achieve sufficient enhancement in both the aorta and liver.
We believe that the contrast dose for hepatic dynamic CT investigated in this study can be applied to single-detector CT to 4-MDCT. However, scanning timing during the HAP should be changed when single-detector CT or 4-MDCT is used for the CT examination. With 64-MDCT, the liver can be scanned in about 2 seconds; thus, the center of the scanning period is about 39 (38 + [2 / 2]) seconds after contrast injection start [10]. In single-detector CT to 4-MDCT, we should set the scanning delay so that the center of the scanning period is 39 seconds after injection start. For example, when the scanning period is 20 seconds, we should set the scanning delay at 29 (39 – [20 / 2]) seconds.
There are potential limitations in our study. First, because the range of body weights and mean body weights of our patients were lower than in patients from North America and Europe, the applicability of our results to populations of greater body weight must be confirmed. Second, there was no histo pathologic confirmation of HCC in approx imately half of our patients. Nonetheless, because 43 of 47 HCC patients underwent CT during hepatic arteriography, we think that the presence and location of HCC were confirmed in more than 90% of our patients. Although receiver operating characteristic (ROC) analysis [35] should be performed to establish the detectability of HCC under each protocol, ROC analysis requires complete confirmation of HCC. Third, we adopted contrast injection protocols in which the dose was adjusted to the patient's body weight and the injection duration was fixed. With these protocols, the injection rate is high in heavy patients. An upper-limit injection rate may need to be set when our contrast injection protocol is used in routine clinical practice. Finally, because our patient population manifested type B, type C, or alcoholic hepatitis, the results from our study cannot be applied directly to patients free of liver damage.
In conclusion, when the injection duration is 30 seconds, a total iodine dose of 525 mg or more per kilogram of patient body weight is desirable for the good or excellent depiction of hypervascular HCC during HAP, although the administration of 450 mg I/kg of body weight can depict hypervascular HCC.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
