| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 All authors: Department of Radiology, Osaka University Graduate School of Medicine, 2-2 Yamadaoka, Suita, Osaka 5650871, Japan.
Received February 18, 2002;
accepted after revision July 16, 2002.
Address correspondence to T. Murakami.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Fifty-three patients with 103 hypervascular hepatocellular carcinoma nodules who underwent both dynamic MR imaging with 3D Fourier transformation-enhanced fast gradient-echo sequences with a special spectral inversion recovery pulse and dynamic helical CT with double arterial phase imaging were enrolled in the study. For dynamic MR imaging, unenhanced, arterial, portal venous, and equilibrium phase images were obtained before and approximately 19, 60, and 120 sec, respectively, after injection of gadopentetate dimeglumine. Three observers independently interpreted the images obtained with each technique in a blinded manner and in random order.
RESULTS. Mean sensitivity and positive predictive values of CT for hypervascular hepatocellular carcinoma (66% and 97%, respectively) were higher than those of MR imaging (63% and 96%, respectively), but there was no significant difference in detecting sensitivity among the observers (p < 0.05). CT and MR imaging were complementary, with some tumors undetected by CT but revealed on MR imaging. There was also no significant difference in Az values between CT (0.74) and MR imaging (0.71) (p < 0.05).
CONCLUSION. Dynamic MR imaging with 3D Fourier transformation-enhanced fast gradient-echo sequences with a special spectral inversion recovery pulse is recommended to improve the detection of hypervascular hepatocellular carcinoma nodules in addition to the use of dynamic helical CT with double arterial phase imaging.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Three-dimensional (3D) Fourier transformation-enhanced fast gradient-echo sequences with a special spectral inversion recovery pulse were developed to allow high-quality MR angiography during a breath-hold of 20-30 sec in combination with the zero-filling interpolation technique [13]. Because the special spectral inversion recovery pulse can suppress only the fat signal, 3D Fourier transformation-enhanced fast gradient-echo sequences with a special spectral inversion recovery pulse are expected to suppress background hepatic tissues and show the hypervascular hepatocellular carcinoma with markedly high signal intensity after the administration of contrast media on optimal arterial phase MR imaging of the entire liver [14]. The result is therefore a loss of signal intensity from the liver that increases the lesion-to-liver contrast. On the other hand, a recently developed 0.5-sec (0.5 sec/tube rotation) helical CT scanner has made it possible to scan the entire liver twice in the arterial dominant phase during a single breath-hold (double arterial phase scan). Double arterial phase imaging reportedly features better detection of hypervascular hepatocellular carcinoma than does single arterial phase imaging because its higher temporal resolution can use optimal scanning time [15].
In this study, we evaluated the efficacy of dynamic MR imaging with 3D Fourier transformation-enhanced fast gradient-echo sequences with a special spectral inversion recovery pulse for the detection of hypervascular hepatocellular carcinoma in comparison with that of dynamic helical CT with double arterial phase imaging.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Hypervascularity was defined as focal lesion hyperattenuation relative to the surrounding liver parenchyma observed on either dynamic helical CT or dynamic MR imaging. Proof of hypervascular hepatocellular carcinoma was obtained by surgical resection of 39 lesions in 28 patients. The other lesions, which were not surgically treated, were confirmed on the basis of a combination of clinical and radiologic criteria, including a response to transcatheter arterial chemoembolization, focal retention of angiographically administered iodized oil, or progression or regression in size. All 53 patients underwent confirmatory imaging studies within 1 month of the dynamic MR imaging and dynamic helical CT examinations in the form of CT hepatic arteriography and CT during arterial portography performed at the time of angiography with a previously described technique [17]. In addition, the patients underwent follow-up CT examinations more than 6 months later.
MR Imaging Technique
MR images were acquired with a 1.5-T MR imager (Signa Horizon; General
Electric Medical Systems, Milwaukee, WI). MR imaging of the entire liver was
performed with 3D Fourier transformation-enhanced fast gradient-echo sequences
with a special spectral inversion recovery pulse during a single breath-hold.
Parameters were section thickness, 5 mm; no interscan gap; TRrange/TErange,
6-6.2/1.3-1.4; inversion time, 20-21 msec; flip angle, 40°; field of view,
32 cm; image matrix, 128 x 256; bandwidth, 62.5 kHz; and excitation, 1.
For the pulse sequence, the fat signal was suppressed with a spectral
selective inversion radiofrequency pulse that selectively inverted the fat
resonance.
After the unenhanced images had been obtained, dynamic MR imaging was performed with IV administration of gadopentetate dimeglumine (Magnevist; Japan Schering, Osaka, Japan) using a power injector. For dynamic MR imaging, unenhanced, arterial phase, portal venous phase, and equilibrium phase images were obtained after injection of gadopentetate dimeglumine. From 24 to 36 slices were obtained for each phase during a single breath-hold of 17-30 sec. The images were obtained during end expiration in all cases, and no presaturation pulses were used.
To determine the scanning delay of arterial phase imaging, we obtained a single axial image for the abdominal aorta of the upper abdomen. This procedure was repeated once per second for 45 sec after the administration of 2 mL of a test bolus of gadopentetate dimeglumine at a rate of 2 mL/sec through a 20-gauge plastic IV catheter placed in an antecubital vein. This procedure was followed by a 20-mL saline flush at a rate of 2 mL/sec. A cursor indicating the region of interest (ROI) was placed over the abdominal aorta, and the time to reach peak aortic enhancement was used as the scanning delay for the early arterial phase images. After the circulation time was determined, 0.1 mmol/kg of gadopentetate dimeglumine was administered at a rate of 2 mL/sec through the same route, followed by a 20-mL saline flush. We used 2 mL of gadopentetate dimeglumine at a rate of 2 mL/sec (injection time, 1 sec) for the test bolus to determine the scanning delay and approximately 10-18 mL of gadopentetate dimeglumine for the dynamic MR imaging study (injection time, 5-9 sec). Thus, true peak enhancement of the aorta occurs approximately 4-8 sec after the measured scanning delay. Peak tumor enhancement after hepatic arterial enhancement is expected to occur approximately 6 sec after peak enhancement of the aorta, that is, approximately 10-14 sec after the measured scanning delay. The k-space center of 3D Fourier transformation-enhanced fast gradient-echo sequences with a special spectral inversion recovery pulse was in the middle of the imaging time. Because the scanning time of the pulse sequence was approximately 17-30 sec, data acquisition of the k-space center commenced approximately 8.5-15 sec after initiation of the scan, which almost corresponded to peak tumor enhancement (10-14 sec after the measured scan delay). The mean scanning delay of the arterial phase was approximately 19 sec (range, 12-31 sec). Portal venous phase imaging and equilibrium phase imaging were performed 60 and 120 sec, respectively, after injection of the contrast medium.
CT Technique
CT examinations were performed with a single-detector helical CT unit with
a 0.5-sec tube rotation (Aquilion; Toshiba Medical, Tokyo, Japan). After
obtaining the unenhanced images through the liver with a 7-mm collimation,
double arterial (early and late arterial) phase and portal phase images were
acquired with a collimation of 5 mm, pitch of 1.5, reconstruction interval of
5 mm, and tube current of 300 mA.
All the patients received 100 mL of low-osmolarity contrast medium ([iopamidol] Isovue, 300mg I/mL; Nihon Schering, Osaka, Japan) administered with a power injector (Multilevel CT; Medrad, Pittsburgh, PA). The contrast medium was injected at a rate of 5 mL/sec through a 20-gauge plastic IV catheter placed in an antecubital vein. An automatic bolus-tracking program (RealPrep; Toshiba Medical) was used to automatically start the early arterial phase scan after injection of the contrast material. This technique is capable of real-time monitoring, automatic calculation of CT values in the ROI, and automatic initiation of a diagnostic CT scan after the CT value of the ROI has reached a trigger threshold level after injection of the contrast material. The anatomic level for monitoring was set just above the diaphragmatic dome, at the same level as the starting position of the diagnostic scan, and the ROI cursor was placed in the aorta. Real-time low-dose (120 KVp, 50 mA) serial monitoring scans were initiated 10 sec after injection of the contrast material. The level of the trigger threshold was set at an increase of 50 H over the baseline for the aortic ROI. Ten seconds after the trigger threshold had been reached, the early arterial phase helical CT scan started automatically. With our CT system, 10 sec was the shortest possible interval between triggering and the initiation of the diagnostic scan. The late arterial phase helical CT scan was initiated 5 sec after the end of the early arterial phase scan; this interval between the early and the late arterial phase scans was also the shortest possible. The early and late arterial phase scans were obtained during a single breath-hold of 25 sec. The portal venous and equilibrium phase CT scans were obtained 15 and 120 sec after the end of the late arterial phase scan, respectively.
Imaging Assessment
The dynamic MR images and the dynamic helical CT scans were interpreted
separately and independently by three experienced abdominal radiologists. The
portal venous or equilibrium phase images were evaluated together with the
arterial phase images. It was difficult to detect the hypervascular tumors on
these phase images because the liver parenchyma showed almost the same
enhancement as the tumor. However, we believe that multidetector images are
usually the most effective for additional characterization of liver tumors and
minimally false-positive lesions. The three observers knew that the patients
were at risk for hepatocellular carcinoma but were unaware of the presence or
location of liver lesions. The interval between the blinded interpretations
was at least 2 weeks. Each observer recorded the size of the focal hepatic
lesions and assigned the following confidence levels to his or her
observations: 1, probably absent; 2, equivocal; 3, probably present; and 4,
definitely present.
The alternative free-response receiver operating characteristic (ROC) curve analysis was used for each imaging technique on a tumor by tumor basis. Although the conventional ROC method allows only one response per image, the alternative free-response ROC method enables the observer to analyze the response for all the lesions present, and all 103 lesions were analyzed in this study [18]. An alternative free-response ROC curve was fitted to each observer's confidence rating using the maximum-likelihood estimation. The diagnostic accuracy of each imaging technique as rated by each of the observers and the corresponding composite data were estimated by calculating the area under the ROC curve (Az). Differences between the imaging techniques in terms of the mean Az values were analyzed statistically by means of the two-tailed Student's t test for paired data. A two-tailed p value of less than 0.05 was considered significant.
Those lesions among the 103 proven hepatocellular carcinomas that were assigned a confidence level of 3 or 4 were considered true-positive findings. A lesion not assigned a level or assigned a confidence level of 1 or 2 when a lesion was actually present was considered a false-negative lesion. Before interpreting the images, the three observers had been informed that the categorization of a confidence level of 3 or greater represented a positive diagnosis of hepatocellular carcinoma. The degree of disagreement was not factored into the calculation. The sensitivity and positive predictive values for dynamic MR imaging and dynamic helical CT were then calculated. The sensitivity of each imaging technique was compared using the McNemar test for individual observers. A two-tailed p value of less than 0.05 was considered significant.
To assess interobserver variability, we calculated the kappa statistic for multiple observers using the nonweighted binary kappa statistic. A kappa value of 0.01-0.20 was considered in slight agreement; 0.21-0.40, fair; 0.41-0.60, moderate; 0.61-0.80, substantial; and 0.81-1.0, almost perfect.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
The detection sensitivity for tumors of two size categories (<2 cm or
2 cm) and the positive predictive values for each of the three observers
are shown in Table 2. Using
dynamic MR imaging alone, observers 1, 2, and 3 detected 57 tumors in 45
patients, 66 tumors in 49 patients, and 73 tumors in 51 patients, respectively
(Fig.
1A,1B,1C).
Using dynamic helical CT alone, they detected 63 tumors in 46 patients, 71
tumors in 49 patients, and 70 tumors in 49 patients, respectively. Mean
sensitivity and positive predictive values of dynamic helical CT for
hypervascular hepatocellular carcinoma (66% and 97%, respectively) were higher
than those of dynamic MR imaging (63% and 96%, respectively) (Fig.
2A,2B,2C,2D),
although there was no significant difference in sensitivity between dynamic MR
imaging and dynamic helical CT for the individual observers (p >
0.05) (Fig.
2A,2B,2C,2D).
|
|
|
|
|
|
|
|
Hepatocellular carcinomas that were missed on dynamic helical CT scans were detected on dynamic MR images by observer 1 (six tumors in five patients), observer 2 (two tumors in two patients), and observer 3 (10 tumors in eight patients). Hepatocellular carcinomas missed on the dynamic MR images were detected on dynamic helical CT scans by observer 1 (12 tumors in 10 patients), observer 2 (seven tumors in seven patients), and observer 3 (seven tumors in six patients).
For hepatocellular carcinomas that were detected on both dynamic MR imaging and dynamic helical CT, the confidence level of detection was higher for dynamic MR imaging for eight tumors in seven patients for observer 1, 10 tumors in 10 patients for observer 2, and 11 tumors in nine patients for observer 3, whereas the reverse was true for seven tumors in six patients for observer 1, six tumors in six patients for observer 2, and seven tumors in seven patients for observer 3.
In considering the patients with hepatocellular carcinoma rather than their individual tumor lesions, observers 1 and 3 each detected tumors only on the dynamic MR images in three patients, and observer 2, in one patient. Observer 1 detected tumors on the dynamic helical CT images alone in four patients, observer 2 in two patients, and observer 3 in one patient.
The kappa values for the three observers, calculated on the basis of each observer's confidence level for the alternative free-response ROC analysis, were 0.77 for MR imaging and 0.79 for CT and showed substantial agreement with regard to the presence of lesions.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In this study, we also used optimal timing scans for both CT and MR imaging, using the test injection or automatic bolus-tracking technique as the time interval between the initiation of contrast material. The start of the arterial phase of hepatic enhancement depends on various factors such as patient weight and cardiac output. For the detection of hepatocellular carcinoma, optimal timing of arterial phase scanning is important [1,2,3,4,5,6,7,8,9]. Yamashita et al. [1] used single arterial phase imaging for both imaging techniques and did not use optimal timing scans for either.
Although there was no significant difference in detection of sensitivity and diagnostic accuracy (Az values) of hepatocellular carcinoma between dynamic MR imaging and dynamic helical CT in this study when these advanced techniques were used, some tumors were detected with only one of these procedures. On the basis of our results, we believe that both dynamic MR imaging and dynamic helical CT should be used for the initial staging examination of hepatocellular carcinoma, whereas either of them may be used for follow-up after treatment.
Our multiple independent observer method and alternative free-response ROC measurements allowed us to measure and control human observer performance and variability. Good agreement was shown among the three observers, as well as an absence of any significant differences between dynamic MR imaging and dynamic helical CT in the detection of hepatocellular carcinoma.
We used a single-detector helical CT scanner to perform the double arterial phase imaging. If we had been able to use multidetector CT, we could have obtained thinner slice images. However, we have data that suggest there is no significant difference in sensitivity between a slice thickness of 2.5 mm and one of 5 mm for the detection of hypervascular hepatocellular carcinoma [19].
To have a good standard of tumor burden, we included only patients in our series who also had undergone a confirmatory invasive imaging study in the form of CT hepatic arteriography and CT during arterial portography at the time of angiography and within 1 month after CT and MR imaging. In addition, these patients underwent follow-up CT examinations more than 6 months later. Thus, all patients were highly suspected of having hepatocellular carcinoma, and as a result, in the patients included in this study, there were none without hepatocellular carcinoma. However, only the study designer and the coordinator knew how the patients were selected, and only the three observers were aware that the patients were at risk for hepatocellular carcinoma, but the observers were unaware of the presence or location of liver lesions. We therefore believe there was no bias at the time of the blinded interpretation.
Some lesions in our study that we believed to represent hepatocellular carcinoma lacked histologic proof. However, all lesions were subjected to several confirmatory studies such as CT hepatic arteriography, CT during arterial portography [17], CT after arterial infusion of iodized oil [20, 21], and CT follow-up. Each of these studies, especially in combination, has been found to detect hypervascular hepatocellular carcinoma with an accuracy approaching 100%. Moreover, we could follow the course of most lesions over time and in response to therapy, especially transcatheter-arterial chemoembolization.
Dynamic MR imaging with 3D Fourier transformation-enhanced fast gradient-echo sequences with a special spectral inversion recovery pulse showed no significant difference in detecting sensitivity and diagnostic accuracy (Az values) for hepatocellular carcinoma compared with dynamic helical CT using double arterial phase imaging. Even more recent advances in MR imaging and CT techniques—for example, the sensitivity encoding technique [22] and multidetector CT—may further improve diagnostic accuracy. Further studies are needed.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
1.5 cm) malignant hepatic neoplasms.
AJR
1995;164:879
-884
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. M. Willatt, H. K. Hussain, S. Adusumilli, and J. A. Marrero MR Imaging of Hepatocellular Carcinoma in the Cirrhotic Liver: Challenges and Controversies Radiology, May 1, 2008; 247(2): 311 - 330. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. T. Schindera, R. C. Nelson, S. Mukundan Jr, E. K. Paulson, T. A. Jaffe, C. M. Miller, D. M. DeLong, K. Kawaji, T. T. Yoshizumi, and E. Samei Hypervascular Liver Tumors: Low Tube Voltage, High Tube Current Multi Detector Row CT for Enhanced Detection Phantom Study Radiology, December 1, 2007; 246(1): 125 - 132. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Schnorr, S. Wagner, C. Abramjuk, R. Drees, T. Schink, E. A. Schellenberger, H. Pilgrimm, B. Hamm, and M. Taupitz Focal Liver Lesions: SPIO-, Gadolinium-, and Ferucarbotran-enhanced Dynamic T1-weighted and Delayed T2-weighted MR Imaging in Rabbits Radiology, July 1, 2006; 240(1): 90 - 100. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z.-H. Fan, M.-H. Chen, Y. Dai, Y.-B. Wang, K. Yan, W. Wu, W. Yang, and S.-S. Yin Evaluation of primary malignancies of the liver using contrast-enhanced sonography: correlation with pathology. Am. J. Roentgenol., June 1, 2006; 186(6): 1512 - 1519. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Giorgio, G. Ferraioli, L. Tarantino, G. de Stefano, V. Scala, F. Scarano, C. Coppola, and L. Del Viscovo Contrast-Enhanced Sonographic Appearance of Hepatocellular Carcinoma in Patients with Cirrhosis: Comparison with Contrast-Enhanced Helical CT Appearance Am. J. Roentgenol., November 1, 2004; 183(5): 1319 - 1326. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Valls, M. Cos, J. Figueras, E. Andia, E. Ramos, A. Sanchez, T. Serrano, and J. Torras Pretransplantation Diagnosis and Staging of Hepatocellular Carcinoma in Patients with Cirrhosis: Value of Dual-Phase Helical CT Am. J. Roentgenol., April 1, 2004; 182(4): 1011 - 1017. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
