| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Gastrointestinal Imaging |
1 From the Departments of Radiology (P.S., M.B., M.A., L.H., R.R.) and Surgery (M.P., Y.P., P.V.), Hôpital Lariboisière-AP-HP, 2 rue Ambroise Paré, 75475 Paris cedex 10, France. Received December 2, 2002; revision requested February 6, 2003; final revision received September 17; accepted October 21. Address correspondence to P.S. (e-mail: philippe.soyer@lrb.ap-hop-paris.fr).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: In this prospective study, 32 patients with 59 surgically and histopathologically proved hypovascular hepatic metastases underwent triple-phase helical CT of the liver, which included unenhanced, arterial-dominant, and portal-dominant phase scanning. Images from each phase were separately analyzed by three readers, and disagreements were resolved with consensus readings. The findings on CT images were compared with intraoperative US and histopathologic findings on a lesion-by-lesion basis to determine the sensitivity of each imaging phase. Statistical review of the lesion-by-lesion analysis was performed by using the Wilcoxon rank sum test.
RESULTS: Among 59 hepatic metastases, unenhanced, arterial-dominant, and portal-dominant phase helical CT imaging depicted 39 (66.1%; 95% CI: 53.3%, 76.8%), 44 (74.5%; 95% CI: 62.2%, 83.9%), and 54 (91.5%; 95% CI: 81.6%, 96.3%) metastases, respectively. Portal-dominant phase imaging depicted significantly more hypovascular hepatic metastases than did unenhanced (P < .001) or arterial-dominant (P < .01) phase imaging (Wilcoxon test).
CONCLUSION: Preoperative use of triple-phase helical CT in patients with hypovascular hepatic metastases may not be warranted. Portal-dominant phase helical CT imaging allows depiction of significantly more hypovascular hepatic metastases than does imaging during any of the other phases.
© RSNA, 2004
Index terms: Computed tomography (CT), phase imaging • Liver neoplasms, CT, 761.1211, 761.12113 • Liver neoplasms, diagnosis, 761.33 • Liver neoplasms, metastases, 761.33
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Double- or triple-phase helical computed tomography (CT) of the liver has been shown to be useful in the detection of hypervascular hepatic tumors, and this has been clearly demonstrated in the detection of hepatocellular carcinoma (3); however, the optimal helical CT technique for the evaluation of hepatic metastases remains a matter of debate. One rule is to tailor the number of acquisitions and the imaging phase to the degree of vascularization in the suspected primary tumor. In patients suspected of having hypervascular hepatic metastases (ie, hepatic metastases that appear more vascular and show greater contrast enhancement than does the adjacent hepatic parenchyma at arterial-dominant phase CT), such as metastases from neuroendocrine tumors, arterial-dominant phase helical CT has proved useful for accurate evaluation of the extent of disease (4).
To our knowledge, however, no definite consensus has been clearly established yet for other types of hepatic metastases with variable degrees of vascularization compared with the adjacent hepatic parenchyma. One possible reason may be that, in some studies, attempts to determine the respective sensitivities of the phases did not include a careful surgical and pathologic correlation to determine the number of metastases present (5,6). Another reason may be that, in studies with a valid standard of reference, only portal-dominant phase helical CT images were evaluated (7–9). Thus, the respective values of the unenhanced, arterial-dominant, and portal-dominant imaging phases in triple-phase helical CT, as well as the overall sensitivity of helical CT, for the detection of hypovascular hepatic metastases remain somewhat unclear.
We conducted this prospective study to compare the sensitivities of unenhanced, arterial-dominant, and portal-dominant phase helical CT in the preoperative depiction of hypovascular hepatic metastases by using intraoperative ultrasonographic (US) and histopathologic findings as the standard of reference.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Thirty-two patients, who were considered suitable candidates for hepatic resection on the basis of preoperative imaging results, underwent partial hepatic resection (including inspection, manual palpation, and intraoperative US exploration) and were included in our study. There were 16 men (age range, 35–76 years; mean age, 60 years) and 16 women (age range, 29–77 years; mean age, 58 years). The remaining 28 patients were excluded from the study because triple-phase helical CT images showed too many lesions, which precluded surgery, or because the patients were found to have benign hepatic lesions at preoperative imaging examinations. For all patients included, triple-phase helical CT and surgery were performed within 3–17 days of each other (mean, 6 days).
Triple-Phase Helical CT
Triple-phase helical CT images of the liver were obtained with commercially available CT scanners. In 13 patients, images were obtained with a Somatom Plus 4 helical scanner (Siemens Medical Systems, Iselin, NJ) with one row of detectors, 5-mm collimation, 5-mm reconstruction interval, gantry rotation speed of 0.75 second, table speed of 5 mm per gantry rotation (pitch of 1), 120 kV, and 200 mAs. In eight patients, images were obtained with a CT Twin RTS (Philips; Eindhoven, the Netherlands) with two rows of detectors, 2 x 5-mm collimation, 6-mm reconstruction interval, gantry rotation speed of 1.0 second, table speed of 10 mm per gantry rotation (pitch of 2), 120 kV, and 200 mAs. In 11 patients, a Somatom Plus 4 Volume Zoom scanner (Siemens Medical Systems) with four rows of detectors, 4 x 2.5-mm collimation, 5-mm reconstruction interval, gantry rotation speed of 0.5 second, table speed of 12.5 mm per gantry rotation (pitch of 5), 120 kVp, and 165 mAs was used.
Once unenhanced helical CT had been performed through the entire abdomen, 120 mL of iohexol (Omnipaque 300; Nycomed, New York, NY) was injected intravenously through a 20-gauge cannula at a rate of 3 mL/sec with an automated power injector (MedRad, Pittsburgh, Pa). For the arterial-dominant phase, the delay between the start of contrast material administration and helical scanning was 25 seconds. For the portal-dominant phase, the delay between the start of contrast material administration and helical scanning was 70 seconds. Helical CT images were originally obtained with the standard settings of window level and width used in our department for viewing liver parenchyma (window level, 50 HU; window width, 300 HU), but in some cases these values were adjusted to the specific case being studied to obtain an optimal window setting.
Image Analysis
Three observers (P.S., M.P., M.B.), who were unaware of patient identification, participated in six different reading sessions at 2-week intervals for a retrospective image analysis. The observers were aware of the overall goal of the study before the reading sessions and knew that the patients had been referred for preoperative assessment of hepatic metastases, but they did not have any other information regarding patient history. The helical CT images were first grouped into three separate image sets (unenhanced phase, arterial-dominant phase, and portal-dominant phase), which were interpreted by the three readers working in consensus and blinded to the results of other imaging techniques and to the results of the original interpretation. Each set, which included all of the cases, was interpreted in a random manner, and readers were blinded to the results of interpretation of the other sets of images by using three separate reading sessions, with an interval of 2 weeks between sessions. The observers were not blinded to phase type, however, because there were obvious differences in appearance between the sets of helical CT images. The number, size, and location of hepatic metastases visible on each set were noted for each patient. The segmental locations of hepatic metastases were categorized according to the Couinaud numbering system (10). If a metastasis crossed segmental boundaries, it was considered to involve more than one segment. Hepatic metastases on individual images were considered to exist if their presence was agreed on by the three observers.
In addition, the observers were asked to determine whether any biliary cysts or cavernous hemangiomas were present on the portal-dominant phase images. Cysts were defined as hypoattenuating (attenuation similar to that of water) round lesions with no visible wall and no enhancement. Cavernous hemangiomas were defined as hyperattenuating round lesions with discontinuous peripheral globular enhancement. The numbers and sizes of biliary cysts and cavernous hemangiomas were noted when present. In cases of disagreement between the three observers, a consensus reading was performed.
After the first three reading sessions were completed, three additional reading sessions were performed to assess the sensitivity achieved by the various combinations of the three phases; these sessions were performed by the same three observers using the same criteria as in the first sessions. The unenhanced and arterial-dominant phase image sets were combined and interpreted to determine the sensitivity of the combination of these two phases. Then the unenhanced and portal-dominant phase sets were combined and interpreted during another reading session. Finally, the unenhanced, arterial-dominant, and portal-dominant phase sets were combined and interpreted jointly in another session to determine the sensitivity of a combination of all three phases. To limit learning bias, a balanced permutation of the order of cases was introduced in each reading session (11).
Standard of Reference
The combination of results from intraoperative US and careful surgical inspection and palpation of the liver, together with histopathologic findings, constituted the standard of reference in our study. In all patients, surgery was performed by one of the surgeons at our institution, who had more than 8 years of experience performing liver surgery and intraoperative US and was aware of the results of all available preoperative imaging examinations. Before surgery, helical CT findings were prospectively analyzed by a staff radiologist who did not participate in the retrospective image analysis and by the surgeon who performed the surgery in each patient, so that a one-to-one correlation could be obtained between helical CT and surgical findings. Intraoperative US (Eccocee; Toshiba, Tokyo, Japan) was performed with a high-resolution 7.5-MHz intraoperative US probe. The transducer, shaped like the letter "T," was sterilized and placed directly on the surface of the liver as previously described (12). When unsuspected nonpalpable deep hepatic lesions were detected at intraoperative US, a biopsy was performed with US guidance and the location and size of the lesion were noted on the surgical report. Also, when a biliary cyst or a cavernous hemangioma was observed at intraoperative US, it was noted on the surgical report.
After surgery, the segmentectomy specimens were sliced at 5-mm intervals by a staff pathologist, who carefully searched for small hepatic tumors and made direct comparisons with the imaging findings. The sizes of the hepatic metastases were determined to the nearest millimeter from measurements made by the pathologist after surgery. The standard of reference was established by the surgeon and the pathologist, who agreed on a single final interpretation of the total number of metastases and benign lesions detected in the resected specimens. At the final lesion-by-lesion analysis, a tumor visible on the helical CT images was scored as a true-positive lesion only if it corresponded in location with a metastasis in the resected specimens.
Statistical Analysis
Sex-related difference in the distribution of metastases in the patient population was evaluated by using the 
2 test. Sensitivity was defined for each phase as the number of hepatic metastases correctly depicted at that phase, divided by the number of hepatic metastases identified at intraoperative US and histopathologic examinations. Sensitivity was defined as the true-positive rate (ie, the number of hepatic metastases correctly depicted at CT divided by the number of metastases present as defined with the standard of reference). Sensitivity and corresponding 95% CIs were thus calculated for each imaging phase. The difference in sensitivity between two given phases was the sensitivity of one phase minus the sensitivity of the other phase, and 95% CIs were calculated for the differences in sensitivity between unenhanced, arterial-dominant, and portal-dominant phase imaging. The calculation of the 95% CI provided a range of plausible differences in sensitivity (13).
Differences in sensitivity between the different phases and the combinations of phases were analyzed by using the Wilcoxon rank sum test. This test takes into account the clustering effect caused by individual patients who may have multiple hypovascular hepatic metastases. A conservative approach was thus used to test the hypothesis that the numbers of metastases detected at each phase and the different combinations of phases were different. First, the hypothesis that the three different phases and combinations of phases were different was tested by means of the Friedman two-way analysis of variance with ranks. When the result of this global test was significant, comparisons between the different phases were made by using the Wilcoxon rank sum test. Significance level for two-by-two comparisons was adjusted for multiplicity by using the Bonferroni correction. A P value of less than .05 was considered to indicate a statistically significant difference between the helical CT phases.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
2 test; 2 = 2.381, P = .7943). The various partial hepatic resections included the following: resection of segments II and III (n = 4); segments V, VI, VII, and VIII (n = 10); segments II, III, and IV (n = 6); segments IV, V, VI, VII, and VIII (n = 6); segments I, V, VI, VII, and VIII (n = 2); segments IV, V, and VI (n = 1); and resection of a single segment (n = 3). In addition, single or multiple wedge resections were performed in five patients in association with one of these resections.
The mean number of hepatic metastases per patient was 1.8 (range, one to seven metastases). Twenty-one patients had one hepatic metastasis each, three patients had two hepatic metastases, four patients had three hepatic metastases, two patients had four hepatic metastases, one patient had five hepatic metastases, and one patient had seven hepatic metastases. The maximum transverse diameter of hepatic metastases ranged from 4 to 105 mm (mean, 22 mm). The metastases were located in segment I (n = 2), segment II (n = 5), segment III (n = 6), segment IV (n = 15), segment V (n = 8), segment VI (n = 11), segment VII (n = 12), and segment VIII (n = 10). A total of 69 segments were involved because seven hepatic metastases involved more than one segment (one metastasis involved four segments, one metastasis involved three segments, and five metastases involved two segments). The metastases originated from colorectal adenocarcinoma in 27 patients, stromal tumor in three patients (originating from the duodenum in two patients and the rectum in one patient), and malignant colonic schwannoma in one patient. Metastases from stromal tumor were considered hypovascular on the basis of the results of color and power Doppler US examinations that showed hypovascularity in all cases. One patient had hypovascular hepatic metastases from adenocarcinoma of unknown origin. Among these 32 patients, two had been referred for recurrence and had previously undergone partial hepatic resection (resection of segments V, VI, VII, and VIII in one patient and resection of segment IV in the other patient).
Unenhanced phase imaging allowed depiction of 39 of 59 hepatic metastases (sensitivity, 66.1%; 95% CI: 53.3%, 76.8%) (Fig 1). Arterial-dominant phase imaging allowed depiction of 44 of 59 hepatic metastases (sensitivity, 74.5%; 95% CI: 62.2%, 83.9%). By using the combination of unenhanced and arterial-dominant phase imaging, a sensitivity of 74.5% (44 of 59 patients; 95% CI: 62.2%, 83.9%) was obtained in the depiction of hepatic metastases, because any metastases that were not depicted on arterial-dominant phase images were also not seen on the unenhanced images (Fig 2). Portal-dominant phase imaging allowed depiction of 54 of 59 hepatic metastases (sensitivity, 91.5%; 95% CI: 81.6%, 96.3%) (Fig 3). By using the combination of unenhanced, arterial-dominant, and portal-dominant phase imaging, a sensitivity of 91.5% (54 of 59; 95% CI: 81.6%, 96.3%) was obtained in the depiction of hepatic metastases, because metastases not depicted on portal-dominant phase images were also not seen with the combination of unenhanced and arterial-dominant phase images. Four hepatic cysts and three cavernous hemangiomas were depicted on portal-dominant phase images and were correctly characterized; this was confirmed histopathologically in three cases or intraoperatively for benign lesions that were not removed in four cases. No false-positive findings of hepatic metastases were made on portal-dominant phase images (seven of seven; specificity, 100%; 95% CI: 64.5%, 100%).
|
|
|
|
|
|
|
|
|
|
2 = 27.333, P < .0001). Portal-dominant phase imaging alone allowed depiction of significantly more hepatic metastases than did unenhanced phase imaging (difference in sensitivities, 25.4%; 95% CI: 12.59%, 38.20%; P < .001 with Wilcoxon test), or arterial-dominant phase imaging (difference in sensitivities, 16.9%; 95% CI: 5.70%, 28.20%; P < .01 with Wilcoxon test), and the combination of unenhanced and arterial-dominant phase imaging (difference in sensitivities, 16.9%; 95% CI: 5.70%, 28.20%; P < .01 with Wilcoxon test) (Fig 3). None of the five hepatic metastases missed at portal-dominant phase imaging was seen with the other imaging phases. No significant difference in sensitivities was found between unenhanced and arterial-dominant phase imaging. The 20 hepatic metastases not seen at unenhanced phase imaging had a mean maximum transverse diameter of 11 mm (range, 4–25 mm). The 15 metastases missed at arterial-dominant phase imaging had a mean maximum transverse diameter of 9 mm (range, 4–15 mm). All five hepatic metastases missed at portal-dominant phase imaging had a maximum diameter of less than 10 mm, with a mean diameter of 5 mm (range, 4–6 mm); they were located in segment VIII at the hepatic dome (n = 2), in the central part of segment V (n = 1), in the subcapsular part of segment IV (n = 1), or in the central part of segment III (n = 1). In no patient was the liver found to be unresectable after the intraoperative detection of additional metastases, but in three patients, the type of surgical resection that had been preoperatively decided was modified. The correlation between the diameter of the hepatic metastasis and the detection at each imaging phase is reported in the Table.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The sensitivity of portal-dominant phase helical CT in our study (91.5%; 95% CI: 81.6%, 96.3%) was within the range of sensitivities reported by authors who investigated portal-dominant phase imaging alone (7–9). In one study, Kuszyk et al (7) found a sensitivity of 88% (36 of 41 metastases) for the detection of hepatic metastases from colorectal cancer by using portal-dominant phase helical CT alone. In a similar study, Valls et al (9) obtained a sensitivity of 85% (247 of 290 metastases from colorectal cancer) by using portal-dominant phase imaging alone.
In recent years, several studies have been performed to evaluate each imaging phase at triple-phase helical CT of the liver (4,6). Some studies have been focused specifically on hypervascular hepatic tumors, and authors of these studies have found that arterial-dominant phase imaging is useful in a subset of hypervascular tumors, such as hepatocellular carcinoma, and hypervascular metastases, such as metastases from islet cell tumors (3,4). For a small subset of lesions, metastases that are typically hypervascular are best detected with the combination of unenhanced and portal-dominant venous phase imaging (14).
In specific cases, such as hepatic metastases from breast carcinoma, a subset of which may be hypervascular compared with the adjacent hepatic parenchyma on arterial-dominant phase images, arterial-dominant phase imaging has been found to be of little value in lesion detection (15,16). Frederick et al (15) compared unenhanced, arterial-dominant, and portal-dominant phase helical CT for the detection of metastases from breast carcinoma and found that portal-dominant phase imaging, with a sensitivity of 85%, detected significantly more hepatic metastases than did arterial-dominant (sensitivity, 59%) or unenhanced (sensitivity, 61%) phase imaging. In addition, Frederick et al did not find any case with a negative result at initial portal-dominaint phase imaging in which the results became positive for metastasis after the addition of unenhanced or arterial-dominant phase imaging (15). This conclusion was further supported by the results of a larger study from the same group, in which the addition of unenhanced or arterial-dominant phase imaging did not reveal substantially more metastases compared with portal-dominant phase imaging alone (17). In contrast, for the detection of hepatic metastases from renal cell carcinoma, Raptopoulos et al (18) found that the highest degree of sensitivity was achieved with the combination of unenhanced, arterial-dominant, and portal-dominant phase imaging.
All hepatic metastases in our study were considered hypovascular either because they originated from primary neoplasms typically considered hypovascular (such as colonic adenocarcinoma) or because color and power Doppler US did not show arterial hypervascularity in those that originated from primary neoplasms not considered typically hypovascular (such as stromal tumors). In hypovascular hepatic tumors, the adjunct use of unenhanced and arterial-dominant phase imaging with portal-dominant phase imaging for lesion detection remains debated. Ch’en et al (5) found that the addition of arterial-dominant phase imaging did not help to detect any additional metastases compared with imaging in the portal-dominant phase alone. In a similar study, Miller et al (6) found that, in patients with hypovascular tumors, more lesions were detected on portal-dominant phase images; they concluded in their study that unenhanced CT is not routinely necessary for the detection of hypovascular metastases. On the other hand, others have found that some hypovascular hepatic metastases missed on portal-dominant phase images were depicted on arterial-dominant phase images, so that a statistically significant improvement in the detection of hypovascular hepatic metastasis was achieved by using the combination of arterial-dominant and portal-dominant phase imaging compared with portal-dominant phase imaging alone (19). Our results, which suggest that images obtained at the portal-dominant phase alone allow detection of significantly more hypovascular hepatic metastases than do images obtained at the two other phases, are consistent with those obtained by Ch’en et al (5) and Miller et al (6).
We found that the arterial-dominant phase did not help to depict additional hepatic metastases not visible on portal-dominant phase images; however, we did not assess the value of the arterial-dominant phase in terms of anatomic information that can be given to the surgeon with the use of three-dimensional reconstructions. Helical CT angiographic images of the hepatic arterial vasculature may be useful to the surgeon before partial hepatic resection, because they may show unexpected variations of the celiac axis or superior mesenteric artery that may make surgery more difficult. However, the effect of three-dimensional arterial images on the surgical decision-making process is not definitely established yet.
In our study, we found that unenhanced phase imaging is significantly less sensitive than imaging during the other two phases for the detection of hypovascular hepatic metastases at helical CT. Our results suggest that unenhanced phase imaging is not routinely necessary for the preoperative detection of hypovascular hepatic metastases. To our knowledge, few studies have investigated the intrinsic or added value of unenhanced imaging for the depiction of hypovascular hepatic metastases at helical CT. In their study, Miller et al (6) found that unenhanced phase imaging was less sensitive than portal-dominant phase imaging, but their results were not based on histopathologic proof, which makes it difficult to draw conclusions.
Several criticisms may be raised with respect to our study. The first criticism relates to the fact that we evaluated triple-phase helical CT by using a reference standard of results at surgery, yet the results of triple-phase helical CT determined whether patients were suitable for surgery. It might have been possible that, in patients with many metastases, some metastases might have been missed at triple-phase helical CT because of their small size, thus potentially resulting in an overestimation of the actual sensitivity of the different phases. For ethical reasons, it was not possible to perform surgery in patients with unresectable disease solely to check for the number of metastases present.
The second criticism is the small number of patients included in our study. Our results need to be verified with further studies involving more patients.
The third criticism is that, in all patients, it was not possible to perform pathologic examination of the part of the liver that was left in place. Although we agree that the more accurate standard of reference would be results of pathologic examination of the entire liver, this would be feasible only in a group of patients undergoing liver transplantation. In portions of the liver that were not resected, the reference standard we used for determining the number of hepatic metastases present was the combination of surgical examination and intraoperative US; this combination has previously been found to have a high sensitivity (96%) for hepatic metastasis detection (12).
The fourth potential criticism is due to the section thickness used for examining the resected specimens. It is reasonable to believe that some hepatic metastases with a diameter smaller than the section thickness may have been missed, potentially causing us to overestimate the true sensitivity of the different phases in the detection of hepatic metastases.
The fifth possible criticism may be raised with respect to the amount of contrast material we used when performing triple-phase helical CT examinations. In several studies, 150–170 mL of iodinated contrast material has been used; we administered only 120 mL to our patients (6,9). It is possible that a higher dose of contrast material, which results in higher degrees of hepatic parenchymal enhancement, might have caused CT to be more sensitive in the depiction of the small hepatic metastases that were missed in our study.
The sixth criticism applies to the fact that three of our patients had hepatic metastases that originated from gastrointestinal stromal tumors. These tumors, as well as their metastases, are generally considered hypervascular. In our study, the three patients with this type of metastasis were included because color Doppler US did not show hypervascularity. It may be argued, however, that our results apply to a subset of patients with hypovascular hepatic metastases from gastrointestinal tumor and that these results may not be valid in a more general population of patients with hepatic metastases from gastrointestinal tumors.
The final potential criticism is that we used a consensus reading and did not compare the readings of each observer independently by using receiver operating characteristic analysis.
In conclusion, we found that the portal-dominant phase is the most sensitive of the three helical CT imaging phases in the preoperative evaluation of patients with hypovascular hepatic metastases because it allows depiction of significantly more hepatic metastases than do any of the other phases. Thus, the preoperative use of triple-phase helical CT in patients with hypovascular hepatic metastases may not be warranted. Our results encourage the use of a single-phase helical CT technique for the evaluation of patients with hypovascular hepatic metastases. This technique, which substantially reduces patient irradiation and examination time compared with a double- or triple-phase technique, can be performed first in patients who are candidates for curative partial hepatic resection and then in follow-up of the subgroup of patients who undergo partial hepatic resection.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. S. Lee, S. H. Park, E. K. Choi, S. Y. Kim, M.-J. Kim, K. H. Lee, and Y. H. Kim Colorectal Polyps on Portal Phase Contrast-Enhanced CT Colonography: Lesion Attenuation and Distinction from Tagged Feces Am. J. Roentgenol., July 1, 2007; 189(1): 35 - 40. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Monzawa, T. Ichikawa, H. Nakajima, Y. Kitanaka, K. Omata, and T. Araki Dynamic CT for Detecting Small Hepatocellular Carcinoma: Usefulness of Delayed Phase Imaging Am. J. Roentgenol., January 1, 2007; 188(1): 147 - 153. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-H. Kim, J. Czernin, M. S. Allen-Auerbach, B. S. Halpern, B. J. Fueger, J. R. Hecht, O. Ratib, M. E. Phelps, and W. A. Weber Comparison Between 18F-FDG PET, In-Line PET/CT, and Software Fusion for Restaging of Recurrent Colorectal Cancer J. Nucl. Med., April 1, 2005; 46(4): 587 - 595. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
