| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Diagnostic Radiology, Aichi Cancer Center, 1-1 Kanokoden
Chikusa-ku, Nagoya 464-8681, Japan.
2
Department of Radiology, Gifu University School of Medicine, 40 Tsukasamachi,
Gifu 500-8705, Japan.
3
Department of Radiology, Kyoto First Red Cross Hospital, Kyoto 605-0981,
Japan.
4
Department of Gastroenterological Surgery, Aichi Cancer Center, Nagoya
464-8681, Japan.
5
Department of Radiology, Institute of Clinical Medicine, University of
Tsukuba, Tsukuba 305-8575, Japan.
Received July 7, 1999;
accepted after revision September 14, 1999.
Address correspondence to Y. Inaba.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Fifty-four patients with hepatic metastases from colorectal cancer preoperatively underwent combined CT during arterial portography and CT hepatic arteriography using the unified CT and angiography system. Three radiologists independently and retrospectively reviewed the images of CT during arterial portography alone, CT hepatic arteriography alone, and combined CT during arterial portography and CT hepatic arteriography. Image review was conducted on a segment-by-segment basis; a total of 432 hepatic segments with (n = 103) 118 metastatic tumors ranging in size from 2 to 160 mm (mean, 25.8 mm) and without (n = 329) tumor were reviewed.
RESULTS. Relative sensitivity of combined CT during arterial portography and CT hepatic arteriography (87%) was higher than that of CT during arterial portography alone (80%, p < 0.0005) and CT hepatic arteriography alone (83%, p < 0.005). Relative specificity of CT hepatic arteriography alone (95%, p < 0.0005) and combined CT during arterial portography and CT hepatic arteriography (96%, p < 0.0001) was higher than that of CT during arterial portography alone (91%). Diagnostic accuracy, determined by a receiver operating characteristic curve analysis, was greater with combined CT during arterial portography and CT hepatic arteriography than with CT during arterial portography alone (p < 0.05) or CT hepatic arteriography alone (p < 0.01).
CONCLUSION. Using a unified CT and angiography system, we found that combined CT during arterial portography and CT hepatic arteriography significantly raised the detectability of hepatic metastases from colorectal cancer.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
With the advent in the early 1990s of the helical CT technique, which enables whole-liver scanning during a single breath-hold, we have been performing angiographically assisted CT for preoperative workup in patients with malignant hepatic tumors. Since we developed and started to clinically use a unified CT and angiography system in 1992 [10] (Fig. 1A,1B), we have performed angiographically assisted helical CT with this device in more than 400 patients with suspected hepatic tumors. This device enables us to obtain helical CT hepatic arteriograms from all hepatic segments by placing an angiographic catheter in all the hepatic arteries supplying the liver under fluoroscopic guidance with the angiographic imager. In contrast, previous studies, limited by CT hepatic arteriography, used a single hepatic artery for CT hepatic arteriography [6, 11,12,13].
|
|
Lobectomy, segmentectomy, or subsegmentectomy has been performed to improve the survival rates of surgical candidates at our institution who are selected after workup that includes angiographically assisted CT [14, 15]. We therefore performed a retrospective comparison of CT during arterial portography alone, CT hepatic arteriography alone, and the combination of both, by means of receiver operating characteristic (ROC) analysis on a hepatic segment-by-segment basis to determine whether combined CT during arterial portography and CT hepatic arteriography is more accurate than CT during arterial portography alone or CT hepatic arteriography alone in the preoperative detection of hepatic metastases from colorectal cancer.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
These patients underwent follow-up imaging by sonography and IV contrast-enhanced helical CT 3-6 months after hepatic surgery. Eleven patients developed recurrent hepatic metastases 6-12 months after surgery. After image review for this study, we retrospectively reviewed angiographically assisted CT images in the 11 patients with recurrence and confirmed the lack of abnormal imaging findings in the sites corresponding to those on the follow-up CT or sonographic images. We did not include those recurrent tumors as standard lesions in the present study because it was impossible to determine whether tiny metastatic deposits were present at the time of the angiographically assisted CT study and no definite metastatic lesions could be detected in the residual liver by intraoperative sonography.
CT Techniques
A 5-French angiographic catheter for CT during arterial portography was
placed in the superior mesenteric artery in all patients using the Seldinger
technique through the femoral artery (Figs.
2A,
3A,
4A). In nine patients with a
replaced right hepatic artery arising from the superior mesenteric artery, the
superior mesenteric artery catheter was inserted well beyond the hepatic
artery origin so that the contrast material did not overflow into the replaced
right hepatic artery. After performing CT during arterial portography, digital
subtraction celiac or hepatic arteriography was performed
(Fig. 4B). Then the tip of the
angiographic catheter was placed in every hepatic artery supplying the liver,
such as the proper or common hepatic artery or replaced right or left hepatic
artery, under fluoroscopic control. As many sessions of CT hepatic
arteriography as the number of hepatic arteries supplying the liver were
performed to obtain the whole-liver CT hepatic arteriography images (Figs.
2B,
3B,
4C, and
4D). Of the 54 patients, 42
had a single hepatic artery, eight had two hepatic arteries, and four had
three hepatic arteries.
|
|
|
|
|
|
|
|
Angiographically assisted CT was performed using a unified CT and angiography system (Interventional-CT system; Toshiba Medical Systems, Tokyo, Japan) under development since 1992. This system consists of a helical CT unit (X-force [before 1995] or X-vigor [1995 and after]) and digital subtraction angiography unit (model DFP-60A; Toshiba Medical Systems) also equipped with screen-film angiographic capability (model KXO-80C; Toshiba Medical Systems). Both components of the system were arranged in a linear configuration to form a 2.95-m-long patient cradle that facilitated safe and rapid patient transfer from one unit to the other with minimal danger of catheter dislodgement [10]. By sliding the single-patient cradle between the CT gantry and the angiographic imager, fluoroscopic monitoring, digital subtraction angiography, or helical CT imaging was available at the operator's discretion.
The helical CT images were obtained in a craniocaudal direction with 7- to 10-mm collimation, 7- to 10-mm/sec table speed, 130 kVp, and 150 mAs during a single breath-hold helical acquisition for 20-32 sec, depending on the liver size. We routinely supplied 100% oxygen at 21/min to the patients through a nasal cannula to assure breath-holding. Breath-holding was successful in all patients. On CT during arterial portography, data acquisition was started 25-30 sec after the initiation of a transcatheter superior mesenteric arterial injection of 50-70 ml of nonionic contrast material, iopamidol (Iopamiron 150; Schering, Berlin, Germany) that contained 150 mg I/ml at a rate of 2 ml/sec, using an automated power injector (Model Mark V Plus; Medrad, Pittsburgh, PA). We did not perform transcatheter administration of vasodilator before CT during arterial portography. On CT hepatic arteriography, data acquisition was started 5-10 sec after the initiation of a transcatheter hepatic arterial injection of 20-30 ml of the nonionic contrast material at a rate of 1 ml/sec.
No technical failure to obtain CT during arterial portography and whole-liver CT hepatic arteriography images, such as catheter placement failure or catheter dislodgement during sliding the patient cradle, was experienced in the series. The total examination time for the angiographically assisted CT study ranged from 30 min to 2 hr (mean, 1 hr) and was well tolerated by all patients. The total dose of iodine after the digital subtraction angiography and angiographically assisted CT study ranged from 22,300 to 52,100 mg (mean, 29,800 mg). Contiguous axial images of 7- to 10-mm thickness with a 5- to 10-mm step were reconstructed from the volumetric data set using a 180° linear interpolation algorithm.
Image Analysis
Three gastrointestinal radiologists independently reviewed the CT images.
They knew that the patients were referred for assessment of suspected liver
metastases but were not provided with any other patient information. They
reviewed images of CT during arterial portography alone, CT hepatic
arteriography alone, and combined CT during arterial portography and CT
hepatic arteriography in 54 patients.
The image review was conducted on a segment-by-segment basis because one of the chief determinants of hepatic resectability is the accurate definition of the number of segments to be resected and because our objective was to compare the ability of the radiologists to detect liver metastases on images obtained with each imaging technique and not to localize lesions. To prevent errors in identification of lesion location by the radiologists, the hepatic segment numbering system of Couinaud [16] was drawn on the images by the study coordinator. The image review was performed in three separate sessions. Images were reviewed in alphabetic order according to the patient's name, but the order in which the images from the three imaging techniques were reviewed was randomized. In other words, images from all patients were reviewed at a single session, but only the images obtained with one of the three imaging techniques in a given patient were reviewed at that session. The images obtained with the other techniques were reviewed at the subsequent two sessions. To minimize learning bias, the name, age, identification number, and imaging parameters for each patient were masked, and the three reviewing sessions were performed at 2-week intervals.
For each imaging technique, the radiologists recorded the size and site (Couinaud segment [16]) of visible abnormalities and indicated, for each segment, whether metastatic deposits could be detected. The radiologists assigned one of five confidence levels (1 = definitely absent, 2 = probably absent, 3 = equivocal, 4 = probably present, 5 = definitely present). When a lesion was located in two or more segments, the radiologist was asked to consider only the segment that was mainly involved and to assess the probability of another metastatic deposit in the other segment. The radiologists were instructed to indicate a score of 1 when no focal attenuation change was seen; a score of 3 when the attenuation change was subtle, ill-defined, and not circular or oval in shape; and a score of 5 when the attenuation change was discrete, well-circumscribed, and circular or oval in shape. Scores of 2 and 4 were assigned on the basis of each radiologist's subjective judgment. A total of 432 hepatic segments, including 103 segments harboring 118 metastatic tumors, were reviewed.
Statistical Analysis
The relative sensitivity of each imaging sequence for hepatic metastasis
was determined by using the number of segments assigned a score of 3 or
greater (equivocal to definitely present) of the 103 segments with metastatic
deposits. The relative specificity of each imaging sequence was determined by
using the number of segments assigned a score of 1 or 2 (definitely absent or
probably absent) of the total 329 segments without metastasis. The relative
sensitivity and specificity were compared using the McNemar test. The relative
accuracy was compared with a chi-square test.
For each imaging technique, a binomial receiver operating characteristic (ROC) curve was fitted to the confidence rating of each radiologist by using a maximum-likelihood estimation [17]. The diagnostic accuracy of each imaging technique for each radiologist was estimated by calculating the area under the ROC curve [18]. Differences between the ROC curves of individual radiologists were tested by using an area test with a univariate z-score test of the difference between the areas under the two ROC curves [19].
To assess interobserver variability in interpreting images, the kappa statistic for multiple observers was used to measure the degree of agreement among the three observers. We used the nonweighted kappa statistic with binary data defined in terms of the presence (definitely present, probably present, equivocal) or absence (probably absent, definitely absent) of metastatic deposits in a hepatic segment. The degree of disagreement was not factored into the calculation. A kappa value less than or equal to 0.40 indicated positive but poor agreement, a value of 0.41-0.75 indicated good agreement, and a value greater than 0.75 indicated excellent agreement.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
The index values, for each radiologist, of the areas under the ROC curves of the three imaging techniques for detection of hepatic metastases are shown in Table 2. Diagnostic accuracy with combined CT during arterial portography and CT hepatic arteriography was significantly higher than that with CT during arterial portography alone (p < 0.05) for two of the three radiologists and was significantly higher than that with CT hepatic arteriography alone (p < 0.01) for one of the three radiologists.
|
The kappa values for CT during arterial portography alone, CT hepatic arteriography alone, and combined CT during arterial portography and CT hepatic arteriography were 0.81, 0.87, and 0.87, respectively. Excellent agreement was obtained among the radiologists with regard to the presence or absence of metastatic deposits in a given segment.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our results show that detectability of hepatic metastases from colorectal cancer with combined CT during arterial portography and CT hepatic arteriography was greater than that with CT during arterial portography alone and CT hepatic arteriography alone. This observation may indicate that CT during arterial portography alone or CT hepatic arteriography alone should not be recommended as a preoperative assessment in patients with suspected hepatic metastases. CT during arterial portography has been considered a very sensitive imaging method, despite its low specificity, especially for the noncirrhotic liver, which does not commonly have the severe liver deformity, multiple nodular changes in the liver parenchyma, and perfusion abnormalities caused by arterioportal shunting found in the cirrhotic liver. Other researchers [6, 11,12,13] have reported that the addition of CT hepatic arteriography to CT during arterial portography increased the accuracy of diagnosis of hepatic tumors; however, the additional value of CT hepatic arteriography combined with CT during arterial portography may be limited when just one of the arteries supplying the liver is used to opacify the liver as described in previous reports [6, 11,12,13]. The entire hepatic artery supply is revealed in only half of the CT hepatic arteriography patients after contrast injection into the common hepatic artery because of the high frequency of anatomic variations in the arterial blood supply to the liver [20]; thus, the whole liver cannot be imaged in nearly half the patients who undergo conventional CT hepatic arteriography. The unified CT and angiography system we used enabled us to obtain whole-liver CT hepatic arteriography images by selecting every hepatic artery under fluoroscopic guidance. We believe that this advantage may further increase the value of combined CT during arterial portography and CT hepatic arteriography.
The usefulness of combined CT during arterial portography and CT hepatic arteriography has been assessed by some previous researchers. Researchers who compared CT during arterial portography and CT during infusion hepatic arteriography with simultaneous use of both techniques concluded that CT during arterial portography alone should be recommended for the detection of malignant hepatic tumors and that CT during infusion hepatic arteriography is of value in differentiating malignant from benign small nodules [11]. Other researchers who evaluated whether the addition of CT hepatic arteriography to CT during arterial portography raised tumor detectability using ROC curve analysis reported that the combined techniques do significantly raise specificity in detecting hepatic tumors and accuracy in characterizing hepatic tumors, although combining CT hepatic arteriography with CT during arterial portography does not significantly increase sensitivity for detection of hepatic tumors [12]. Another group of researchers concluded that the combination has significantly higher sensitivity than either CT during arterial portography or CT hepatic arteriography alone for revealing hypervascular hepatocellular carcinoma smaller than 20 mm in diameter [13]. However, all these groups of researchers performed CT hepatic arteriography by choosing one hepatic artery, and their study populations included some patients in whom whole-liver CT hepatic arteriography was not obtained because multiple hepatic arteries were present. This failure to examine all hepatic arteries might have decreased the additional value of combined CT hepatic arteriography in the previous studies. On the other hand, our results indicate the statistically significant superiority of combined CT during arterial portography and CT hepatic arteriography over CT during arterial portography alone in the ROC curve analysis for two of the three observers. Furthermore, the value of the area under the ROC curve, relative sensitivity, specificity, and accuracy with CT hepatic arteriography alone was higher than with CT during arterial portography alone. These observations suggest that whole-liver CT hepatic arteriography obtained by opacifying every hepatic artery may have a higher diagnostic value than CT during arterial portography alone compared with CT hepatic arteriography performed by selecting just a single hepatic artery.
The three radiologists who reviewed the CT images frequently encountered various types of false-positive findings on CT during arterial portography or CT hepatic arteriography that were depicted as focal perfusion abnormalities, presumably caused by cystic venous drainage [21], aberrant right gastric venous drainage [22], nonportal supply via the parabiliary venous system [23], focal arterioportal shunt [8], and rib compression effect [24]. The three radiologists in our study were knowledgeable about such pseudolesions and were experienced in differentiating them, and their performances as observers were satisfactory. Generally, some experience in interpreting angiographically assisted CT images may be necessary to efficiently differentiate small pseudolesions from small hypervascular neoplasms. Regarding false-negative findings, some segmental or subsegmental portal perfusion defects on CT during arterial portography probably caused by the portal venous obstruction or portal venous laminar flow obscured some metastases. Uneven opacification of the liver on CT hepatic arteriography probably caused by the hepatic artery laminar flow obscured some lesions. Also, tiny subcapsular metastases were commonly hard to detect on CT during arterial portography or CT hepatic arteriography.
MR images can detect hepatic metastases without any contrast agents or with gadopentetate dimeglumine or tissue-specific contrast agents. The use of tissue-specific contrast agents may improve the accuracy of hepatic neoplasm detection [25], and such MR images are believed to be free from false-positive findings caused by benign perfusion abnormalities that are common with angiographically assisted CT [8, 9]. However, whether the accuracy of MR imaging with a tissue-specific contrast agent is better than that of CT during arterial portography alone is still controversial [25,26,27]. Combining CT hepatic arteriography with CT during arterial portography may achieve a tumor detectability higher than that of MR imaging with a tissue-specific contrast agent and enable tissue characterization by observing enhancement characteristics with CT hepatic arteriography. We need to further compare the diagnostic accuracy of combined CT during arterial portography and CT hepatic arteriography images obtained with the unified CT and angiography system with MR images.
This study has some limitations. We considered the patients who developed recurrent metastases after hepatic surgery to have no definite tumors in their residual liver at the time of preoperative angiographically assisted CT because the retrospective review indicated no abnormal imaging findings; however, tiny metastatic deposits that showed no perfusion abnormalities on angiographically assisted CT may have existed before they were radiographically detectable. Because statistical analyses were performed on a segment-by-segment basis and confidence levels of the individual lesions were not obtained, subtle imaging findings of some small lesions might have been ignored in the presence of well-demarcated metastases located in the same hepatic segment. Other potential limitations of this study include reading-order bias and recall bias; however, we believe that these biases were minimal because the three reading sessions were conducted in a random order, the intervals between the three sessions were at least 2 weeks, and the tumors evaluated were relatively small (mean, 25.8 mm).
In conclusion, we found that combined CT during arterial portography and CT hepatic arteriography using a unified CT and angiography system significantly improved the detectability of hepatic metastases from colorectal cancer compared with CT during arterial portography alone or CT hepatic arteriography alone. Our results encourage performing angiographically assisted CT using a unified CT and angiography system for preoperative assessment in patients with hepatic metastases from colorectal cancer.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Toyoda, T. Kumada, and Y. Sone Impact of a Unified CT Angiography System on Outcome of Patients with Hepatocellular Carcinoma Am. J. Roentgenol., March 1, 2009; 192(3): 766 - 774. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kakeda, Y. Korogi, Y. Miyaguni, J. Moriya, N. Ohnari, N. Oda, K. Nishino, and W. Miyamoto A Cone-Beam Volume CT Using a 3D Angiography System with a Flat Panel Detector of Direct Conversion Type: Usefulness for Superselective Intra-arterial Chemotherapy for Head and Neck Tumors AJNR Am. J. Neuroradiol., October 1, 2007; 28(9): 1783 - 1788. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
