| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, Universidad de Antioquia, Hospital Universitario San
Vicente de Paúl, Calle 64 x Kra. 51D,
Medellín, Colombia.
2
Department of Gastroenterology, Universidad de Antioquia, Hospital
Universitario San Vicente de Paúl,
Medellín, Colombia.
3
Department of Surgery, Universidad de Antioquia, Hospital Universitario San
Vicente de Paúl,
Medellín, Colombia.
Received November 4, 1999;
accepted after revision March 2, 2000.
Address correspondence to J. A. Soto.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Fifty-one patients referred for endoscopic retrograde cholangiography of suspected biliary stones were studied with unenhanced helical CT, MR cholangiography, and helical CT performed after oral administration of a cholangiographic contrast agent (iopodic acid). The studies were randomized for interpretation. Two radiologists evaluated the images by consensus and determined the presence and location of stones. We used retrograde cholangiography findings as the standard of reference. Sensitivity and specificity (with 95% confidence intervals [CIs]) of the three examinations were calculated and compared using the exact form of the McNemar test.
RESULTS. Bile duct stones were revealed with retrograde cholangiography in 26 patients (51%). Sensitivity was 65% (95% CI, 44.4-82%) for unenhanced helical CT, 92% (95% CI, 73-99%) for CT cholangiography, and 96% (95% CI, 78-99%) for MR cholangiography. Specificity was 84% (95% CI, 63-95%) for unenhanced helical CT, 92% (95% CI, 73-99%) for CT cholangiography, and 100% (95% CI, 83-100%) for MR cholangiography. The sensitivity of CT cholangiography and MR cholangiography was significantly higher than that of unenhanced helical CT (p < 0.01). Differences in specificity were not significant.
CONCLUSION. Our results indicate that oral contrast—enhanced CT cholangiography and MR cholangiography are significantly more sensitive than unenhanced helical CT for the detection of bile duct calculi.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
During the study period, 68 patients with clinical findings suggestive of choledocholithiasis were referred for endoscopic retrograde cholangiography. Twelve patients did not meet the inclusion or exclusion criteria. Two additional patients did not complete the MR imaging examination because of claustrophobia. In three patients, the scheduled retrograde cholangiography was not attempted (n = 1) or not completed (n = 2). In the remaining 51 patients (32 women, 19 men), all four studies were completed; these patients constituted the study population. Mean age was 53 years (age range, 18-84 years). The examinations were performed within a 48-hr period in the following order: unenhanced helical CT first, MR cholangiography next, and CT cholangiography last. Endoscopic retrograde cholangiography was performed after the tests under comparison were completed. Forty-four patients (86%) had a prior history of cholecystectomy. Forty-four patients had bilirubin levels less than 1 mg/dL, and seven patients had bilirubin levels between 1 and 5 mg/dL: 1.1, 1.4, 1.7, 1.8, 2.9, 3.6, and 4.1 mg/dL.
Unenhanced Helical CT and Oral Contrast-Enhanced CT
Cholangiography
All CT examinations were performed in a helical CT scanner (ProSpeed;
General Electric Medical Systems, Milwaukee, WI). Unenhanced helical CT
studies were obtained in a supine position after a fasting period of at least
4 hr. To define the area of interest to be included in the helical
acquisitions, low-dose axial images of the liver and pancreas were initially
obtained. Scanning parameters for these initial images were nonhelical
acquisition, 10-mm collimation, 10-mm table feed, 120 kVp, and 160 mAs.
Helical CT scans (unenhanced and CT cholangiographic scans) were obtained
during a single acquisition in a caudocranial direction, starting at the third
portion of the duodenum. Parameters used for helical CT acquisitions were 30-
to 36-sec scan time, one tube rotation per second at a current of 250 mAs and
120 kVp, 3-mm collimation, and 5-mm/sec table speed (pitch, 1.67). The field
of view for scanning depended on patient size and ranged from 280 to 380 mm.
Patients who were unable to hold their breath were instructed to breathe
quietly; although there were motion artifacts in six patients (12 CT
examinations) as a result of incomplete breath-holding, the quality of all CT
scans was considered adequate for diagnosis. The volumetric data set was
reconstructed at 1-mm intervals using a 180° linear interpolation
algorithm and a high-density kernel. For CT techniques, no additional oral or
IV contrast material was administered.
CT cholangiography was performed after acquisition of unenhanced helical CT and MR cholangiographic studies. The total dose of iopodic acid administered was 6 g (12 capsules of 500 mg each), divided into two doses of 3 g each, as described in a previous study [10]. Patients were instructed to take the first dose 2 hr after dinner and the second dose 2 hr later. CT cholangiograms were acquired between 7:00 and 9:00 the following morning. Patients were examined in a fasting state after cholecystectomy. In an attempt to induce physiologic contraction of the gallbladder and improve opacification of the common bile duct, a fatty meal was given to the six patients with gallbladder in situ 20-30 min before the CT examination. The fatty meal did not cause pain or other discomfort in these six patients. No other contrast agent was used. Patients were encouraged to drink abundant fluids for 4 days after the administration of iopodic acid to decrease the risk of renal tubular deposition of uric acid.
MR Cholangiography
MR examinations were performed on a 1.5-T system (ACS NT; Philips Medical
Systems, Best, The Netherlands) with a body coil. MR cholangiograms were
acquired using three different pulse sequences: breath-hold, single-shot
half-Fourier rapid acquisition with relaxation enhancement (in single- and
multislice modes), and non—breath-hold three-dimensional fast spin-echo
with respiratory triggering. At our institution, we prefer the single-shot
breath-hold techniques because of the short scanning times.
Non—breath-hold sequences are the preferred method for patients with
dyspnea and for patients in whom communication is difficult. The
three-dimensional fast spin-echo sequence is also acquired when breath-hold
images are of poor quality or are not diagnostic. Using these general
guidelines, the three MR sequences were acquired in 17 patients, only the
breath-hold sequences were acquired in 31 patients, and only the
non—breath-hold sequence was acquired in three patients who had
difficulty following breathing instructions. With this approach, the quality
of all MR examinations was considered adequate for interpretation.
Scanning parameters for the single-slice half-Fourier rapid acquisition with relaxation enhancement (RARE) sequence were infinite/300 (TR/TEeff), 30-mm slice thickness, one acquisition, 128 x 256 matrix, 128 echo train length, 9.9-msec echo spacing, and acquisition time of 2.5 sec per slice. For the multislice half-Fourier RARE sequence, parameters were infinite/290 (TR/TEeff), 5-mm slice thickness, 10 slices, one acquisition, 128 x 256 matrix, 128 echo train length, 9-msec echo spacing, and 20-sec acquisition time. Finally, for the three-dimensional fast spin-echo sequence we used the following parameters: 2000-2300/240 (TR range/TEeff), 2-mm partition thickness (40 partitions, eight slabs), two acquisitions, 128 x 256 matrix, 39-43 echo train length, 12-msec echo spacing, and nominal acquisition time of 5 min 40 sec to 6 min 20 sec. We used a chemically selective fat-saturation prepulse for the three sequences. A 65% partial K-space filling factor was applied for both half-Fourier RARE sequences. The three-dimensional fast spin-echo and multislice half-Fourier RARE sequences were acquired in a right anterior oblique—coronal plane. For the single-slice half-Fourier sequence, four different projections, each with a different orientation, were obtained. No oral contrast or antiperistaltic agent was administered.
Image Analysis
The source images obtained with CT and MR examinations were transferred to
separate workstations (General Electric Advantage Windows for CT studies and
Phillips Easy Vision for MR studies) for interpretation and postprocessing.
Postprocessing capabilities were similar for both workstations and included
maximum intensity pixel projection, shaded-surface display, and multiplanar
reformatting algorithms. All studies were interpreted in an interactive manner
by consensus of two radiologists at the workstations. Differences in
interpretation were resolved by conference between the two radiologists. No
clinical information or results of other diagnostic studies were provided to
the interpreting radiologists. Both radiologists were fellowship-trained in
cross-sectional abdominal imaging. They were asked to generate as many
two-dimensional and three-dimensional renderings as needed, but we did not
record how many projections were obtained per patient. For interpretation of
CT and MR imaging studies, the radiologists used the complete sets of source
images as well as the projectional renderings that they generated.
We randomized the sequence of interpretation of studies from different patients as well as the order of interpretation of the three studies from the same patient. This randomization was accomplished by an investigator who was not involved in image interpretation. For each of the three imaging techniques, the 51 examinations were independently and randomly divided into three groups of 17 each; thus, a total of nine groups (each with 17 examinations) were available for interpretation. Interpretation was carried out retrospectively over a period of 3 weeks. One set of studies per imaging technique was analyzed per week. During the first week the order was unenhanced helical CT first, MR cholangiography next, and oral CT cholangiography last. During the second week the order was MR cholangiography first, CT cholangiography next, and unenhanced helical CT last. Finally, during the third week the order was CT cholangiography first, unenhanced helical CT next, and MR cholangiography last. Data regarding identification of patients, such as name, age, and dates, were not presented with the images.
All studies were interpreted by the same two radiologists. For every study, the radiologists were asked to determine the presence and location (intrahepatic, common bile duct, or both) of stones. On unenhanced helical CT and CT cholangiography studies, stones were seen as intraductal foci with an attenuation coefficient that differed from that of the surrounding bile. The appearance of biliary stones on CT scans varied with the internal composition of the stone and the attenuation of the bile. The radiologists classified stones on unenhanced helical CT scans as follows: homogeneously hyperattenuating (calcified), soft-tissue attenuating, or hypoattenuating with a hyperattenuating rim. For CT cholangiography studies, stones were classified as hypo- or hyperattenuating compared with bile. On MR cholangiography, stones appeared as intraductal foci of absent or reduced signal intensity. The radiologists also measured stone size using electronic calipers and the software available with the workstations. If more than one stone was present, the smallest stone visible with each technique was measured and this measurement was recorded. For CT cholangiography examinations, the degree of biliary excretion of contrast material was determined by measuring the attenuation of opacified bile in the intrapancreatic portion of the common bile duct, using circular or oval regions of interest. These measurements were made from the axial source images by a radiologist who was also involved in image interpretation.
Endoscopic Retrograde Cholangiography
All endoscopic retrograde cholangiography procedures were performed by one
of three gastrointestinal endoscopists using a standard technique. Retrograde
cholangiograms were interpreted by consensus by the endoscopist who performed
the procedure and a radiologist who was not involved in interpretation of the
tests under comparison. No information about the findings of unenhanced
helical CT, CT cholangiography, or MR cholangiography was provided. The
interpreting physicians determined the presence, location, and size of stones.
For calculation of stone size, the outer caliber of the endoscope was used as
the correction factor for magnification. As done with the other imaging
techniques, the smallest stone visible on retrograde cholangiography was
measured. In one patient, retrograde cholangiography did not reveal filling
defects, but a 3-mm stone was retrieved after sphincterotomy. This patient was
considered to have a positive diagnosis of choledocholithiasis for the purpose
of statistical analysis. In all other patients with a positive diagnosis of
choledocholithiasis, the stones were well depicted on retrograde
cholangiography.
Statistical Analysis
The consensus interpretation provided by the radiologists was used to
determine sensitivity, specificity, and predictive values of each technique.
We used the interpretation of retrograde cholangiograms as the standard of
reference for these determinations. Ninety-five percent confidence intervals
(CIs) were calculated [18].
Data of sensitivity and specificity of the three techniques were compared
using the exact from of the McNemar test
[19]. A p value of
0.05 or less indicated a statistically significant difference.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
For unenhanced helical CT studies, the radiologists had 17 true-positive, nine false-negative, 21 true-negative, and four false-positive interpretations. For CT cholangiography, there were 24 true-positive, two false-negative, 23 true-negative, and two false-positive interpretations. Finally, for MR cholangiography, there were 25 true-positive, one false-negative, 25 true-negative, and no false-positive interpretations. According to these results, unenhanced helical CT had a sensitivity of 65% (95% CI, 44-82%), a specificity of 84% (95% CI, 63-95%), a positive predictive value of 81% (95% CI, 57-94%), and a negative predictive value of 70% (95% CI, 50-85%). For CT cholangiography, the sensitivity was 92% (95% CI, 73-99%), the specificity was 92% (95% CI, 73-99%), the positive predictive value was 92% (95% CI, 73-99%), and the negative predictive value was 92% (95% CI, 73-99%). Finally, for MR cholangiography, the sensitivity was 96% (95% CI, 78-99%), the specificity was 100% (95% CI, 84-100%), the positive predictive value was 100% (95% CI, 83-100%), and the negative predictive value was 96% (95% CI, 78-100%).
The appearance of bile duct stones in the 17 patients in whom they were identified on unenhanced helical CT was hypoattenuating on hyperattenuating rim (n = 7) (Fig. 1A,1B,1C), homogeneously hyperattenuating or calcified (n = 6) (Fig. 2A,2B,2C,2D), and soft-tissue attenuating (n = 4) (Fig. 3A,3B). False-negative interpretations of unenhanced helical CT studies were caused by not identifying stones that were isoattenuating with bile. Eight of these stones were located in the common bile duct (Fig. 4A,4B,4C,4D) and the remaining one in an intrahepatic duct in the left hepatic lobe. The size of stones that were not identified with unenhanced helical CT ranged from 3 to 12 mm (Fig. 4A,4B,4C,4D). False-positive results of unenhanced helical CT studies were caused by misinterpreting apparent intraductal density as stones (Fig. 5A,5B,5C).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Stones had lower attenuation than opacified bile in 21 (87.5%) of 24 patients with true-positive CT cholangiograms (Figs. 1A,1B,1C, 3A,3B, and 4A,4B,4C,4D) and higher attenuation in three patients (Fig. 2A,2B,2C,2D). In one patient, the single calcified stone was easier to detect on maximum-intensity-pixel-projection reformations than on axial sections viewed with regular soft-tissue window settings (Fig. 2A,2B,2C,2D). One false-negative interpretation of CT cholangiography occurred in a patient with isolated ductal dilatation and large stones in the left hepatic ducts (Fig. 6A,6B,6C). The contrast agent did not opacify the abnormal ducts, and the stones were not depicted. MR cholangiography clearly showed the stones in this patient, and they were confirmed with retrograde cholangiography (Fig. 6A,6B,6C). The other case of false-negative interpretation of CT cholangiography occurred in a patient in whom there was no biliary excretion of contrast material because of an elevated bilirubin level and in whom retrograde cholangiography showed a single stone in the common bile duct. False-positive interpretations of CT cholangiography were attributed to image noise resulting from large patient size. Noise can mimic intraductal density that can be mistaken for stones. The single false-negative interpretation of MR cholangiography occurred in a patient in whom only the non—breath-hold sequence was acquired; image quality was degraded by patient motion. Unenhanced helical CT and CT cholangiography showed a single 6-mm common bile duct stone, which was confirmed on retrograde cholangiography.
|
|
|
Biliary Excretion and Tolerance of Iopodic Acid
In four patients (8%), no biliary excretion of cholangiographic contrast
agent was evident. Two of these patients had elevated bilirubin levels (2.9
and 4.1 mg/dL) and this was likely the reason for the failure to excrete the
contrast agent. The third patient had vomiting 1 hr after the second dose of
the contrast agent. In the fourth patient, no definite cause could be
identified. In the remaining 46 patients, there was biliary opacification with
contrast and the mean attenuation of bile was 114 H (range, 65-243 H). Adverse
events occurred in five patients (10%) after administration of the contrast
agent. Two patients complained of epigastric pain, two had mild diarrhea, and
one had vomiting after ingestion of iopodic acid capsules. These symptoms were
self-limited in all patients and none required specific therapy.
Statistics
Using the McNemar test, we found significant differences between the
sensitivity of unenhanced CT and CT cholangiography (p = 0.008) and
between the sensitivity of unenhanced CT and MR cholangiography (p =
0.005). The difference between the sensitivity of CT cholangiography and MR
cholangiography was not significant (p = 0.32). No significant
difference was found between the specificity of the three tests
(p>0.05 for the three comparisons).
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Helical CT techniques have also been used for the detection of biliary calculi. For example, Neitlich et al. [7] used unenhanced thin-section helical CT with overlapping retrospective reconstruction and found a sensitivity of 88% and a specificity of 97%. However, as noted by Baron [22], the results of Neitlich et al. may be difficult to reproduce in a broader patient population; in fact, their experience has not been duplicated to date. In CT cholangiography, exogenous contrast agents are used to opacify the bile ducts [8,9,10]. This technique has been shown to also be highly effective for the diagnosis of bile duct stones. Cholangiographic contrast agents can be administered IV [8, 9] or orally [10]. The toxicity of IV agents limits their usefulness [23], and they are not currently approved for administration in the United States. Oral cholangiographic contrast agents are safe and are well tolerated by most patients; adverse reactions are rare and usually mild [24]. In a recent study [10], helical CT and iopanoic acid were used to produce cholangiograms in a group of patients with suspected choledocholithiasis with good results (sensitivity, 86-93%; specificity, 100%). A limitation of CT cholangiography is the insufficient opacification of bile ducts that may occur in patients with elevated bilirubin levels or liver insufficiency.
In this study, we compared the sensitivity and specificity for detection of bile duct stones of unenhanced helical CT, CT cholangiography obtained with an orally administered contrast agent, and MR cholangiography in the same patient population. For unenhanced helical CT, we used a technique similar to the one suggested by Neitlich et al. [7]. For CT cholangiography, we used iopodic acid as the contrast agent. Like most oral cholangiographic contrast agents, iopodic acid is a triiodobenze ring compound that is absorbed primarily in the jejunum and transported to the liver, where it is conjugated with glucuronic acid and excreted in bile [17, 24]. For MR cholangiography, we used three pulse sequences that are used in clinical practice for this indication. Sensitivity and specificity of MR cholangiography both exceed 90% [13,14,15,16] regardless of the technique used, and results of trials that compare the various sequences [25,26,27,28] are not conclusive as to which is most appropriate for the depiction of stones. In clinical practice, the choice of sequence depends on personal preference and possibilities offered by the equipment available at each center. At our institution, we prefer single-shot breath-hold techniques because of the short scanning times required. Non—breath-hold sequences are the preferred method for patients with dyspnea and for patients in whom communication of breath-holding instructions is difficult.
Our results show that the sensitivities for detection of bile duct stones of CT cholangiography performed with iopodic acid and of MR cholangiography are similar (>90%) and that both methods have a significantly higher sensitivity than unenhanced helical CT (sensitivity, 69%). The low sensitivity of unenhanced helical CT in our study resulted from the high frequency of false-negative interpretations when stones were isoattenuating with bile. Stone detection is easier with CT cholangiograms and MR cholangiograms because of the greater difference in attenuation and signal intensity, respectively, with surrounding bile. Although the high attenuation of bile achieved with cholangiographic contrast agents increases interpreter confidence for detection of stones, there is a theoretic risk that bile may become isoattenuating with calcified stones, resulting in false-negative interpretations of the CT cholangiogram. In our study, for example, the stones of three patients had higher attenuation than bile, but the difference in attenuation was enough to allow detection. We routinely use the maximum-intensity-pixel-projection algorithm for reconstruction of CT cholangiograms when review of the source images does not disclose hypoattenuating stones. Because this algorithm shows only the pixels with the highest CT attenuation, every hyperattenuating stone should be well seen with this approach. This occurred with one of the patients in our study.
The difference between our results and those of other studies [4,5,6,7] regarding sensitivity for stone detection with CT performed without cholangiographic contrast material has several possible explanations. Stone composition and size in the patient population of the various studies may vary. For example, Jeffrey et al. [4] reported a sensitivity of 90% with CT. However, the composition of stones in 13 (62%) of 21 patients with choledocholithiasis was predominantly calcium bilirubinate. Stones composed predominantly of calcium bilirubinate have higher attenuation and are easier to detect with CT. Calcified stones were seen in five (29%) of 17 patients in the series of Neitlich et al. [7] and in six (23%) of 26 patients in our series. At our institution, most patients with bile duct stones have cholesterol stones, as is true for most of the Western world [29]. Stones composed predominantly of calcium bilirubinate are more common in Eastern countries [30]. The impact that stone size may have on the difference in sensitivity between the studies is difficult to establish because in some of the earlier reports [4, 6] the size of confirmed stones is not clear. In the study by Baron et al. [5], 12 patients had common bile duct stones and all were 5 mm or larger in diameter. In our series, three patients had a single stone in the common bile duct measuring four mm or smaller and only one (a densely calcified stone) was properly detected with unenhanced CT.
The difference in sensitivity between our study and that of Neitlich et al. [7] (69% versus 88%) is more difficult to explain because the technique used is similar and no significant differences in patient population are apparent. In their study, images were presented to the observers with fixed narrow windows, and in our study the observers were free to use the window settings that they preferred because images were evaluated interactively at a workstation. Although this may affect the results, the observers in our study were asked to include narrow windows in their assessment. Another factor that may account for the differences between the two studies is observer experience. The observers who evaluated the images in the study of Neitlich et al. have extensive experience in the interpretation of unenhanced abdominal CT studies, mostly for the detection of ureteral stones [31, 32]. There is a learning curve associated with interpretation of every imaging technique. The greater amount of experience of the observers in the study of Neitlich et al. may be partly responsible for the higher sensitivity found in their study, especially considering that detection of biliary stones is more difficult than detection of ureteral stones.
CT cholangiography is limited by the requirement for hepatic uptake, conjugation, and excretion of the contrast agent to obtain adequate opacification of the bile ducts. Insufficient excretion of the bile ducts may occur in patients with elevated bilirubin levels or liver insufficiency. Four patients in our study had poor excretion of the contrast agent. In an additional patient with segmental intrahepatic duct dilatation and stones, no excretion was noted in the affected ducts. False-negative interpretations occurred in two of these patients. The number of examinations with poor biliary excretion of the contrast agent can be decreased by excluding patients with elevated bilirubin levels or known hepatic disease.
CT techniques offer advantages over MR cholangiography in certain situations. In MR cholangiograms, artifacts may arise from surgical clips or metallic elements located in the vicinity of the common bile duct. Because the clips themselves are not seen with MR imaging, the artifacts may be misinterpreted as stones [33, 34]. CT techniques show the metallic clips, and artifacts are minimal. In MR cholangiograms acquired in a coronal or oblique coronal plane, intraductal air (which is hypointense) may mimic stones. This pitfall can be partially avoided with axial images, acquired either by directly scanning in the axial plane or by multiplanar reformations of coronal raw data. Air rises to the nondependent aspect of the duct, and stones remain in the dependent side [34]. CT in the axial acquisition plane offers advantages because the attenuation of air differs from that of stones. This may be particularly useful for patients with biliary—enteric anastomoses, who have abundant air in their bile ducts. Further studies in this patient population are warranted. Some limitations are common to both CT cholangiography and MR cholangiography. For example, filling defects such as tumefactive sludge may be the source of false-positive interpretations and affect the specificity of both techniques.
In summary, our results show that the sensitivity of oral contrast—enhanced CT cholangiography and MR cholangiography for detection of bile duct stones is higher than that of unenhanced helical CT. Although increasing observer experience may improve the performance of unenhanced helical CT, interpretation of CT cholangiograms is easier because of the higher contrast between bile and stones. For detection of bile duct calculi with CT, we suggest using cholangiographic contrast agents because the rate of false-negative interpretations can be decreased significantly. Furthermore, CT cholangiography can replace MR cholangiography when it is either not available or contraindicated. However, use of CT cholangiography is limited when there is significant bilirubin level elevation or liver disease.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  E J Williams, J Green, I Beckingham, R Parks, D Martin, and M Lombard Guidelines on the management of common bile duct stones (CBDS) Gut, July 1, 2008; 57(7): 1004 - 1021. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. W. Anderson, E. Rho, and J. A. Soto Detection of Biliary Duct Narrowing and Choledocholithiasis: Accuracy of Portal Venous Phase Multidetector CT Radiology, May 1, 2008; 247(2): 418 - 427. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. C. Chan, B. N. Joe, F. V. Coakley, E. L. Prien Jr, R. G. Gould, S. Prevrhal, W. C. Barber, K. S. Kirkwood, A. Qayyum, and B. M. Yeh Gallstone Detection at CT in Vitro: Effect of Peak Voltage Setting. Radiology, November 1, 2006; 241(2): 546 - 553. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. W. Anderson, B. C. Lucey, J. C. Varghese, and J. A. Soto Accuracy of MDCT in the diagnosis of choledocholithiasis. Am. J. Roentgenol., July 1, 2006; 187(1): 174 - 180. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K S Gurusamy, M Farouk, and J H Tweedie UK guidelines for management of acute pancreatitis: is it time to change? Gut, September 1, 2005; 54(9): 1344 - 1345. [Full Text] [PDF]
|
|
|
|
|
|
J. S. Chen, B. M. Yeh, Z. J. Wang, J. P. Roberts, R. S. Breiman, A. Qayyum, and F. V. Coakley Concordance of Second-Order Portal Venous and Biliary Tract Anatomies on MDCT Angiography and MDCT Cholangiography Am. J. Roentgenol., January 1, 2005; 184(1): 70 - 74. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Rucker, C. O. Menias, and S. Bhalla Mimics of Renal Colic: Alternative Diagnoses at Unenhanced Helical CT RadioGraphics, October 1, 2004; 24(suppl_1): S11 - S28. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R Razzaq, MBchB, MRCP, FRCR and S A Sukumar Imaging of the jaundiced adult Imaging, September 1, 2004; 16(4): 287 - 300. [Abstract] [Full Text] [PDF]
|
|
|
|
|
