79
Downloaded from www.ajronline.org by 190.129.29.192 on 10/12/15 from IP address 190.129.29.192. Copyright ARRS. For personal use only; all rights reserved
CT of the Extrahepatic
Bile Ducts:
Wall Thickness and Contrast Enhancement in Normal and Abnormal Ducts
.a1 Scott J. Schulte1 Richard L. Baron1 Sharlene A. Teefey2 Charles A. Rohrmann, Jr.1 Patrick C. Freeny3 William P. Shuman1 Michael A. Foster1
Recent reports have described thickening and enhancement of the extrahepatic bile duct wall on CT scans obtained after administration of IV contrast material. We undertook this study to establish parameters for the normal thickness and enhancement of the bile duct wall on CT, and to develop a differential diagnosis for thickening of the duct wall. Routine CT examinations of 100 patients without biliary disease were evaluated pro-
spectively.
The common
hepatic
duct and common
bile duct could be visualized
in 66%
and 82% of cases, respectively; the walls of these ducts could be separately discerned in 59% and 52%. The mean thickness of the duct wall was 1 mm, with a maximal thickness of 1.5 mm. Wall enhancement was similar to (51%), slightly greater than (44%), or markedly greater than (5%) the enhancement of adjacent pancreatic parenchyma. A review of records covering a 5-year period identified 52 patients in whom CT showed thickening of the bile duct wall (2 mm). These patients could be categorized by seven underlying diseases, and analysis of the CT scans revealed four general patterns of thickening. Focal, concentric wall thickening in the distal common bile duct was associated with pancreatitis, pancreatic cancer, and common bile duct stones; focal, eccentric thickening tended to occur with cholangiocarcinoma and sclerosing cholangitis. Diffuse, concentric thickening was seen with acute cholangitis; diffuse,
eccentric
thickening
was associated
with
oriental
cholangiohepatitis
and sclerosing
cholangitis. Thickening of greater than 5 mm was seen only with cholangiocarcinoma. Enhancement of the duct wall in these groups varied and was of no predictive value. In summary, the extrahepatic bile ducts can be visualized in the majority of patients, and the normal duct wall should be 1.5 mm or less in thickness. Contrast enhancement of the duct wall occurs in patients without biliary tract disease and alone is not predictive of pathology. Pancreatitis, pancreatic cancer, common bile duct stones, cholangiocarcinoma, sclerosing cholangitis, acute cholangitis, and oriental cholangiohepatitis are associated with thickening of the duct wall. AJR
Received June 23, 1989; accepted after revision September 6, 1989. Presented at the annual meeting of the American Roentgen Ray Society, New Orleans, May 1989. 1 Department of Radiology, SB-05, University Hospital, Seattle, WA 98195. Address reprint requests to S. J. Schulte. 2 Department of Radiology (114), ZB-20, Veterans Administration Medical Center, 1660S. Columbian Way, Seattle, WA 98108. of Radiology, Virginia Mason Hospital, 925 Seneca St., Seattle, WA 98111. 0361 -803X/90/1 541-0079 © American Roentgen Ray Society
154:79-85,
January
1990
Technologic advances in CT have resulted in continued improvement in spatial resolution and in the ability of CT to evaluate the biliary system. Visualization of the normal common bile duct (CBD) by CT has been reported in 30-53% of individuals [1 , 2]. We suspect that the ability of CT to visualize the extrahepatic bile
ducts
has
improved
since
these
studies,
although
a more
recent
evaluation
using present CT technology has not been reported. In the past, most CT studies focused on visualization of dilated bile ducts and the characterization of the obstructing process. More recent reports have described the capability of CT to show changes in the bile duct wall; several reports have described thickening and/ or enhancement of the bile duct wall after IV administration of contrast material [38]. Despite these reports of pathologic thickening and contrast enhancement of the
duct
wall,
the
thickness
and
enhancement
characteristics
duct wall on CT have not been established, nor has a differential presented for duct wall thickening. This study was undertaken to address these issues and was
of the
normal
diagnosis divided
bile
been into two
Downloaded from www.ajronline.org by 190.129.29.192 on 10/12/15 from IP address 190.129.29.192. Copyright ARRS. For personal use only; all rights reserved
80
SCHULTE
basic components. The first was an evaluation of normal extrahepatic bile ducts to determine the frequency of CT visualization of the common hepatic duct (CHD) and CBD lumen and wall, and to establish parameters for normal thickness and contrast enhancement of the duct wall on CT. The second component was an analysis of abnormal extrahepatic bile ducts to define the processes that can cause thickening and contrast enhancement of the duct walls and to determine characteristic patterns to the thickening or enhancement that would allow differentiation among types of disease.
Subjects, Patients
Materials, Without
From
December
or Pancreatic
1 987 through
Disease March
1 988,
1 16 consecutive
for abdominal
CT and without
creatic
were evaluated
prospectively. CT examinations were (n = 68) or Picker 1 200 (n = 48) was optimized for patient body habitus; mA were used. Consecutive dynamic
disease
known
pa-
tients referred performed
biliary or pan-
on either GE 9800
scanners. 1 20-140
The scan technique kVp and 1 20-140
scans at 10-mm intervals with 5- or 10-mm collimation were obtained through the hepatobiliary system. IV contrast material (Reno-M-60, Squibb) was administered with a power injector, as a 2 mI/sec, 50 ml bolus, followed by 100 ml infused at 1 mI/sec. Oral contrast
material also was administered At the completion
of each examination,
the images
separate indicated
of the wall of the CBD after administration of IV was compared with that of the adjacent pancreatic and subjectively graded as 0, 1 , or 2. Grade 0 signified similar
the bile duct enhancement
to that
of pancreatic
parenchyma
or inability
after the CT examinations
were done, all patient
charts were reviewed or referring physicians were contacted to determine if interval biliary or pancreatic disease had developed. In 16 patients, biliary or pancreatic disease developed or no clinical follow-up was available. They were excluded from the study, resulting in 1 00
patients
in the
final
with Biliary
there was no clinical patients.
or Pancreatic
in five. At the time of or laboratory
evidence
Disease
To identify patients with thickening wall, medical, surgical, and radiologic
of the extrahepatic bile duct records from participating in-
stitutions from the preceding 5 years were reviewed, and available CT scans from patients with any potential biliary or pancreatic disease
were examined.
GE 8800, GE 9800, or Picker
1200 scanners
and
ments directly from the recorded CT image. On the basis of our findings of normal bile duct wall thickness, all CT examinations showing a duct wall 2 mm or thicker were included in the disease group. The studies were analyzed by two of the investigators and consensus was reached. The thickness of the duct wall was measured and the thickening was characterized as to its location and whether it was diffuse or focal and concentric or eccentric. The degree of duct wall enhancement was graded by using the same scale used for the normal
population.
Results
study
group.
There
were
63
men
and
Without
Biliary
or Pancreatic
Disease
In the group of 1 00 patients without biliary or pancreatic disease, the lumen of the CHD was identified in 66 patients and the lumen of the CBD in 82 patients. The mean diameter of the CHD lumen was 2.8 mm (range, 1 .3-8.0 mm) and the mean diameter of the CBD lumen was 3.6 mm (range, 1.51 0.9 mm). The wall of the CHD could be delineated in 59 patients (Fig. 1) and the CBD wall in 52 (Fig. 2). The mean thickness ofthe wall of the CHD was 0.94 mm, with a maximal thickness of 1 .5 mm. The mean thickness of the wall of the CBD was 1 .0 mm, with a maximal thickness of 1 .5 mm. Enhancement of the wall of the CBD was equal to the enhancement of adjacent pancreatic parenchyma or could not be distinguished from the pancreatic parenchyma (grade 0) in 51 patients, slightly greater than that of pancreatic parenchyma (grade 1) in 44, and markedly greater than that of pancreatic parenchyma (grade 2) in five.
to
wall from the surrounding pancreas, grade 1 slightly greater than that of pancreatic paren-
chyma, and grade 2 indicated enhancement markedly greater than that of pancreatic parenchyma. Grading of the CHD wall was not attempted because of the lack of adjacent reference tissue. Nine to 1 2 months
Patients
Patients
were reviewed,
Enhancement contrast material enhancement
in nine and lost to further follow-up
the last clinical evaluation, of biliary disease in these
routinely.
and the extrahepatic bile ducts were assessed at two levels: the CHD at the level of the porta hepatis and the CBD at the level of the pancreatic head. These ducts were identified as tubular structures (lumen density equivalent to water) detected on serial scans in the expected location of the bile ducts: the CHD was anterior to the portal vein, lateral to the hepatic artery, and surrounded by fat in the porta hepatis, and the CBD was within or adjacent to the parenchyma of the pancreatic head. At each level, the presence or absence of the duct was assessed; if visualized, the images were magnified two or three times, and, by using digital calipers, the lumen diameter (inner to inner wall) and wall thickness were measured. Measurements were taken along the short axis of the duct lumen and wall to avoid falsely elevated measurements related to oblique scanning [2]. The standard and magnified images, including those with the cursor measurements, were filmed and reviewed later by three of the investigators with consensus agreement on each measurement.
parenchyma
evaluation
AJR:154, January 1990
conventional abdominal scanning techniques had been used. The thickness of the bile duct wall was measured by using a magnifying glass and custom-made rulers capable of measuring 0.5-mm incre-
and Methods
Biliary
ET AL.
37
women 20-89 years old (mean, 55). Of these 100 patients, 6-month clinical follow-up or autopsy evaluation was available for 86. Maximal follow-up for the remaining study patients was 3-6 months in 10, 13 months in three, and 2 weeks in one. Reasons for the shorter
follow-up courses were death from a nonbiliary cause without autopsy
Patients
with Billary
or Pancreatic
Disease
Bile duct walls 2 mm thick or greater were present in 52 patients with proved diagnoses. There were seven underlying diseases, which could be divided into eight categories. Nonoperated pancreatic carcinoma.-The walls of the bile ducts were thickened in eight patients with pancreatic carcinoma who had not undergone surgical bypass. In all cases, the thickening was focal, occurring within or just above the pancreatic head (Fig. 3). The wall thickening was concentric in all cases except one, in which it was minimally eccentric. The walls ranged from 2.5 to 3 mm thick. Wall enhancement was grade 1 in all cases. Surgically treated pancreatic carcinoma. -The bile duct walls were thickened in two patients with pancreatic carci-
Downloaded from www.ajronline.org by 190.129.29.192 on 10/12/15 from IP address 190.129.29.192. Copyright ARRS. For personal use only; all rights reserved
AJA:154, January 1990
CT
OF
EXTRAHEPATIC
BILE
DUCTS
81
Fig. 1.-CT scan shows normal common hepatic duct at level of porta hepatis. Lumen (density equivalent to water) is surrounded by a thin but perceptible duct wall (arrows).
Fig. 2.-CT scan shows normal common bile duct within pancreatic parenchyma. Lumen (density equivalent to water) is surrounded by thin, mildly enhancing duct wall (arrow).
Fig 3.-CT scan shows pancreatic carcinoma enlarging head of pancreas and surrounding distal common bile duct. There is mild eccentric thickening and contrast enhancement of duct wall (arrows),
which
was focal
and identified
within
pan-
creatic head only.
Fig. 4.-Marked eccentric thickening of common hepatic duct wall due to cholangiocarcinoma. A, CT scan shows eccentric thickening of common hepatic duct wall posteromedially (arrows), displacing duct lumen (density equivalent to water) from portal vein posterioriy. B, Cholangiogram shows eccentric narrowing of common hepatic duct lumen (arrows), with
obliteration just distal to narrowing. (Modified and reprinted with permission from Teefey et al. [5].)
noma who previously had undergone Whipple procedures with choledochojejunostomies. In one case, the thickening was diffuse, involving the remaining extrahepatic bile ducts. In the other case, the thickening was focal, occurring near the
anastomosis.
In both
cases,
the
thickening
was
concen-
tric. Walls were 3.5 and 4 mm thick; wall enhancement was assigned grades 1 and 2. Neoplasms of bile ducts. -This group comprised 10 patients with cholangiocarcinoma and one with a benign papillary neoplasm. Focal thickening of the duct wall at the level of the neoplastic mass was present in all cases (Figs. 4 and 5). Eccentric thickening occurred in nine and concentric thicken-
ing in two
Fig. 5.-CT ening
and
mild
scan shows subtle eccentric thickcontrast
common hepatic duct cholangiocarcinoma. this level only.
of the cases.
The
range
enhancement of wall of (arrows) In a patient with Thickening was present at
of duct
wall
thickening
was
3-1 1 mm. Wall enhancement was grade 0 in three, grade 1 in seven, and grade 2 in one. Acute cholangitis.-Bile duct walls were thickened in six patients with proved acute cholangitis. In two cases CBD stones were known to be present. The thickening was diffuse in all cases but one, in which focal thickening was seen at the level of a CBD stone. The thickening was concentric in all cases and was from 2 to 4 mm thick. Duct wall enhancement was grade 1 in four and grade 2 in two cases (Fig. 6). Oriental cholangiohepatitis.-Duct walls were thickened in three patients with oriental cholangiohepatitis. The thickening
82
SCHULTE
ET AL.
AJR:154, January 1990
Fig. 6.-CT
scan
shows
concentric
thickening
Downloaded from www.ajronline.org by 190.129.29.192 on 10/12/15 from IP address 190.129.29.192. Copyright ARRS. For personal use only; all rights reserved
and mild contrast enhancement of wall of common bile duct (arrows) in a patient with acute cholan. gitis. This thickening and enhancement was present throughout extrahepatic bile ducts.
Fig.
7.-CT
scan
shows
eccentric
thickening
and mild contrast enhancement of wall of common hepatic duct (arrows), which was present at multiple levels, in a patient with oriental cholangiohepatitis. Duct lumen is dilated markedly and contains stones and debris.
A
B
Fig. 8.-Focal concentric thickening and marked contrast enhancement of distal common bile duct wall in patient with choledocholithiasis. A, CT scan of intrapancreatic common bile duct, well above level of stone, reveals no discemible duct wall (arrows). B, Scan 2 cm inferior to A shows marked concentric thickening duct wall (arrows).
and
contrast enhancement
of
C, Delayed, small-field-of-view scan with 3-mm collimation, obtained 1 cm below B, shows a decrease In contrast enhancement of thickened duct wall (arrows), as well as a soft-tissue density stone within duct lumen. D, Cholangiogram
shows
stricture
of distal com-
mon bile duct (arrows) just proximal to visualized stone. Wall thickening seen on CT corresponds to region of stricture demonstrated here.
was diffuse and eccentric in all cases (Fig. 7). Thickening was 3 mm in one and 4 mm in two. All had grade 1 duct wall enhancement. CBD stones without cholangitis. -This group comprised nine patients with CBD stones without associated cholangitis or oriental cholangiohepatitis. The duct wall thickening was
focal in eight cases and diffuse in one. In six patients the CBD stones were visualized on CT; in these cases focal thickening was identified only at the level of the stones, which were in the distal CBD (Fig. 8). In the other two cases of focal thickening, the thickening was also in the distal CBD; however, the stones were not visualized definitely on CT. The
AJA:154,
January
CT
1990
OF
EXTRAHEPATIC
BILE
DUCTS
83
Downloaded from www.ajronline.org by 190.129.29.192 on 10/12/15 from IP address 190.129.29.192. Copyright ARRS. For personal use only; all rights reserved
Fig. 9.-CT scan shows marked contrast enhancement of eccentrically thickened distal wall of common bile duct (arrows) in a patient with sclerosing cholangitis.
Fig. 10.-CT scan shows marked contrast enhancement and concentric thickening of distal wall of common bile duct (arrow) in a patient with chronic pancreatitis, as evidenced by a markedly dilated pancreatic duct. Thickening and enhancement
were only present distally.
thickening was concentric in seven and eccentric in two. The wall thickening ranged from 2 to 4 mm. Wall enhancement was grade 0 in one, grade 1 in seven, and grade 2 in one. Sclerosing cholangitis. -Six patients were identified with sclerosing cholangitis and thickening of the duct wall. The thickening was diffuse in three and focal in three. Eccentric thickening was present in all but one case and ranged from 2.5 to 5 mm thick. Wall enhancement was grade 0 in three, grade 1 in one, and grade 2 in two (Fig. 9). Pancreatitis. -Seven patients with pancreatitis had thickening of the duct wall. In all cases, the thickening was focal and concentric, occurring in the distal CBD within or just above the inflamed pancreatic head (Fig. 10). Thickening ranged from 2 to 3 mm. Wall enhancement was grade 0 in one, grade 1 in five, and grade 2 in one.
Analysis
of the Pattern
of Duct
Wall Thickening
From these data it is evident that there are four general patterns of duct wall thickening and that these patterns are associated with certain diseases. Focal and concentric thickening in the distal common bile duct occurred only in patients with pancreatitis (7/7), pancreatic carcinoma (7/8), CBD stones (6/9), and acute cholangitis (1/6). Focal and eccentric thickening occurred in patients with neoplasms of the bile duct wall (9/1 1), sclerosing cholangitis (2/6), CBD stones (2/9), and pancreatic cancer (1/8). Only neoplasms of the bile duct wall were associated with thickening greater than 5 mm. Diffuse and concentric thickening was noted in patients with acute cholangitis (5/6), surgically treated pancreatic carcinoma (1/2), and CBD stones (1/9). Diffuse and eccentric thickening was detected in cases of oriental cholangiohepatitis (3/3) and sclerosing cholangitis (3/6).
Discussion Current CT technology provides sufficient anatomic resolution to visualize the extrahepatic bile ducts in the majority of patients. Using a bolus, rapid-infusion technique for admin-
istration of IV contrast material and dynamic scanning, we were able to visualize the CHD and CBD lumens in 66% and 82%, respectively, of 1 00 patients without biliary disease. These values are higher than previously reported [1 , 2] and reflect the continued improvement in CT performance and resolution. The CHD and CBD walls could be distinguished in 59% and 52% of these patients, respectively, and by using magnified
images
and digital
calipers,
the wall thickness
could
be measured easily and accurately in most cases. In these 1 00 patients without biliary tract disease, the mean thickness of the duct wall was 1 mm; no duct walls were more than 1.5 mm thick. Recent reports have described several cases of bile duct wall thickening and enhancement with IV contrast material, evident on CT. Five cases of thickening were found in AIDS cholangitis, although CT was noted to be less successful than sonography in showing this finding [3]. One case of thickening and one case of duct wall enhancement in AIDS cholangitis were noted in another report [4]. Seven cases of duct wall thickening have been described in sclerosing cholangitis [5, 6]. In one of these studies [5], 1 1 cases of duct wall enhancement in sclerosing cholangitis and two cases of duct wall thickening in cholangiocarcinoma were also reported. Five cases of contrast enhancement of the duct wall in oriental cholangiohepatitis have also been reported [7]. The degree of thickening or contrast enhancement of the bile duct wall was not quantified precisely in these reports, nor were the specific criteria for abnormal duct wall thickening or enhancement indicated. Our study provides an objective standard for normal thickness of the bile duct wall on CT and establishes a value of 1 .5 mm, above which the thickness of the duct wall should be considered abnormal. This study also establishes the enhancement charactenstics of the wall of the normal CBD. In the 1 00 patients without biliary tract disease, CBD wall enhancement after administration of IV contrast material varied. CBD wall enhancement was similar to that of the pancreatic parenchyma (grade 0) in 51 %, slightly greater than that of the pancreatic parenchyma (grade 1) in 44%, and markedly greater than that of the pancreatic parenchyma (grade 2) in 5%. In the patients with
Downloaded from www.ajronline.org by 190.129.29.192 on 10/12/15 from IP address 190.129.29.192. Copyright ARRS. For personal use only; all rights reserved
84
SCHULTE
biliary or pancreatic disease and duct wall thickening, wall enhancement occurred with greater frequency than in the normal group: grade 0, 1 5%; grade 1 , 70%; and grade 2, 1 5%. This may in part be biased by selection of grades 1 and 2 enhancing walls during collection of the abnormal cases, as thickened walls that enhance greater than the pancreas are visualized more readily. On the other hand, it is also possible that the enhancement reflects the pathologic process. Regardless, both grades 1 and 2 enhancement occurred in patients without any evidence of biliary or pancreatic disease, and therefore reliance on enhancement alone is an unreliable predictor
of biliary
disease.
The abnormal bile duct component of this study had several limitations. First, because of the retrospective design, a vanety of CT scanners and techniques had been used. The limited quality of many of the scans and the lack of standardization did limit evaluation of the CT scans; however, it did not falsely produce duct wall thickening or significantly alter the pattern of thickening. Second, because of the method of collecting abnormal cases, we were unable to determine the frequency of duct wall thickening within each disease category. Despite these limitations, our study does produce useful information about the abnormal duct wall. Our review of CT examinations from patients with any potential biliary or pancreatic disease revealed several cases of thickening of the duct wall in each of the following categories: pancreatitis, pancreatic carcinoma, CBD stones, bile duct wall neoplasms, acute cholangitis, oriental cholangiohepatitis, and sclerosing cholangitis. Within these select groups of patients with abnormal thickening of the duct wall, we found general trends in the pattern of thickening, and certain patterns of thickening were associated with specific diseases. Enhancement of the duct wall occurred throughout the abnormal groups, with no characteristic trends in the degree of enhancement. The pattern offocal, concentric thickening in the distal CBD was the characteristic pattern of thickening in patients with pancreatitis, pancreatic cancer, and CBD stones. All patients with pancreatitis and duct wall thickening had this pattern of thickening (Fig. 1 0). The cause of this thickening with pancreatitis is speculative, as little histopathologic information is available about the bile duct wall. It is known that pancreatitis produces peniductal inflammation and fibrosis of the intrapancreatic portion of the CBD [9, 1 0], and this could extend into and thicken the adjacent duct wall. Pancreatitis is also known to result in strictures of the distal CBD [1 0-1 3], and partial obstruction may contribute to the wall thickening. This pattern of concentric thickening in the distal CBD was present also in all but one patient with pancreatic carcinoma and duct wall thickening; in the one exception, the thickening was in a similar location but slightly eccentric (Fig. 3). This thickening could be the result of associated pancreatitis, or could be related to partial obstruction of the CBD. Other possible causes include neoplastic infiltration of the duct wall and lymphedema. The pattern of concentric thickening in the distal CBD was present in six of nine patients with CBD stones (Fig. 8). In two, the thickening was focal in the distal CBD, but eccentric; in one patient, the thickening was diffuse and concentric.
ET AL.
AJR:154,
January 1990
Again, the cause of the thickening is unknown, as no histopathologic studies are available. Mechanical irritation from the stones could result in inflammatory and fibrotic changes in the duct wall, and in all six cases in which stones were visualized on CT, the thickening was at the level of the stones. Partial obstruction and associated pancreatitis may also be contributory; in one patient with CBD stones, pancreatitis was clinically evident. In the one case in which the thickening was diffuse, multiple stones and debris were present in the CBD, which may account for the diffuse nature of the thickening. Although cholangitis was not evident clinically, it is also possible that subclinical cholangitis was present. Focal and eccentric thickening of the duct wall was the predominant pattern with neoplasms of the bile duct wall only; within this category the thickening was focal in all cases and eccentric in nine of 1 1 . The greatest degree of wall thickening, up to 1 1 mm, occurred in this group, and thickening greater than 5 mm was found in this group only. The thickening was at the level of the neoplasm in all cases. The neoplastic mass may obliterate the lumen (density equivalent to water), and appear as a solid mass on CT; however, thin collimated sections just above this level may reveal thickening of the duct wall, presumably the result of tumor infiltration (Fig. 4). Focal and eccentric thickening of the duct wall also occurred in patients with sclerosing cholangitis, CBD stones, and pancreatic cancer; however, this pattern was present in a minority of cases, and the thickening was 5 mm or less. Diffuse and concentric thickening of the duct wall was characteristic of acute cholangitis (five of six cases); this was the only category in which this pattern predominated (Fig. 6). This pattern correlates with the expected pathologic changes, as acute cholangitis is usually a diffuse process, known to result in an inflamed, thickened duct wall [14]. In contrast to acute cholangitis, the wall thickening in patients with oriental cholangiohepatitis was diffuse but eccentric in all cases (Fig. 7). This reflects the recurrent chronic nature of the process, with areas undergoing various degrees of inflammation and fibrosis [1 4, 1 5]. Sclenosing cholangitis was also associated with a diffuse and eccentric pattern of duct wall thickening in three of six cases. Differentiation was possible because of the marked ductal dilatation, intraluminal debris, and stones present in all three cases of oriental cholangiohepatitis. Sclerosing cholangitis was associated with both diffuse and focal thickening of the duct wall that was nearly always eccentric (Fig. 9). The cholangiographic findings of bile duct wall irregularities, diverticula, and nodules are well known [1 6], and these changes may contribute to the thickening of the duct wall evident on CT [5]. Histopathologic examination reveals nonspecific, chronic inflammatory change and fibrosis within the walls of the bile duct [14, 17, 18]. In summary, the extrahepatic bile ducts can be visualized and characterized in a majority of patients. Our analysis of patients without biliary or pancreatic disease showed that no duct walls were more than 1 .5 mm thick; this establishes a value above which the duct wall should be considered abnormal. Enhancement of the bile duct wall was found to occur in patients without biliary tract disease; this establishes that
CT
Downloaded from www.ajronline.org by 190.129.29.192 on 10/12/15 from IP address 190.129.29.192. Copyright ARRS. For personal use only; all rights reserved
AJR:154, January 1990
OF
EXTRAHEPATIC
enhancement alone is not predictive of pathology. Several diseases were found to be associated with thickening of the duct wall, and a differential diagnosis for thickening is established. Four general patterns of thickening were identified, and there appears to be an association between each pattern and specific diseases. Enhancement of the duct wall occurred throughout the disease categories, with no specific trends to the degree of enhancement.
REFERENCES 1 . Foley WD, Wilson CR, Quiroz extrahepatic biliary tract Tomogr 1980;4:48-52
with
FA, Lawson computed
TL. Demonstration of the normal tomography. J Comput Assist
2. Co CS, Shea WJ, Goldberg HI. Evaluation of common bile duct diameter using high resolution computed tomography. J Comput Assist Tomogr 1986;10:424-427 3. Dolmatch BL, Laing FC, Federle MP, Jeffrey RB, Cello J. AIDS-related cholangitis: radiographic findings in nine patients. Radiology 1987;163: 31 3-316 4. Radin DR, Cohen H, Halls JM. Acalculous inflammatory disease of the biliary tree in acquired immunodeficiency syndrome: CT demonstration. J Comput Assist Tomogr 1987;1 1:775-778 5. Teefey SA, Baron RL, Rohrmann CA, Shuman WP, Freeny PC. Sclerosing cholangitis: CT findings. Radiology 1988;169:635-639
BILE
DUCTS
6. Carroll BA, Oppenheimer stration
85
DA. Sclerosing cholangitis:
of bile duct wall thickening.
AJR 1982;139:
sonographic
demon-
1016-1018
7. Chan F-L, Man S-W, Leong LLY, Fan S-T. Evaluation ofrecurrent pyogenic cholangitis with CT: analysis of 50 patients. Radiology 1989;170: 165-1 69 8. Baron AL. Common bile duct stones: reassessment of criteria for CT diagnosis. Radiology 1987;162:419-424 9. Sarles H, Sarles J-C, Camatte A, et al. Observations on 205 confirmed cases of acute pancreatitis, recurring pancreatitis, and chronic pancreatitis. Gut 1965;6:545-549 10. Littenberg G, Afroudakis A, Kaplowitz N. Common bile duct stenosis from chronic pancreatitis: a clinical and pathologic spectrum. Medicine (Baltimore) 1979;58:385-412 1 1 . Scott J, Summerfield JA, Elias E, Dick A, Sherlock S. Chronic pancreatitis: a cause of cholestasis. Gut 1977;18:196-201 12. Rohrmann CA Jr, Baron AL. Biliary complications of pancreatitis. Radiol Olin North Am 1989;27:93-104 13. Sarles H, Sahel J. Cholestasis and lesions of the biliary tract in chronic pancreatitis. Gut 1978;19:851-857 14. Schiff L, Schiff ER, eds. Diseases of the liver. Philadelphia: Lippincott, 1987 15. Chou ST. Chan CW. Recurrent pyogenic cholangitis: a necropsy study. Pathology
1980;12:415-428
16. MacCarty AL, LaRusso NF, Wiesner RH, Ludwig J. Primary sclerosing cholangitis: findings on cholangiography and pancreatography. Radiology 1983;149:39-44
17. LaRusso NF, Wiesner RH, Ludwig J, MacCarty AL Primary scierosing cholangitis. N EnglJ Med 1984;310:899-903 18. Thorpe MEC, Scheuer PJ, Sherlock S. Primary sclerosing cholangitis, the biliary tree, and ulcerative colitis. Gut 1967;8:435-448
Downloaded from www.ajronline.org by 190.129.29.192 on 10/12/15 from IP address 190.129.29.192. Copyright ARRS. For personal use only; all rights reserved
CT of the Extrahepatic
Bile Ducts:
Wall Thickness and Contrast Enhancement in Normal and Abnormal Ducts
.a1 Scott J. Schulte1 Richard L. Baron1 Sharlene A. Teefey2 Charles A. Rohrmann, Jr.1 Patrick C. Freeny3 William P. Shuman1 Michael A. Foster1
Recent reports have described thickening and enhancement of the extrahepatic bile duct wall on CT scans obtained after administration of IV contrast material. We undertook this study to establish parameters for the normal thickness and enhancement of the bile duct wall on CT, and to develop a differential diagnosis for thickening of the duct wall. Routine CT examinations of 100 patients without biliary disease were evaluated pro-
spectively.
The common
hepatic
duct and common
bile duct could be visualized
in 66%
and 82% of cases, respectively; the walls of these ducts could be separately discerned in 59% and 52%. The mean thickness of the duct wall was 1 mm, with a maximal thickness of 1.5 mm. Wall enhancement was similar to (51%), slightly greater than (44%), or markedly greater than (5%) the enhancement of adjacent pancreatic parenchyma. A review of records covering a 5-year period identified 52 patients in whom CT showed thickening of the bile duct wall (2 mm). These patients could be categorized by seven underlying diseases, and analysis of the CT scans revealed four general patterns of thickening. Focal, concentric wall thickening in the distal common bile duct was associated with pancreatitis, pancreatic cancer, and common bile duct stones; focal, eccentric thickening tended to occur with cholangiocarcinoma and sclerosing cholangitis. Diffuse, concentric thickening was seen with acute cholangitis; diffuse,
eccentric
thickening
was associated
with
oriental
cholangiohepatitis
and sclerosing
cholangitis. Thickening of greater than 5 mm was seen only with cholangiocarcinoma. Enhancement of the duct wall in these groups varied and was of no predictive value. In summary, the extrahepatic bile ducts can be visualized in the majority of patients, and the normal duct wall should be 1.5 mm or less in thickness. Contrast enhancement of the duct wall occurs in patients without biliary tract disease and alone is not predictive of pathology. Pancreatitis, pancreatic cancer, common bile duct stones, cholangiocarcinoma, sclerosing cholangitis, acute cholangitis, and oriental cholangiohepatitis are associated with thickening of the duct wall. AJR
Received June 23, 1989; accepted after revision September 6, 1989. Presented at the annual meeting of the American Roentgen Ray Society, New Orleans, May 1989. 1 Department of Radiology, SB-05, University Hospital, Seattle, WA 98195. Address reprint requests to S. J. Schulte. 2 Department of Radiology (114), ZB-20, Veterans Administration Medical Center, 1660S. Columbian Way, Seattle, WA 98108. of Radiology, Virginia Mason Hospital, 925 Seneca St., Seattle, WA 98111. 0361 -803X/90/1 541-0079 © American Roentgen Ray Society
154:79-85,
January
1990
Technologic advances in CT have resulted in continued improvement in spatial resolution and in the ability of CT to evaluate the biliary system. Visualization of the normal common bile duct (CBD) by CT has been reported in 30-53% of individuals [1 , 2]. We suspect that the ability of CT to visualize the extrahepatic bile
ducts
has
improved
since
these
studies,
although
a more
recent
evaluation
using present CT technology has not been reported. In the past, most CT studies focused on visualization of dilated bile ducts and the characterization of the obstructing process. More recent reports have described the capability of CT to show changes in the bile duct wall; several reports have described thickening and/ or enhancement of the bile duct wall after IV administration of contrast material [38]. Despite these reports of pathologic thickening and contrast enhancement of the
duct
wall,
the
thickness
and
enhancement
characteristics
duct wall on CT have not been established, nor has a differential presented for duct wall thickening. This study was undertaken to address these issues and was
of the
normal
diagnosis divided
bile
been into two
Downloaded from www.ajronline.org by 190.129.29.192 on 10/12/15 from IP address 190.129.29.192. Copyright ARRS. For personal use only; all rights reserved
80
SCHULTE
basic components. The first was an evaluation of normal extrahepatic bile ducts to determine the frequency of CT visualization of the common hepatic duct (CHD) and CBD lumen and wall, and to establish parameters for normal thickness and contrast enhancement of the duct wall on CT. The second component was an analysis of abnormal extrahepatic bile ducts to define the processes that can cause thickening and contrast enhancement of the duct walls and to determine characteristic patterns to the thickening or enhancement that would allow differentiation among types of disease.
Subjects, Patients
Materials, Without
From
December
or Pancreatic
1 987 through
Disease March
1 988,
1 16 consecutive
for abdominal
CT and without
creatic
were evaluated
prospectively. CT examinations were (n = 68) or Picker 1 200 (n = 48) was optimized for patient body habitus; mA were used. Consecutive dynamic
disease
known
pa-
tients referred performed
biliary or pan-
on either GE 9800
scanners. 1 20-140
The scan technique kVp and 1 20-140
scans at 10-mm intervals with 5- or 10-mm collimation were obtained through the hepatobiliary system. IV contrast material (Reno-M-60, Squibb) was administered with a power injector, as a 2 mI/sec, 50 ml bolus, followed by 100 ml infused at 1 mI/sec. Oral contrast
material also was administered At the completion
of each examination,
the images
separate indicated
of the wall of the CBD after administration of IV was compared with that of the adjacent pancreatic and subjectively graded as 0, 1 , or 2. Grade 0 signified similar
the bile duct enhancement
to that
of pancreatic
parenchyma
or inability
after the CT examinations
were done, all patient
charts were reviewed or referring physicians were contacted to determine if interval biliary or pancreatic disease had developed. In 16 patients, biliary or pancreatic disease developed or no clinical follow-up was available. They were excluded from the study, resulting in 1 00
patients
in the
final
with Biliary
there was no clinical patients.
or Pancreatic
in five. At the time of or laboratory
evidence
Disease
To identify patients with thickening wall, medical, surgical, and radiologic
of the extrahepatic bile duct records from participating in-
stitutions from the preceding 5 years were reviewed, and available CT scans from patients with any potential biliary or pancreatic disease
were examined.
GE 8800, GE 9800, or Picker
1200 scanners
and
ments directly from the recorded CT image. On the basis of our findings of normal bile duct wall thickness, all CT examinations showing a duct wall 2 mm or thicker were included in the disease group. The studies were analyzed by two of the investigators and consensus was reached. The thickness of the duct wall was measured and the thickening was characterized as to its location and whether it was diffuse or focal and concentric or eccentric. The degree of duct wall enhancement was graded by using the same scale used for the normal
population.
Results
study
group.
There
were
63
men
and
Without
Biliary
or Pancreatic
Disease
In the group of 1 00 patients without biliary or pancreatic disease, the lumen of the CHD was identified in 66 patients and the lumen of the CBD in 82 patients. The mean diameter of the CHD lumen was 2.8 mm (range, 1 .3-8.0 mm) and the mean diameter of the CBD lumen was 3.6 mm (range, 1.51 0.9 mm). The wall of the CHD could be delineated in 59 patients (Fig. 1) and the CBD wall in 52 (Fig. 2). The mean thickness ofthe wall of the CHD was 0.94 mm, with a maximal thickness of 1 .5 mm. The mean thickness of the wall of the CBD was 1 .0 mm, with a maximal thickness of 1 .5 mm. Enhancement of the wall of the CBD was equal to the enhancement of adjacent pancreatic parenchyma or could not be distinguished from the pancreatic parenchyma (grade 0) in 51 patients, slightly greater than that of pancreatic parenchyma (grade 1) in 44, and markedly greater than that of pancreatic parenchyma (grade 2) in five.
to
wall from the surrounding pancreas, grade 1 slightly greater than that of pancreatic paren-
chyma, and grade 2 indicated enhancement markedly greater than that of pancreatic parenchyma. Grading of the CHD wall was not attempted because of the lack of adjacent reference tissue. Nine to 1 2 months
Patients
Patients
were reviewed,
Enhancement contrast material enhancement
in nine and lost to further follow-up
the last clinical evaluation, of biliary disease in these
routinely.
and the extrahepatic bile ducts were assessed at two levels: the CHD at the level of the porta hepatis and the CBD at the level of the pancreatic head. These ducts were identified as tubular structures (lumen density equivalent to water) detected on serial scans in the expected location of the bile ducts: the CHD was anterior to the portal vein, lateral to the hepatic artery, and surrounded by fat in the porta hepatis, and the CBD was within or adjacent to the parenchyma of the pancreatic head. At each level, the presence or absence of the duct was assessed; if visualized, the images were magnified two or three times, and, by using digital calipers, the lumen diameter (inner to inner wall) and wall thickness were measured. Measurements were taken along the short axis of the duct lumen and wall to avoid falsely elevated measurements related to oblique scanning [2]. The standard and magnified images, including those with the cursor measurements, were filmed and reviewed later by three of the investigators with consensus agreement on each measurement.
parenchyma
evaluation
AJR:154, January 1990
conventional abdominal scanning techniques had been used. The thickness of the bile duct wall was measured by using a magnifying glass and custom-made rulers capable of measuring 0.5-mm incre-
and Methods
Biliary
ET AL.
37
women 20-89 years old (mean, 55). Of these 100 patients, 6-month clinical follow-up or autopsy evaluation was available for 86. Maximal follow-up for the remaining study patients was 3-6 months in 10, 13 months in three, and 2 weeks in one. Reasons for the shorter
follow-up courses were death from a nonbiliary cause without autopsy
Patients
with Billary
or Pancreatic
Disease
Bile duct walls 2 mm thick or greater were present in 52 patients with proved diagnoses. There were seven underlying diseases, which could be divided into eight categories. Nonoperated pancreatic carcinoma.-The walls of the bile ducts were thickened in eight patients with pancreatic carcinoma who had not undergone surgical bypass. In all cases, the thickening was focal, occurring within or just above the pancreatic head (Fig. 3). The wall thickening was concentric in all cases except one, in which it was minimally eccentric. The walls ranged from 2.5 to 3 mm thick. Wall enhancement was grade 1 in all cases. Surgically treated pancreatic carcinoma. -The bile duct walls were thickened in two patients with pancreatic carci-
Downloaded from www.ajronline.org by 190.129.29.192 on 10/12/15 from IP address 190.129.29.192. Copyright ARRS. For personal use only; all rights reserved
AJA:154, January 1990
CT
OF
EXTRAHEPATIC
BILE
DUCTS
81
Fig. 1.-CT scan shows normal common hepatic duct at level of porta hepatis. Lumen (density equivalent to water) is surrounded by a thin but perceptible duct wall (arrows).
Fig. 2.-CT scan shows normal common bile duct within pancreatic parenchyma. Lumen (density equivalent to water) is surrounded by thin, mildly enhancing duct wall (arrow).
Fig 3.-CT scan shows pancreatic carcinoma enlarging head of pancreas and surrounding distal common bile duct. There is mild eccentric thickening and contrast enhancement of duct wall (arrows),
which
was focal
and identified
within
pan-
creatic head only.
Fig. 4.-Marked eccentric thickening of common hepatic duct wall due to cholangiocarcinoma. A, CT scan shows eccentric thickening of common hepatic duct wall posteromedially (arrows), displacing duct lumen (density equivalent to water) from portal vein posterioriy. B, Cholangiogram shows eccentric narrowing of common hepatic duct lumen (arrows), with
obliteration just distal to narrowing. (Modified and reprinted with permission from Teefey et al. [5].)
noma who previously had undergone Whipple procedures with choledochojejunostomies. In one case, the thickening was diffuse, involving the remaining extrahepatic bile ducts. In the other case, the thickening was focal, occurring near the
anastomosis.
In both
cases,
the
thickening
was
concen-
tric. Walls were 3.5 and 4 mm thick; wall enhancement was assigned grades 1 and 2. Neoplasms of bile ducts. -This group comprised 10 patients with cholangiocarcinoma and one with a benign papillary neoplasm. Focal thickening of the duct wall at the level of the neoplastic mass was present in all cases (Figs. 4 and 5). Eccentric thickening occurred in nine and concentric thicken-
ing in two
Fig. 5.-CT ening
and
mild
scan shows subtle eccentric thickcontrast
common hepatic duct cholangiocarcinoma. this level only.
of the cases.
The
range
enhancement of wall of (arrows) In a patient with Thickening was present at
of duct
wall
thickening
was
3-1 1 mm. Wall enhancement was grade 0 in three, grade 1 in seven, and grade 2 in one. Acute cholangitis.-Bile duct walls were thickened in six patients with proved acute cholangitis. In two cases CBD stones were known to be present. The thickening was diffuse in all cases but one, in which focal thickening was seen at the level of a CBD stone. The thickening was concentric in all cases and was from 2 to 4 mm thick. Duct wall enhancement was grade 1 in four and grade 2 in two cases (Fig. 6). Oriental cholangiohepatitis.-Duct walls were thickened in three patients with oriental cholangiohepatitis. The thickening
82
SCHULTE
ET AL.
AJR:154, January 1990
Fig. 6.-CT
scan
shows
concentric
thickening
Downloaded from www.ajronline.org by 190.129.29.192 on 10/12/15 from IP address 190.129.29.192. Copyright ARRS. For personal use only; all rights reserved
and mild contrast enhancement of wall of common bile duct (arrows) in a patient with acute cholan. gitis. This thickening and enhancement was present throughout extrahepatic bile ducts.
Fig.
7.-CT
scan
shows
eccentric
thickening
and mild contrast enhancement of wall of common hepatic duct (arrows), which was present at multiple levels, in a patient with oriental cholangiohepatitis. Duct lumen is dilated markedly and contains stones and debris.
A
B
Fig. 8.-Focal concentric thickening and marked contrast enhancement of distal common bile duct wall in patient with choledocholithiasis. A, CT scan of intrapancreatic common bile duct, well above level of stone, reveals no discemible duct wall (arrows). B, Scan 2 cm inferior to A shows marked concentric thickening duct wall (arrows).
and
contrast enhancement
of
C, Delayed, small-field-of-view scan with 3-mm collimation, obtained 1 cm below B, shows a decrease In contrast enhancement of thickened duct wall (arrows), as well as a soft-tissue density stone within duct lumen. D, Cholangiogram
shows
stricture
of distal com-
mon bile duct (arrows) just proximal to visualized stone. Wall thickening seen on CT corresponds to region of stricture demonstrated here.
was diffuse and eccentric in all cases (Fig. 7). Thickening was 3 mm in one and 4 mm in two. All had grade 1 duct wall enhancement. CBD stones without cholangitis. -This group comprised nine patients with CBD stones without associated cholangitis or oriental cholangiohepatitis. The duct wall thickening was
focal in eight cases and diffuse in one. In six patients the CBD stones were visualized on CT; in these cases focal thickening was identified only at the level of the stones, which were in the distal CBD (Fig. 8). In the other two cases of focal thickening, the thickening was also in the distal CBD; however, the stones were not visualized definitely on CT. The
AJA:154,
January
CT
1990
OF
EXTRAHEPATIC
BILE
DUCTS
83
Downloaded from www.ajronline.org by 190.129.29.192 on 10/12/15 from IP address 190.129.29.192. Copyright ARRS. For personal use only; all rights reserved
Fig. 9.-CT scan shows marked contrast enhancement of eccentrically thickened distal wall of common bile duct (arrows) in a patient with sclerosing cholangitis.
Fig. 10.-CT scan shows marked contrast enhancement and concentric thickening of distal wall of common bile duct (arrow) in a patient with chronic pancreatitis, as evidenced by a markedly dilated pancreatic duct. Thickening and enhancement
were only present distally.
thickening was concentric in seven and eccentric in two. The wall thickening ranged from 2 to 4 mm. Wall enhancement was grade 0 in one, grade 1 in seven, and grade 2 in one. Sclerosing cholangitis. -Six patients were identified with sclerosing cholangitis and thickening of the duct wall. The thickening was diffuse in three and focal in three. Eccentric thickening was present in all but one case and ranged from 2.5 to 5 mm thick. Wall enhancement was grade 0 in three, grade 1 in one, and grade 2 in two (Fig. 9). Pancreatitis. -Seven patients with pancreatitis had thickening of the duct wall. In all cases, the thickening was focal and concentric, occurring in the distal CBD within or just above the inflamed pancreatic head (Fig. 10). Thickening ranged from 2 to 3 mm. Wall enhancement was grade 0 in one, grade 1 in five, and grade 2 in one.
Analysis
of the Pattern
of Duct
Wall Thickening
From these data it is evident that there are four general patterns of duct wall thickening and that these patterns are associated with certain diseases. Focal and concentric thickening in the distal common bile duct occurred only in patients with pancreatitis (7/7), pancreatic carcinoma (7/8), CBD stones (6/9), and acute cholangitis (1/6). Focal and eccentric thickening occurred in patients with neoplasms of the bile duct wall (9/1 1), sclerosing cholangitis (2/6), CBD stones (2/9), and pancreatic cancer (1/8). Only neoplasms of the bile duct wall were associated with thickening greater than 5 mm. Diffuse and concentric thickening was noted in patients with acute cholangitis (5/6), surgically treated pancreatic carcinoma (1/2), and CBD stones (1/9). Diffuse and eccentric thickening was detected in cases of oriental cholangiohepatitis (3/3) and sclerosing cholangitis (3/6).
Discussion Current CT technology provides sufficient anatomic resolution to visualize the extrahepatic bile ducts in the majority of patients. Using a bolus, rapid-infusion technique for admin-
istration of IV contrast material and dynamic scanning, we were able to visualize the CHD and CBD lumens in 66% and 82%, respectively, of 1 00 patients without biliary disease. These values are higher than previously reported [1 , 2] and reflect the continued improvement in CT performance and resolution. The CHD and CBD walls could be distinguished in 59% and 52% of these patients, respectively, and by using magnified
images
and digital
calipers,
the wall thickness
could
be measured easily and accurately in most cases. In these 1 00 patients without biliary tract disease, the mean thickness of the duct wall was 1 mm; no duct walls were more than 1.5 mm thick. Recent reports have described several cases of bile duct wall thickening and enhancement with IV contrast material, evident on CT. Five cases of thickening were found in AIDS cholangitis, although CT was noted to be less successful than sonography in showing this finding [3]. One case of thickening and one case of duct wall enhancement in AIDS cholangitis were noted in another report [4]. Seven cases of duct wall thickening have been described in sclerosing cholangitis [5, 6]. In one of these studies [5], 1 1 cases of duct wall enhancement in sclerosing cholangitis and two cases of duct wall thickening in cholangiocarcinoma were also reported. Five cases of contrast enhancement of the duct wall in oriental cholangiohepatitis have also been reported [7]. The degree of thickening or contrast enhancement of the bile duct wall was not quantified precisely in these reports, nor were the specific criteria for abnormal duct wall thickening or enhancement indicated. Our study provides an objective standard for normal thickness of the bile duct wall on CT and establishes a value of 1 .5 mm, above which the thickness of the duct wall should be considered abnormal. This study also establishes the enhancement charactenstics of the wall of the normal CBD. In the 1 00 patients without biliary tract disease, CBD wall enhancement after administration of IV contrast material varied. CBD wall enhancement was similar to that of the pancreatic parenchyma (grade 0) in 51 %, slightly greater than that of the pancreatic parenchyma (grade 1) in 44%, and markedly greater than that of the pancreatic parenchyma (grade 2) in 5%. In the patients with
Downloaded from www.ajronline.org by 190.129.29.192 on 10/12/15 from IP address 190.129.29.192. Copyright ARRS. For personal use only; all rights reserved
84
SCHULTE
biliary or pancreatic disease and duct wall thickening, wall enhancement occurred with greater frequency than in the normal group: grade 0, 1 5%; grade 1 , 70%; and grade 2, 1 5%. This may in part be biased by selection of grades 1 and 2 enhancing walls during collection of the abnormal cases, as thickened walls that enhance greater than the pancreas are visualized more readily. On the other hand, it is also possible that the enhancement reflects the pathologic process. Regardless, both grades 1 and 2 enhancement occurred in patients without any evidence of biliary or pancreatic disease, and therefore reliance on enhancement alone is an unreliable predictor
of biliary
disease.
The abnormal bile duct component of this study had several limitations. First, because of the retrospective design, a vanety of CT scanners and techniques had been used. The limited quality of many of the scans and the lack of standardization did limit evaluation of the CT scans; however, it did not falsely produce duct wall thickening or significantly alter the pattern of thickening. Second, because of the method of collecting abnormal cases, we were unable to determine the frequency of duct wall thickening within each disease category. Despite these limitations, our study does produce useful information about the abnormal duct wall. Our review of CT examinations from patients with any potential biliary or pancreatic disease revealed several cases of thickening of the duct wall in each of the following categories: pancreatitis, pancreatic carcinoma, CBD stones, bile duct wall neoplasms, acute cholangitis, oriental cholangiohepatitis, and sclerosing cholangitis. Within these select groups of patients with abnormal thickening of the duct wall, we found general trends in the pattern of thickening, and certain patterns of thickening were associated with specific diseases. Enhancement of the duct wall occurred throughout the abnormal groups, with no characteristic trends in the degree of enhancement. The pattern offocal, concentric thickening in the distal CBD was the characteristic pattern of thickening in patients with pancreatitis, pancreatic cancer, and CBD stones. All patients with pancreatitis and duct wall thickening had this pattern of thickening (Fig. 1 0). The cause of this thickening with pancreatitis is speculative, as little histopathologic information is available about the bile duct wall. It is known that pancreatitis produces peniductal inflammation and fibrosis of the intrapancreatic portion of the CBD [9, 1 0], and this could extend into and thicken the adjacent duct wall. Pancreatitis is also known to result in strictures of the distal CBD [1 0-1 3], and partial obstruction may contribute to the wall thickening. This pattern of concentric thickening in the distal CBD was present also in all but one patient with pancreatic carcinoma and duct wall thickening; in the one exception, the thickening was in a similar location but slightly eccentric (Fig. 3). This thickening could be the result of associated pancreatitis, or could be related to partial obstruction of the CBD. Other possible causes include neoplastic infiltration of the duct wall and lymphedema. The pattern of concentric thickening in the distal CBD was present in six of nine patients with CBD stones (Fig. 8). In two, the thickening was focal in the distal CBD, but eccentric; in one patient, the thickening was diffuse and concentric.
ET AL.
AJR:154,
January 1990
Again, the cause of the thickening is unknown, as no histopathologic studies are available. Mechanical irritation from the stones could result in inflammatory and fibrotic changes in the duct wall, and in all six cases in which stones were visualized on CT, the thickening was at the level of the stones. Partial obstruction and associated pancreatitis may also be contributory; in one patient with CBD stones, pancreatitis was clinically evident. In the one case in which the thickening was diffuse, multiple stones and debris were present in the CBD, which may account for the diffuse nature of the thickening. Although cholangitis was not evident clinically, it is also possible that subclinical cholangitis was present. Focal and eccentric thickening of the duct wall was the predominant pattern with neoplasms of the bile duct wall only; within this category the thickening was focal in all cases and eccentric in nine of 1 1 . The greatest degree of wall thickening, up to 1 1 mm, occurred in this group, and thickening greater than 5 mm was found in this group only. The thickening was at the level of the neoplasm in all cases. The neoplastic mass may obliterate the lumen (density equivalent to water), and appear as a solid mass on CT; however, thin collimated sections just above this level may reveal thickening of the duct wall, presumably the result of tumor infiltration (Fig. 4). Focal and eccentric thickening of the duct wall also occurred in patients with sclerosing cholangitis, CBD stones, and pancreatic cancer; however, this pattern was present in a minority of cases, and the thickening was 5 mm or less. Diffuse and concentric thickening of the duct wall was characteristic of acute cholangitis (five of six cases); this was the only category in which this pattern predominated (Fig. 6). This pattern correlates with the expected pathologic changes, as acute cholangitis is usually a diffuse process, known to result in an inflamed, thickened duct wall [14]. In contrast to acute cholangitis, the wall thickening in patients with oriental cholangiohepatitis was diffuse but eccentric in all cases (Fig. 7). This reflects the recurrent chronic nature of the process, with areas undergoing various degrees of inflammation and fibrosis [1 4, 1 5]. Sclenosing cholangitis was also associated with a diffuse and eccentric pattern of duct wall thickening in three of six cases. Differentiation was possible because of the marked ductal dilatation, intraluminal debris, and stones present in all three cases of oriental cholangiohepatitis. Sclerosing cholangitis was associated with both diffuse and focal thickening of the duct wall that was nearly always eccentric (Fig. 9). The cholangiographic findings of bile duct wall irregularities, diverticula, and nodules are well known [1 6], and these changes may contribute to the thickening of the duct wall evident on CT [5]. Histopathologic examination reveals nonspecific, chronic inflammatory change and fibrosis within the walls of the bile duct [14, 17, 18]. In summary, the extrahepatic bile ducts can be visualized and characterized in a majority of patients. Our analysis of patients without biliary or pancreatic disease showed that no duct walls were more than 1 .5 mm thick; this establishes a value above which the duct wall should be considered abnormal. Enhancement of the bile duct wall was found to occur in patients without biliary tract disease; this establishes that
CT
Downloaded from www.ajronline.org by 190.129.29.192 on 10/12/15 from IP address 190.129.29.192. Copyright ARRS. For personal use only; all rights reserved
AJR:154, January 1990
OF
EXTRAHEPATIC
enhancement alone is not predictive of pathology. Several diseases were found to be associated with thickening of the duct wall, and a differential diagnosis for thickening is established. Four general patterns of thickening were identified, and there appears to be an association between each pattern and specific diseases. Enhancement of the duct wall occurred throughout the disease categories, with no specific trends to the degree of enhancement.
REFERENCES 1 . Foley WD, Wilson CR, Quiroz extrahepatic biliary tract Tomogr 1980;4:48-52
with
FA, Lawson computed
TL. Demonstration of the normal tomography. J Comput Assist
2. Co CS, Shea WJ, Goldberg HI. Evaluation of common bile duct diameter using high resolution computed tomography. J Comput Assist Tomogr 1986;10:424-427 3. Dolmatch BL, Laing FC, Federle MP, Jeffrey RB, Cello J. AIDS-related cholangitis: radiographic findings in nine patients. Radiology 1987;163: 31 3-316 4. Radin DR, Cohen H, Halls JM. Acalculous inflammatory disease of the biliary tree in acquired immunodeficiency syndrome: CT demonstration. J Comput Assist Tomogr 1987;1 1:775-778 5. Teefey SA, Baron RL, Rohrmann CA, Shuman WP, Freeny PC. Sclerosing cholangitis: CT findings. Radiology 1988;169:635-639
BILE
DUCTS
6. Carroll BA, Oppenheimer stration
85
DA. Sclerosing cholangitis:
of bile duct wall thickening.
AJR 1982;139:
sonographic
demon-
1016-1018
7. Chan F-L, Man S-W, Leong LLY, Fan S-T. Evaluation ofrecurrent pyogenic cholangitis with CT: analysis of 50 patients. Radiology 1989;170: 165-1 69 8. Baron AL. Common bile duct stones: reassessment of criteria for CT diagnosis. Radiology 1987;162:419-424 9. Sarles H, Sarles J-C, Camatte A, et al. Observations on 205 confirmed cases of acute pancreatitis, recurring pancreatitis, and chronic pancreatitis. Gut 1965;6:545-549 10. Littenberg G, Afroudakis A, Kaplowitz N. Common bile duct stenosis from chronic pancreatitis: a clinical and pathologic spectrum. Medicine (Baltimore) 1979;58:385-412 1 1 . Scott J, Summerfield JA, Elias E, Dick A, Sherlock S. Chronic pancreatitis: a cause of cholestasis. Gut 1977;18:196-201 12. Rohrmann CA Jr, Baron AL. Biliary complications of pancreatitis. Radiol Olin North Am 1989;27:93-104 13. Sarles H, Sahel J. Cholestasis and lesions of the biliary tract in chronic pancreatitis. Gut 1978;19:851-857 14. Schiff L, Schiff ER, eds. Diseases of the liver. Philadelphia: Lippincott, 1987 15. Chou ST. Chan CW. Recurrent pyogenic cholangitis: a necropsy study. Pathology
1980;12:415-428
16. MacCarty AL, LaRusso NF, Wiesner RH, Ludwig J. Primary sclerosing cholangitis: findings on cholangiography and pancreatography. Radiology 1983;149:39-44
17. LaRusso NF, Wiesner RH, Ludwig J, MacCarty AL Primary scierosing cholangitis. N EnglJ Med 1984;310:899-903 18. Thorpe MEC, Scheuer PJ, Sherlock S. Primary sclerosing cholangitis, the biliary tree, and ulcerative colitis. Gut 1967;8:435-448
