SEARCH

SEARCH BY CITATION

Keywords:

  • biliary papillomatosis (BP);
  • mucin-hypersecreting biliary papillomatosis (MBP);
  • clinical manifestations;
  • cholangioscopy;
  • survivors

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

BACKGROUND

Biliary papillomatosis (BP) is a rare disease that is characterized by multiple numerous papillary adenomas in the biliary tree. The clinical features and outcome, however, are not well known. The authors retrospectively analyzed their clinicopathologic features and long-term follow-up results.

METHODS

Between March 1995 and January 2003, 58 patients were diagnosed with BP by cholangioscopic and histologic findings at a tertiary referral center, Asan Medical Center (University of Ulsan College of Medicine, Seoul, Korea). The authors retrospectively reviewed the medical records to obtain demographic, radiologic, cholangioscopic, and pathologic data.

RESULTS

The common clinical manifestations at the presentation of patients were repeated episodes of abdominal pain, jaundice, and acute cholangitis. Acute cholangitis was more common in patients with mucin-hypersecreting BP (MBP), whereas patients with nonmucin-producing BP (NMBP) were more asymptomatic (P < 0.05). Papillary adenocarcinoma and mucinous carcinoma were detected in 48 patients (83%) with papillary adenomas. Overall survival rates of NMBP and MBP were 89% and 69% at 1 year, 57% and 37% at 3 years, and 52% and 19% at 5 years, respectively. The mean survival period of NMBP and MBP was 52.27 ± 6.72 months and 30.84 ± 8.36 months, respectively.

CONCLUSIONS

BP should be regarded as a premalignant disease with high malignant potential. The pathogenesis of progression from benign to malignant disease may follow the adenomacarcinoma sequence. Although clinical presentations were somewhat different for patients with NMBP and MBP, the long-term survival rate was similar. Cancer 2004;100:783–93. © 2004 American Cancer Society.

Biliary papillomatosis (BP) is a disease characterized by multiple papillary tumors of variable distribution and extent in the intrahepatic and/or extrahepatic biliary tree. Although BP is basically a collection of benign papillary adenomas, papillary adenocarcinoma can develop within these lesions1, 2 and has a tendency to spread superficially along the bile duct mucosa.3, 4

Because BP is a scarce entity, the clinical features and outcome are not well known. To our knowledge, few reports to date have described the clinical characteristics and prognosis of this biliary papillary neoplasm.5–8 These reports comprise a small sample of patients but do not include long-term follow-up results of patients who did not undergo curative surgery. BP is classified as either mucin-hypersecreting type (MBP) or nonmucin-producing type (NMBP) because both disease types have similar macroscopic and microscopic findings, except for mucin hypersecretion. In the current study, we classified patients as having NMBP or MBP and compared their clinicopathologic features, as well as the long-term follow-up results of patients who received either curative or palliative treatment. To our knowledge, this is the first investigation of its kind and is currently the largest series in the English literature on BP.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Data Collection and Follow-Up

Between March 1995 and January 2003, 58 patients were diagnosed with BP based on cholangioscopic and histologic findings. We retrospectively reviewed the medical records to obtain demographic, radiologic, cholangioscopic, and pathologic data.

For patients who had not visited the clinic between January 2002 and January 2003, we obtained their follow-up information either by sending questionnaires or by telephone contact. Only three patients were lost during follow-up.

Diagnosis of Biliary Papillomatosis

The diagnosis of BP was based on cholangioscopic or macroscopic evidence of numerous papillary growing tumors in the intrahepatic duct lumen with or without extrahepatic duct involvement and on microscopic evidence of branching papillary fronds covered by columnar epithelium and supported by delicate fibrovascular stalks.5, 9

Cholangioscopy revealed a large amount of secreted mucin in the bile duct lumen (mucobilia) in some patients. Overproduced mucin resulted in massive mucin retention and bile stasis within the intrahepatic (IHD) and/or the extrahepatic bile duct (EHD) lumen. This type of tumor was defined as MBP.

Study Design

We classified BP as either MBP or NMBP according to the presence or absence of mucobilia on cholangioscopic evaluation. We compared the clinical, radiologic, and pathologic features and the survival rates between the two groups of patients.

Analyses of Radiologic Findings

Radiologic findings were reviewed by two experienced gastrointestinal radiologists. Interpretations of the abdominal ultrasonography (US), computed tomography scan (CT), direct cholangiogram, and magnetic resonance cholangiopancreatography (MRCP) findings were made by the consensus of the two gastrointestinal radiologists.

Method of Cholangioscopic Evaluation

All patients underwent percutaneous transhepatic cholangioscopic evaluation (PTCS). For PTCS, initial percutaneous transhepatic biliary drainage (PTBD) was performed by using a pigtail catheter (7.5-–8.0-French; Cook, Bloomington, IN) under fluoroscopic guidance. Before PTBD, 50 mg of meperidine and 3–5 mg of midazolam were administered intravenously to control pain and anxiety. Two or 3 days after PTBD, the sinus tract was dilated to receive a 16- or18-French catheter after intercostal nerve block with 2% lidocaine. PTCS was then performed 7 days after the tract dilatation. This interval was required for the stabilization and maturation of the sinus tract. Cholangioscopic evaluation was performed with a cholangioscope with either a 4.9-mm external dimension (FCN-15X; Pentax, Tokyo, Japan) or a 5.1-mm external dimension (FCN-1530; Pentax). During cholangioscopic evaluation, the various mucosal appearances of bile duct tumors and strictures were studied. Multiple targeted biopsies were performed with forceps (KA 1811S, Pentax) under direct cholangioscopic visualization.

Pathologic Review

Three pathologists (S.J.J., E.Y., and D.E.S.) reviewed the tissue specimens of 58 patients with BP (34 cholangioscopic biopsy specimens and 24 surgically resected specimens).

Statistical Analyses

All statistical analyses were performed with SPSS statistical software, version 11.0 (SPSS Inc., Chicago, IL). Categoric continuous variables were compared with the chi-square test, the Fisher exact test, and the Student t test. The log-rank test was used to compare Kaplan–Meier curves of survival rates. P values < 0.05 were statistically significant.

The institutional review board of the University of Ulsan College of Medicine (Seoul, Korea) approved the study. Patients were contacted by either telephone or mail.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Demographic Data and Clinical Characteristics

Thirty nine patients were men and 19 were women. Their mean age was 60.3 years (range, 37–87 years). Of the 58 patients, 40 (69%) were diagnosed with NMBP and 18 (31%) with MBP. The demographic data and clinical characteristics of patients with NMBP and MBP are summarized in Table 1. Age and gender distributions were not significantly different between the two groups.

Table 1. Demographic Data and Clinical Characteristics of the Patients
CharacteristicsNMBP (n = 40)MBP (n = 18)
  • NMBP: nonmucin-producing biliary papillomatosis; MBP: mucin-hypersecreting biliary papillomatosis.

  • a

    P < 0.05.

Demographic data  
 Mean age (yrs) (range)60.0 (37–79)61.2 (39–87)
 Gender (male:female)30:109:9
Clinical manifestations  
 Abdominal pain14 (35.0%)8 (44.4%)
 Jaundice 8 (20%)2 (11.1%)
 Acute cholangitisa 5 (12.5%)8 (44.4%)
 Weight loss 2 (5.0%)0
 No symptomsa11 (27.5%)0
Associated diseases  
 Biliary stone12 (30%)9 (50%)
 Clonorchiasis 9 (22.5%)1 (5.6%)
 Biliary stone and clonorchiasis 1 (2.5%)0
 None18 (45.0%)8 (44.4%)

The principal clinical manifestations at admission were repeated episodes of right upper quadrant or epigastric pain, jaundice, acute cholangitis, and weight loss. It is noteworthy that acute cholangitis was a more common manifestation in patients with MBP, whereas asymptomatic patients were diagnosed more frequently with NMBP (P < 0.05). Asymptomatic patients with NMBP initially were suspected by abdominal US or CT scan performed during a routine health checkup.

Associated biliary stones were observed in 21 patients (36%). All these stones were determined to be pigment stones by macroscopic and cross-sectional appearance. Eleven patients (19%) had been treated previously for clonorchiasis or were diagnosed with clonorchiasis during the diagnostic workup. One patient with NMBP was diagnosed with both biliary stones and clonorchiasis. Patients with MBP and NMBP showed no difference regarding the number of associated diseases (Table 1).

Radiologic Findings

Abdominal US were performed in 34 patients and revealed hyperechoic or hypoechoic masses within dilated IHD in 14 patients (41.2%). On CT scans, 29 (50%) of 58 patients revealed soft tissue densities (hypoattenuated or isoattenuated mass) in the dilated bile ducts. In some patients, the bile duct walls were thickened and enhanced. In all patients with MBP, marked dilatation of IHD and EHD was detected both proximal and distal to the intraductal tumors. However, the detection of mucobilia itself was difficult using US and CT scans.

Thirty-eight of 58 patients underwent MRCP, which showed multiple round, oval papillary signal voids within the lumen or attached at the wall of the bile ducts. Intraductal papillary tumors were detected in 26 (68%) of 38 patients. In 6 patients (16%), only bile duct wall thickening and dilatation of the bile duct with or without stricture were detected.

On direct cholangiographic findings obtained by endoscopic transpapillary or by the percutaneous transhepatic route, the dilated ducts revealed multiple, round-to-ovoid filling defects and ragged irregularity of the bile duct wall (Fig. 1A). In the patients with MBP, a cholangiogram revealed extremely dilated IHD and EHD with amorphous filling defects (Fig. 1B). During endoscopic retrograde cholangiopancreatography (ERCP), a wide open ampulla of Vater with extrusion of mucoid material was observed on duodenoscopic evaluation in the majority of patients with MBP (Fig. 2A).

thumbnail image

Figure 1. Direct cholangiogram of biliary papillomatosis. (A) Nonmucin-producing biliary papillomatosis. The dilated main hepatic ducts and common bile duct (CBD) reveal multiple, round to ovoid filling defects (black arrows) and ragged irregularity (arrowheads) of the bile duct wall. Round filling defects (white arrow) at the distal CBD are air bubbles. (B) Mucin-hypersecreting biliary papillomatosis. Extremely dilated extrahepatic bile duct with amorphous filling defects (arrowheads). Left intrahepatic bile duct is not visualized by mucin (black arrows).

Download figure to PowerPoint

thumbnail image

Figure 2. Duodenoscopic finding of mucin-hypersecreting biliary (MBP) papillomatosis and nonmucin-producing biliary papillomatosis (NMBP). (A) Patulous ampulla extruding bile- tinged mucin (*) is observed in patients with MBP. (B) NMBP. After sphincterotomy, a balloon catheter was inserted, inflated, and retracted carefully. This resulted in extrusion of friable polypoid masses with a smooth, convoluted surface. The specimen was consistent with papillary adenocarcinoma.

Download figure to PowerPoint

Cholangioscopic Findings

On cholangioscopic evaluation, sludge material intermingled with pus frequently covered the papillary masses and multiple papillary masses were observed within the bile duct lumen (Fig. 3A). The masses were soft and friable and the surface color was usually bright yellow or pinkish. Some tissue specimens showed normal mucosa intervening between papillary lesions. The heights of the papillary projections varied. Prominent papillary tumors were found to correlate histologically with papillary adenocarcinoma, whereas tiny villous (Fig. 3B) or coral reef-like projections (Fig. 3C) or fish egg-like mucosal changes (Fig. 3D) were found to correlate with papillary adenoma. The cholangioscopic appearance of MBP was very similar to that of NMBP with the exception of the intraductal amorphous mucin, which had a whitish or bile-tinged color. Various papillary mucosal lesions were covered with mucin and detected on the wall of the bile duct after saline irrigation. The viscosity of mucin varied.

thumbnail image

Figure 3. Various cholangioscopic findings of biliary papillomatosis. (A) Pinkish or red multiple papillary masses and tiny papillary projections are scattered within the bile duct lumen. Normal intervening mucosa is observed between papillary masses. (B) A number of tiny villous projections are observed in the bile duct lumen. (C) Coral reef-like mucosal projections are observed in the bile duct lumen. (D) Fish egg-like mucosal changes are observed in the bile duct lumen.

Download figure to PowerPoint

Location of the Tumors

Seven tumors were located in both lobes, 16 in the right IHD, 14 in the left IHD, 12 in both the lobes and the EHD, 5 in the right IHD and EHD, and 4 in the left IHD and EHD. The location of the NMBP and MBP is summarized in Table 2. There was no significant difference with regard to the overall tumor distribution in the liver between the two groups. However, involvement of the EHD was more common among patients with NMBP (P < 0.05).

Table 2. Location of the Tumors
CharacteristicsNMBP (n = 40)MBP (n = 18)Total
  1. NMBP: nonmucin-producing biliary papillomatosis; MBP: mucin-hypersecreting biliary papillomatosis; RIHD: right intrahepatic duct; LIHD: left intrahepatic duct; EHD: extrahepatic duct.

Both lobes347
RIHD11516
LIHD8614
Both lobes and EHD12012
RIHD and EHD325
LIHD and EHD314

Histopathologic Findings

Macroscopic findings

The 24 patients who underwent surgical resection demonstrated grayish tan to yellow, friable, papillary granular masses that filled the dilated IHD (Fig. 4A) and/or EHD (Fig. 4B). There was no macroscopic ot morphologic difference noted between tissue specimens of NMBP and MBP, except for mucin in the ductal lumen.

thumbnail image

Figure 4. Macroscopic findings of biliary papillomatosis (BP). (A) Macroscopic finding of intrahepatic BP. Intrahepatic ducts are dilated and filled with yellowish-tan friable papillary masses(arrows). (B) Macroscopic finding of extrahepatic BP. The mucosa is covered by numerous yellowish-tan friable papillary masses.

Download figure to PowerPoint

Microscopic findings

In the current study, BP was classified into five classes according to the degree of cytologic and structural atypia including increased nuclear-to-cytoplasmic ratio, loss of polarity, hyperchromatism, pleomorphism, prominent nucleoli, abnormal mitosis, cribriform pattern and multilayering, and presence of invasion.10 Class 1 was defined as BP with low-grade dysplasia showing mild nuclear atypia, focal multilayering, and no invasion. Class 2 was defined as BP with high-grade dysplasia showing moderate nuclear atypia, cribriform pattern, and multilayering. Class 3 was defined as BP with carcinoma in situ (CIS), which is characterized by severe nuclear atypia with pleomorphism, atypical mitosis, and occasional necrosis but no stromal invasion. Class 4 was defined as CIS with microscopic foci of stromal invasion. Finally, Class 5 was defined as definite invasion into the hepatic parenchyma or a fibromuscular layer of the bile duct wall (Table 3). When a lesion showed mixed types of BP or a patient received multiple cholangioscopic biopsies at different sites, the most advanced type was used for final classification. Classes 1 and 2 were included in the category of benign adenoma and > Class 3 were considered as adenocarcinoma.

Table 3. Histologic Criteria of BP According to the Degree of Cytologic, Structural Atypia, and Liver Invasion
CharacteristicsClass 1Class 2Class 3Class 4Class 5
  1. BP: biliary papillomatosis; N/C ratio: nuclear-to-cytoplasmic ratio.

Nuclear atypia     
 Increase N/C ratioMildModerateSevereSevereSevere
 Loss of polarityMildModerateSevereSevereSevere
 HyperchromatismMildModerateSevereSevereSevere
 PleomorphismMildModerateSevereSevereSevere
 Prominent nucleoliInconspicuousConspicuousConspicuousConspicuousConspicuous
 Abnormal mitosisAbsentFewPresentPresentPresent
Structural atypia     
 Cribriform patternAbsentPresentPresentPresentPresent
 MultilayeringPresentPresentPresentPresentPresent
Invasion     
 StromaAbsentAbsentAbsentPresentPresent
 Liver parenchymaPresentAbsentAbsentAbsentPresent

The final classification of cholangioscopic biopsy specimens from 34 patients included Class 1 (n = 1), Class 2 (n = 7), and > Class 3 (n = 26). All patients with > Class 3 showed adenomas in the background. Twenty-four surgically resected tissue specimens were comprised of Class 1 (n = 1), Class 2 (n = 1), Class 3 (n = 7), Class 4 (n = 5), and Class 5 (n = 10). Similarly, adenomas coexisted in the adjacent region in the 22 patients who were diagnosed with cancer after surgical resection (Fig. 5). Results of the cholangioscopic biopsy and surgical resected specimens of NMBP and MBP are presented in Table 4. There was no significant difference with regard to histologic grading between the two groups (P = 0.156). Of the 58 tissue specimens, 42 (72%) were compatible with papillary adenocarcinoma (37 well differentiated tumors and 5 moderately differentiated tumors). Mucinous carcinoma was found in 3 of 34 patients (1 patient with NMBP and 2 patients with MBP) diagnosed by cholangioscopic biopsy and in 3 of 24 patients (1 NMBP and 2 MBP) who received hepatic resection. Lymph node metastasis was not detected in any of the resected tissue specimens. Perineural invasion was detected in two patients (one with NMBP and one with MBP).

thumbnail image

Figure 5. Microscopic findings of biliary papillomatosis (BP). (A) Class 3 BP (carcinoma in situ [CIS]) shows fine fibrovascular cores lined by biliary epithelia with severe nuclear atypia, focal cribriform pattern, and no invasion. Adenoma coexists in the adjacent region. (B) Class 4 BP (CIS with microscopic stromal invasion) shows well differentiated CIS with microscopic stromal invasion (arrow head) (A and B: hematoxylin-eosin, original magnification × 10.)

Download figure to PowerPoint

Table 4. Results of Cholangioscopic Biopsies and Surgical Resected Specimens According to the Pathologic Types
ClassCholangioscopic biopsy (n = 34)Surgical resection (n = 24)
NMBPMBPNMBPMBP
  • NMBP: nonmucin-producing biliary papillomatosis; MBP: mucin-hypersecreting biliary papillomatosis.

  • a

    The invasiveness could not be determined based on biopsy specimens.

Class 11010
Class 26110
Class 316a10a52
Class 4  32
Class 5  82

Epithelial Nature

The lining epithelia of the BP tissue specimens demonstrated biliary epithelial type and various gastroenteric metaplasias such as goblet cell metaplasia, foveolar metaplasia, and colon-like metaplasia. Colon-like metaplasia was found in all tissue specimens. Goblet cells tended to predominate in MBP (P < 0.01). However, the frequency of foveolar metaplasia was similar between NMBP and MBP (Table 5).

Table 5. Differentiation of the Epithelial Nature between NMBP and MBP
CharacteristicsNMBP (n = 40)MBP (n = 18)
  • NMBP: nonmucin-producing biliary papillomatosis; MBP: mucin-hypersecreting biliary papillomatosis.

  • a

    P < 0.01.

Goblet cell metaplasiaa13 (32.5%)18 (100%)
Foveolar metaplasia11 (27.5%) 7 (38.8%)
Colonic metaplasia40 (100%)18 (100%)
Biliary type40 (100%)18 (100%)

Treatment

Twenty-four (41.4%) of 58 patients underwent partial hepatic resections (Table 6). Of the 24 patients, 20 underwent curative resection with an adequate resection margin and no lymph node metastasis or vascular invasion. In 9 of 20 patients who underwent curative resection, the tumors were confined to the mucosa of the resected specimens. In four patients who underwent palliative resection, the main lesion was resected but residual lesions remained in the unresected hepatic lobe. There was no mortality related to the operative procedure.

Table 6. Type of Liver Resection Performed in 24 Patients with Biliary Papillomatosis
CharacteristicsNMBP (n = 18)MBP (n = 6)
  1. NMBP: nonmucin-producing biliary papillomatosis; MBP: mucin-hypersecreting biliary papillomatosis.

Right lobectomy81
Left lobectomy74
Right posterior segmentectomy10
Left lateral segmentectomy11
Extended left lobectomy10

Thirty-four patients underwent PTBD only, with two patients receiving additional radiotherapy.

Survival and Disease Recurrence

The mean duration of follow-up was 24 months (range, 6–84 months). For patients who underwent surgery, the mean follow-up was 30 months (range, 6–73 months). For patients who underwent palliative drainage, the mean follow-up was 18 months (range, 3–84 months).

Actuarial Patient Survival

Kaplan–Meier survivals of patients with NMBP and MBP are shown in Figure 6. The survival rates of patients with NMBP and MBP were 89% and 69%, respectively, at 1 year, 57% and 37%, respectively, at 3 years, and 52% and 19%, respectively, at 5 years. The mean survival period of patients with NMBP and MBP was 52.27 ± 6.72 months (95% confidence interval [95% CI], 39.11–65.43 months) and 30.84 ± 8.36 months (95% CI, 14.45–47.23 months), respectively. There was no significant difference noted between the two groups (P = 0.1555).

thumbnail image

Figure 6. The Kaplan–Meier survival curve of overall patients with biliary papillomatosis. Straight line: nonmucin-producing biliary papillomatosis; dotted line: mucin-hypersecreting biliary papillomatosis. P = 0.156 (log-rank test).

Download figure to PowerPoint

Patient survival according to the type of treatment is shown Figure 7. The mean survival period of the curative resection and palliative groups was 60.87 ± 5.86 months (95% CI, 49.38–72.36 months) and 36.72 ± 6.61 months (95% CI, 23.77–49.67 months), respectively. There was a significant difference noted between the two groups (P = 0.0103).

thumbnail image

Figure 7. The Kaplan–Meier survival curve of patients with biliary papillomatosis who underwent curative resection and palliative treatment. Straight line: curative resection group; dotted line: palliative treatment group. P < 0.05 (log-rank test).

Download figure to PowerPoint

We also analyzed patient survival according to the histologic class (Classes 1 and 2 vs. ≥ Class 3). However, there was no significant difference between the two groups (P = 0.932). It was impossible to perform a subgroup analysis of patient survival according to the histologic class of patients who received resection because of small number of Class 1 or 2 patients.

Disease-Free Survival

The disease-free survival rates for patients who underwent curative resection were 100% at 1 year, 81% at 3 years, and 81% at 5 years.

In two patients who underwent curative surgical resection, disease recurrence developed despite an initial negative resection margin. One patient experienced disease recurrence at the surgical margin 17 months after undergoing a left lobectomy and died 26 months after resection. Another patient developed disease recurrence in the residual intrahepatic ducts 25 months after undergoing right lobectomy. This patient died 34 months after the surgical resection.

Four patients who had a positive resection margin were alive at the end of the study. Their follow-up period ranged from 9–72 months.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Because BP is a rare disease, to our knowledge reported series published to date contained only a few patients. We believe that the current study is the largest study published to date concerning BP. The Asan Medical Center (University of Ulsan College of Medicine, Seoul, Korea) is a tertiary referral center. Greater than 4000 PTCS have been performed during the past 7 years at this institution. We have carefully followed the course of patients with BP in the last decade and have presented the results in the current study.

The definition of BP puts emphasis on the macroscopic morphology and multiplicity of the neoplasm. MBP and NMBP show very similar macroscopic and microscopic findings. Therefore, we classified MBP as one subtype of BP. In the current study, the division was necessary because the clinicoradiologic presentation of BP may differ according to the presence/absence of mucobilia. The goal of the current study was to determine whether this difference translates into change in therapeutic outcome and prognosis.

The demographic data from the current study demonstrated that BP is more common in men at their sixth to seventh decade of life. This is not so different from usual cholangiocarcinoma. For NMBP, males were involved three times more than females, but in MBP, male and female were equally affected. However, to our knowledge there was no significant statistical difference with regard to gender ratios and age between two groups.

The most common clinical manifestations of patients with BP were recurrent abdominal pain, repeated episodes of acute cholangitis, and obstructive jaundice. Acute cholangitis, which is not a common presentation for typical cholangiocarcinoma, was the second most common manifestation of BP. This manifestation may be the result of tumor emboli, which being friable and easily detached from its origin (Fig. 2B), lead to acute obstruction of the bile duct such as gallstones. Patients with MBP had more frequent episodes of acute cholangitis than patients with NMBP, which may be due to thick mucin obstructing the bile duct. It is noteworthy that 11 of 40 patients with NMBP were asymptomatic and the disease was detected incidentally. However, there was no significant difference in the degree of tumor extent or resection rate between symptomatic and asymptomatic patients.

The level of CA 19-9 antigen was elevated in 20 (40%) of 50 patients. The mean level was much higher in patients with MBP compared with patients with NMBP (P < 0.05; data not shown). We believe that the significantly higher elevation of CA 19-9 antigen in patients with MBP may be due to more common cholestasis or cholangitis associated with mucin overproduction.

Although there have been several studies on BP published to date,5, 11 to our knowledge the pathogenesis of this disease has not been well characterized. Most authors suggest that bile stasis and recurrent infection induced by chronic biliary diseases such as hepatolithiasis or clonorchiasis may lead to the development of periductal inflammation, mucosal hyperplasia, dysplasia (low and high grade), CIS, and adenocarcinoma.12, 13 The investigations regarding BP are mainly of patients in Far Eastern countries such as Korea, Taiwan, and Japan.5, 10, 12–14 These countries are endemic areas of hepatolithiasis and clonorchiasis, which are risk factors for BP. In the current study, approximately half of the patients were diagnosed with biliary stones or clonorchiasis. These findings suggest that BP is associated strongly with these biliary diseases.

BP has been considered to be a benign disease with low malignant potential.15, 16 Some series have reported only anecdotal cases that progressed to malignancy.5–7, 17–20 However, in the current study, 48 (83%) of 58 patients were diagnosed with BP that revealed papillary carcinoma (42 patients with adenocarcinoma and 6 patients with mucinous carcinoma) in the background of adenoma on cholangioscopic biopsy or surgically resected specimens. Most of the patients in the current study were diagnosed with cancer after a thorough examination of the entire biliary tree by PTCS and cholangioscopic-targeted multiple biopsies for all of the lesions. Therefore, BP is not a benign disease with low malignant potential as previously believed. Rather, it is a premalignant disease with high malignant potential. Moreover, the findings of mixed pathologic types in each case as well as papillary adenocarcinoma in the background of papillary adenoma suggest that BP represents a continuous histologic spectrum of neoplastic biliary disease from dysplasia to carcinoma with papillary configuration.

The tissue specimens in the current study demonstrated lining epithelia of biliary type. Goblet cell metaplasia was found to be more predominant in patients with MBP. However, there was no significant difference noted in the frequency of colonic and foveolar metaplasia between the two groups, which is somewhat different from the results provided by Chen et al.10

US and CT scan detected intraductal masses in only 41.2% and 50% of patients with BP, respectively. The most common abnormal finding of US and CT scans in these patients was IHD and/or common bile duct dilatation. When intraductal masses were not detected on US or CT scan, they were diagnosed as biliary stones, clonorchiasis, and benign biliary strictures. ERCP may be useful in making the diagnosis of BP. The characteristic ERCP findings of BP are multiple small filling defects and the serrated irregularity of the bile duct wall. In patients with MBP, hypersecreted mucin is drained through the ampulla and a patulous ampulla is the characteristic duodenoscopic finding. On cholangiography, diffuse bile duct dilatation and amorphous filling defects in the bile duct are characteristic. However, a large amount of mucin secretion and obstruction by the tumor prevent complete opacification of the entire biliary tract. As a result, precise evaluation of ductal extent by ERCP is often suboptimal.5, 21

Cholangioscopic evaluation provided detailed information regarding the extent of disease, which enabled the appropriate surgical treatment to be provided to patients. PTCS evaluation has some advantages over conventional radiologic imaging. First, it may visualize the bile duct mucosa directly and detect small or subtle mucosal lesions that are not evident by direct cholangiogram. The recurrence rate of BP after surgery is reported to be high.19 We suggest that incomplete preoperative evaluation of the extent of BP might be an important contributing factor for the high disease recurrence rate. Because small papillomas may not be detected by conventional radiologic methods,5, 21, 22 these undetected lesions, usually remote from the main tumor, may be the foci of recurrence.23, 24 In 6 of the 21 patients in the current study in whom the disease was confined to the EHD by radiologic images including ERCP and MRCP, a Whipple procedure was planned initially. However, as cholangioscopy revealed additional IHD lesions, the surgical procedure was changed to a more aggressive one. Therefore, the patients underwent partial hepatectomy in addition to pancreaticoduodenectomy and all tumors were resected. Another advantage of cholangioscopy is to confirm the histology of the lesions shown by conventional radiologic imaging. We prefer PTCS to peroral cholangioscopy because peroral cholangioscopic examination may be inferior to PTCS for complete evaluation of the IHD due to the difficulty in using the remote control of the babyscope. Therefore, PTCS examination is an indispensable preoperative procedure for determining treatment modalities and the appropriate extent of resection in BP. It is more useful in patients with MBP because mucin is observed as filling defects on direct cholangiography.

MBP was reported previously as mucin-hypersecreting cholangiocarcinoma.25 Mucin-hypersecreting cholangiocarcinoma is the same disease entity as MBP. It is interesting to note that MBP has many similarities with intraductal papillary mucinous tumor (IPMT) of the pancreas.2, 12, 26–29 The fundamental similarity is the hypersecretion of mucin by an intraductal tumor. Overproduced mucin is relatively viscous and obstructs ducts. Hypersecreted mucin can be drained through the ampulla, and a patulous ampulla is a characteristic duodenoscopic finding in both entities. Using cholangioscopic examination, the majority of patients had a distinct papillary lesion that was accompanied by villous, coral reef-like mucosal projections or fish egg-like mucosal changes (Fig. 3). These cholangioscopic findings closely resembled pancreatoscopic findings associated with IPMT of the pancreas. Another similarity is the pathologic findings and adenoma-carcinoma sequence that is suggested in both disease entities.2, 26, 30 MBP causes frequent attacks of obstructive cholangitis by mucus plugging. Similarly, IPMT of the pancreas frequently induces obstructive pancreatitis.

In the current study, 24 (41%) of 58 patients underwent surgical resection. None of these patients had regional lymph node metastasis or vascular invasion and tumors were confined to the mucosa in nine patients. Twenty of 24 patients underwent curative surgical resection and the 5-year survival rate of these patients was 81%. After curative surgery, the prognosis of patients with BP was excellent in the current series and was similar to that reported in previous studies.5, 14, 31–35 Thirty-four of 58 patients underwent only palliative drainage. Their mean survival was 37 months, which is significantly longer than that of patients with usual cholangiocarcinoma.25, 36, 37 The reason is that BP is less invasive and grows slowly. In the current study, there was no significant survival difference noted based on the histologic class. Moreover, two of seven patients with a Class 2 histologic diagnosis as determined by cholangioscopic biopsy died of disease progression. Although these patients were classified only as having high-grade dysplasia, they showed clinicoradiologic features of malignancy on follow-up. One patient had portal vein encasement on magnetic resonance angiography. The other patient, who underwent only PTBD because of the involvement of both lobes by the tumor, showed disease progression on follow-up CT scans. This discrepancy between pathologic class by biopsy and clinical diagnosis may reflect the difficulty in excluding a carcinoma with a limited biopsy specimen. Therefore, when patients are diagnosed as Class 1 or Class 2 on the basis of a cholangioscopic biopsy, resection also should be considered because carcinoma cannot be excluded by cholangioscopic biopsy only.

BP should not be regarded as a benign disease with low malignant potential but, rather, as a premalignant lesion with high malignant potential. The pathogenesis of progression from benign to malignant disease may follow the adenomacarcinoma sequence. Although clinical presentations and gastroenteric metaplasia were somewhat different, the long-term survival rate was not found to be significantly different between patients with NMBP and MBP. Because patients with BP have a better prognosis than patients with usual cholangiocarcinoma, surgical resection can be curative. Therefore, early and accurate diagnosis is important in this disease.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES