Ayman A. Abdo,' Vincent G. Bain,2 Krikor Kichiaq2 and Samuel S. Lee1

Transcription

1 Ayman A. Abdo,' Vincent G. Bain,2 Krikor Kichiaq2 and Samuel S. Lee1 Recently, the autoimmune hepatitis (AIH)/primary sclerosing cholangitis (PSC) overlap syndrome has been reported increasingly. In this syndrome, patients present with features of both AIH and PSC. It has been suggested that the 2 diseases may be sequential in their occurrence, whereby patients have features of AIH and then after a number of years develop features of PSC, but clear confirmation of evolution has not been documented in adults. We describe 6 adult cases in which PSC was diagnosed many years after well-established AIH. Six patients are described in whom AIH definitely was diagnosed at presentation. No evidence of biliary disease was noted on the initial liver biopsy or endoscopic retrograde cholangiography (ERCP). All patients responded well to immunosuppressive therapy. After an average duration of follow-up of 4.6 years they became resistant to immunosuppression, and developed clear features of PSC, which was confirmed by ERCP in all patients. The average age of the patients at first presentation was 31.3 years, 2 were women and 4 were men, and 3 had ulcerative colitis. We found no specific features at presentation that could predict this evolutionary outcome. In conclusion, patients with well-established AIH can, after variable duration of follow-up, develop PSC. In patients with AIH who become resistant to immunosuppression or develop significant cholestasis, PSC should be ruled out by ERCP. (HEPATOLOGY 2002;36: ) A utoimmune hepatitis (AIH) and primary sclerosing cholangitis (PSC) generally are considered to be 2 separate autoimmune diseases of the liver. ore recently, the overlap syndrome, in which patients present with features of both AIH and PSC, has been well documented although no clear definition for this entity exists at this time.'-3 In all reported cases with overlap syndrome the patients had some features of both diseases at presentation. Some reports have suggested that the 2 diseases may be sequential in their occurrence, whereby patients have features of AIH and then after a number of years develop features of PSC.4,5 In all previous adult Abbreviations: AIH, autoimmune hepatitis; PSC, prima y sclerosing cholangitis; ERCP, endoscopic retrograde cholangiopancreatography; UHG, International Autoimmune Hepatitis Group; ANA, antinuclear antibody; ASA, anti-smooth muscle antibody; Ig, immunoglobulin; AL T, alanine aminonansferase; ALP, alkaline phosphatase; ULN, upper limit of normal. From the ILiver Unit, Division of Gastroenterology, University of Calgay, Calgay, Alberta; and2liver Unit, Division of Gastroenterology, University ofalberta, Edmonton, Alberta, Canada. ReceivedJuly 23, 2002; accepted September, Supported by an Alberta Heritage Foundation for edical Research Senior Scholarship award (S.S. L.). Address reprint requests to: Samuel S.Lee,.D., Liver Unit, Division of Gastroenterology, University of Calgay, 3330 Hospital Dr. NK Calgay, Alberta, T2N 4NI Canada. samlee@ucalgay.ca; fm: Copyright by the American Association for the Study of Liver Diseases /O2/3GOG-OO14$35.00/0 doi:i /jhep cases, PSC was not ruled out completely at presentation, either because no cholangiogram had been performed or because bile duct abnormalities were present on the initial liver biopsy. Therefore, many of these patients actually may have been cases of overlap syndrome with occult PSC at presentation or some variant form of AIH such as autoimmune cholangitis,6,7 rather than a true evolution of AIH to PSC. A single pediatric case, with endoscopic retrograde cholangiography (ERCP)-documented evolution from AIH to PSC, has been reported by Gregorio et 4.1 as part of a prospective study of 55 children with either AIH or AIH/PSC overlap syndrome. In this report, we describe 6 adult patients who clearly had AIH at presentation with definite diagnosis according to the 1999 International Autoimmune Hepatitis Group (IAHG) criteria,8 and subsequently evolved to a cholangiographically confirmed diagnosis of PSC. These patients were part of a group of 91 patients with AIH, followed-up by 2 hepatologists (V.G.B. and S.S.L.) with large referral practices. The average age of the entire AIH group was 51 years, and 14% were men. Seventy-five percent of the patients had a significantly positive antinuclear antibody (ANA), 42% had an anti-smooth muscle antibody (ASA), and 77% had elevated immunoglobulin G (IgG). Antineutrophil cytoplasmic antibody was not checked routinely. Besides the 3 patients described 1393

2 1394 ABDO ET AL. HEPATOLOGY, December 2002 Table 1. Patient Characteristics at Time of AlH Diagnosis Parameter Patient 1 Patient 2 Patient 3 Patlent 4 Patient 5 Patient 6 Age at presentation Sex Ulcerative colitis ALT (XULN)* ALP (XULN)t GGT (XULN)f Bilirubin (:mol/l) ANA (titer) ASA (titer) IgG (#L) ERCP IAHG scores (pretreatment) IAHG total score 15 Yes /640 1/ Normal No /80 1/ Normal Yes Negative Negative Normal F No /1280 1/80 33 Not performed F Yes /320 1/ Not performed No /1280 1/80 33 Not performed NOTE. Liver biopsy examination in all patients showed lymphoplasmacytic infiltrate interface hepatitis, and rosetting of liver cells. Abbreviations:, male: F, female; GGT, y-glutamyltransferase. *ALT ULN = 40 U/L. talp ULN = 130 U/L. fggt ULN = 85 U/L. Score >15 = definite diagnosis of autoimmune hepatitis. later with ulcerative colitis, among the entire group only 2 others had a concomitant diagnosis of ulcerative colitis with AIH. Of the 6 cases described later, the average age when they presented with AIH was 31.3 years (range, years). Two were women and 4 were men. Three had ulcerative colitis. Patients normalized their alanine aminotransferase (ALT) levels after diagnosis ofaih and start of immunosuppression in an average of 3.1 months. The average time from the first diagnosis to the second was 4 years (1.5-9 years). A summary of each case is presented, with the summary of the diagnostic tests and liver biopsy of the initial AIH presentation in Table 1. None of the patients had a past history of liver disease or any significant alcohol history, except patient number 5, who had a brief history of alcohol abuse 3 years before presentation. No patients had any significant family history of liver disease. All patients had a negative work-up for other causes of liver disease including: hepatitis B and C serology, antimitochondrial antibody, iron and copper studies, and a-1-antitrypsin deficiency. All patients had histologic features compatible with the diagnosis of AIH on the initial liver biopsy including interface hepatitis, predominant lymphoplasmacytic infiltration, and rosetting of liver cells. No significant biliary abnormalities were visible, such as bile duct tortuosity, duct proliferation, destructive or nondestructive cholarigitis, or ductopenia. The biopsy specimens averaged approximately 2 cm in length and contained about 5 portal tracts per specimen. After a variable duration of good control with standard immunosuppressive therapy (prednisone with or without azathioprine), with normal liver biochemistry tests (including complete normalization of ALT, alkaline phosphatase [ALP], and y-glutamyltransferase), the patients presented with features of PSC. Table 2 summarizes important features of the patients in their subsequent presentation with I SC. Patients and ethods Case 1. A 15-year-old boy known to have ulcerative colitis in remission with maintenance 5-aminosalicylate therapy presented with symptoms of fatigue and jaundice. Table 2. Patient Characteristics With Evolution to a Diagnosis of Primary Sclerosing Cholangitis Parameter Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 Patient 6 Time to ALT level normalization (mo) Time to PSC diagnosis (yr) ALT (XULN) ALP (XULN) Bilirubin (mol/l) Current status Received transplant Stable Awaiting transplant Stable Stable Received transplant

3 HEPATOLOGY, Vol. 36, No. 6, 2002 ABDO ET AL He had no abdominal pain or fever. The physical examination was normal, apart from jaundice. Liver biopsy confirmed a diagnosis ofaih, and the patient was started on prednisone 40 mg/d for 4 weeks, then slowly tapered. The ALT level decreased by half in 3 weeks and normalized by 2 months. The liver enzymes started to increase as the prednisone dose was further decreased. He was then started on azathioprine 150 mg/d, allowing the prednisone to be gradually tapered and discontinued. He remained in complete remission on azathioprine for 1 year. The ALT level then flared to 8 X the upper limit of normal (ULN) and ALP to 2 X ULN. Prednisone was restarted with normalization of liver enzymes. ERCP was performed at that time because of recurrent right upperquadrant abdominal pain, and was completely normal (Fig. 1). He remained in remission with normal liver enzymes for 2 more years on azathioprine. Then the liver enzyme levels started increasing again but in a cholestatic pattern. This flare was unresponsive to prednisone 50 mgld plus azathioprine 150 mg. ERCP was repeated, and this time showed evidence of PSC (Fig. 2). The patient later developed signs of liver failure with repeated bouts of cholangitis. He received a successful liver transplant and currently is doing well. Case 2. A 5 1-year-old male schoolteacher presented with fatigue, weight loss, and mild right upper-quadrant abdominal pain. He had no other significant medical history. Physical examination was unremarkable. Because his Fig. 1. Endoscopic retrograde cholangiogram in case 1, performed 14 months after the initial diagnosis of autoimmune hepatitis, shows a normal appearance. Fig. 2. Endoscopic retrograde cholangiogram in case 1, performed 2 years later, clearly shows biliary stricturing characteristic of sclerosing cholangitis. initial presentation was associated with right upper-quadrant abdominal pain, an ERCP was performed; this was normal. Liver biopsy, together with typical serologic findings (Table l), revealed the diagnosis of AIH and prednisone 30 mg/d was started. He had a complete response with normalization of ALT and ALP levels in 4 weeks. Prednisone was tapered successfully over several months, and the patient remained in clinical and biochemical remission for the next 5 years on no medications. At that point, he had a flare with recurrence of the original symptoms and increased liver enzymes: ALT level was 8 X ULN, ALP level was 2.5 X ULN, but bilirubin level was normal. He was treated with prednisone 20 mg/d and 150 mg of azathioprine with normalization of liver enzyme levels in 2 months and resolution of symptoms. However, he proved to be immunosuppressive dependent because the liver enzyme levels would flare with attempts to taper the prednisone. After approximately 3 years of follow-up evaluation with normal liver enzyme levels on prednisone plus azathioprine immunosuppression, he became symptomatic again with minimal elevation of liver enzyme levels. An ERCP was again performed, which showed a normal common bile duct but irregularities and pruning of intrahepatic ducts consistent with PSC. The patient currently is doing well on ursodeoxycholic acid therapy, and the immunosuppressive medications were discontinued. Case 3. A 54-year-old man had been diagnosed with ulcerative colitis at age 18, which was very well controlled

4 1396 ABDO ET AL. HEPATOLOGY, December 2002 on sulfasalazine maintenance therapy. One year earlier he had developed a bout of acute alcohol-induced pancreatitis from which he recovered completely. He stopped drinking alcohol after that event. He presented with fatigue and anorexia. No other active medical issues were detected. Physical examination revealed mild splenomegaly and spider nevi. Liver biopsy was compatible with AIH and prednisone 30 mg/d was started. The ALT level normalized by 3 months and the symptoms regressed. However, the ALP level remained slightly abnormal ( X ULN). Because of this and a history of mild intermittent upper abdominal pain, in addition to the ulcerative colitis, an ERCP was performed and was normal. The prednisone was tapered and stopped, and azathioprine was started. The ALT level was monitored closely and remained normal. Eventually, after approximately 18 months, the azathioprine was discontinued. About 3 years later, he presented with right upper-quadrant abdominal pain and elevated liver enzyme levels. Prednisone was restarted, but this time induced no significant improvement of the liver enzyme levels. A repeat liver biopsy was performed that showed well-established cirrhosis with bile ductular proliferation and neutrophilic inflammation consistent with PSC. ERCP was performed and showed pruning and irregularity of the intrahepatic ducts consistent with PSC. Prednisone was stopped, and ursodeoxycholic acid was started. The liver enzyme levels and liver function deteriorated further and ascites developed. He currently is awaiting liver transplantation. Case 4. A 25-year-old woman presented with epigastric discomfort, jaundice, pruritis, and fatigue. She had been healthy previously. The physical examination revealed jaundice and hepatomegaly. Liver biopsy diagnosed AIH and prednisone 30 mg/d was begun. The liver enzyme levels improved significantly but did not normalize until 1 year later on 40 mg of prednisone. At that time, a repeat liver biopsy showed marked improvement in inflammation. She proved to be steroid dependent because the liver enzyme levels and symptoms would flare on lower doses of prednisone. Accordingly, azathioprine was added at a dose of 1 mglkg. On these 2 drugs, she remained asymptomatic with normal liver enzyme levels for 3 years. She then had a flare of her symptoms with marked elevation of liver enzyme levels in a cholestatic pattern. This was unresponsive to high-dose steroids. An ERCP was performed that showed typical features of PSC. Immunosuppression was then stopped and ursodeoxycholate was started. She currently is in remission. Case 5. A 23-year-old woman with a 9-year history of ulcerative colitis that was well controlled on 5-aminosalicylates presented with fatigue and anorexia. The physical examination was normal except for fine reticular rash in- volving the trunk. The diagnosis of AIH was made, and prednisone 30 mg/d was started. The liver enzyme levels were completely normal within 3 months. The repeat liver biopsy showed marked improvement. However, her symptoms and liver enzyme levels would flare on withdrawal of prednisone. Treatment was switched to azathioprine 50 mg/d with maintenance of the liver enzyme levels in a normal range for 5 years without any complications. She then presented with recurrent symptoms and elevation in the liver enzyme levels now in a cholestatic pattern with no response to high doses of prednisone and azathioprine. ERCP was performed and showed typical PSC. Ursodeoxycholate was started, and the patient currently is doing well. Case 6. A 20-year-old man presented with right upper-quadrant discomfort and weight loss. There was no jaundice or obstructive symptoms. Physical examination revealed hepatasplenomegaly with no other stigmata of chronic liver disease. Abdominal ultrasound showed no evidence of biliary disease. Liver biopsy showed typical features of AIH with established cirrhosis. Prednisone 40 mg/d was started, with marked improvement in the symptoms and normalization of the ALT level after 4 months. After 18 months, jaundice and abnormal liver chemistry developed. Because of the cholestasis and lack of response to steroids and azathioprine, an ERCP was attempted, but biliary cannulation was unsuccessful. Subsequently, magnetic resonance cholangiography showed evidence of PSC. The immunosuppressive medications were stopped, (and ursodeoxycholate therapy was started. His condition continued to deteriorate, and a liver transplantation was performed successfully. Discussion Although AIH and PSC are considered 2 distinct diseases, they have many common features.9 First, although AIH usually presents with mainly a hepatitic enzyme pattern (it., mainly ALT and AST level elevations) abnormal increases in the ALP level occurs in 8 1 % of patients with AIH and up to 10% will have increases of more than 4-fold.lo The IAHG has recognized a subgroup of patients with AIH who present with a predominantly cholestatic liver enzyme patterns and has labeled them cbolestatic AIH. Second, some PSC patients may have similar autoantibody profiles to patients with AIH. Chapman et al. reported low-titer ASA in 16%, ANA in 32%, and increased serum IgG levels in 25% of PSC patients. Even higher percentages (approximately 70%) of PSC patients were reported by Zauli et a1.i2 to have these 2 autoantibodies.

5 HEPATOLOGY, Vol. 36, No. 6, 2002 ABDO ET AL Third, human leukocyte antigen haplotype studies in PSC and AIH suggest a common genetic basis for the immunologic mechanisms attributed to the diseases.13 They both have been associated with human leukocyte antigen-b8, DR3, and DR4.I4,'5 Autoimmunity seems to play an important role in PSC as it does in AIH. The majority of PSC patients have circulating immune complexes. l6 The predominant infiltrating lymphocytes usually are T4 cells with an elevated T4/T8 ratio.17 ultiple other immunologic similarities to AIH have been described including reduction of T-suppressor cytotoxic cells in blood and increased numbers of circulating B cells. 173 In their comprehensive 1993 study of AIH features in 114 PSC patients, Boberg et al19 found that 2% had diagnostic criteria of definite AIH, and 33% had criteria of probable disease. IgG level was elevated in 61% of patients tested, and ANA or ASA was positive in about 22%. In contrast, applying the 1999 IAHG criteria for AIH diagnosis, Kaya et a1.*0 reported that 1.4% of 1 patients with classic PSC could be classified as definite and only 6% as probable AIH. Despite the many similarities, a diagnosis of either AIH or PSC usually can be made with a subsequent disease course consistent with that specific diagnosis. Recently, several reports have described the simultaneous occurrence of features ofaih and PSC in the same patient, which has been called the AIHIPSC overlap yndrorne.1-3 In addition to this AIH/PSC overlap syndrome, 3 other reports have suggested that another form of overlap syndrome can occur in which the initial presentation is clearly that ofaih but later the clinical picture changes to that of PSC.4,5J However, in these 3 publications, many of the patients already showed some abnormality in the bile ducts at the initial presentation, which suggests a classic overlap syndrome rather than a clear evolution of AIH to PSC. It must be underscored that no initial ERCP studies in these patients were reported-therefore an initial overlap syndrome cannot be ruled out. To our knowledge, only one previous case of documented evolution from AIH to PSC has been reported. In that study, which followed-up 28 children with AIH and 27 with AIH/PSC overlap that the authors termed autoimmune sclerosing cholangitis, one patient (a young girl with ulcerative colitis) presented with AIH and had a normal initial ERCP. Eight years after the initial diagnosis, a follow-up ERCP confirmed the diagnosis of PSC.I Interestingly, this patient was ANA and ASA negative. Her initial liver biopsy (at the time of AIH diagnosis) did show some mild cholangitis, but surprisingly the final liver biopsy (after the ERCP diagnosis of sclerosing cholangitis) was not diagnostic of sclerosing cholangitis. cnair et al4 reported 5 adult cases presenting with features of AIH satisfying criteria of definitive AIH. All 5 patients responded well to prednisone and azathioprine with relapses occurring during reduction or withdrawal of treatment. All patients subsequently developed features of PSC and eventually were diagnosed with this condition. Two of these patients were definitely overlap syndrome because they had biliary epithelial damage on their initial liver biopsy and characteristic PSC features on ERCP at presentation. One patient had predominant biliary changes on initial liver biopsy and no ERCP was performed, whereas in the 2 remaining cases no biliary pathology was found initially but later developed with time (after 2 years in one case and after 14 years in the other case). In both cases no ERCP was performed initially. Similarly, Luketic et al.5 reported 5 patients who presented with typical features ofaih. Only one patient had normalization of ALT levels with therapy whereas the other 4 had only partial responses. With time (as much as 13 years later), increased ALP levels became the predominant abnormality and ERCP showed evidence of PSC. Again, none of these patients had an ERCP at presentation and 2 had bile duct changes on initial liver biopsy. ore importantly, when this group of patients (called atypical PSC) was compared with patients with typical AIH and patients with typical PSC the only differentiating feature between them was the response to immunosuppression, which suggests that they were actually occult overlap syndrome or some other variant form of AIH from the beginning. In the third report, van Buuren et al. described 113 PSC patients of which 4 initially were diagnosed with AIH. These 4 had marked biochemical improvement with prednisone and azathioprine therapy. After a mean follow-up duration of 8.9 years, they developed a cholestatic liver enzyme pattern and were diagnosed by ERCP as PSC. Again, ERCP was not performed at the time of the initial diagnosis. Recently, Czaja and CarpenteF reported on liver biopsies of 84 patients who satisfied international scoring criteria for AIH. Of these, 24% showed some form of bile duct abnormality including 6 with destructive cholangitis, 4 with ductopenia, and 10 with nondestructive cholangitis. Patients with or without bile duct abnormalities had similar laboratory findings, similar responses to therapy, and a similar outcome. They concluded that biliary changes can occur in classic AIH and that they are not associated with distinctive clinical features or treatment response. This suggests that some of the earlier-mentioned reported cases may have indeed been cases in which PSC occurred after well-established AIH.

6 1398 ABDO ET AL. HEPATOLOGY, December 2002 In contrast to the earlier-mentioned reports, our patients clearly represent sequential occurrence of the 2 diseases. All patients had definite AIH diagnosis according to the 1999 IAHG scoring system, and had no evidence of biliary disease on the initial liver biopsy and ERCP. They all had an excellent initial response to immunosuppression with complete normalization of liver enzyme levels and so are very unlikely to have been typical overlap syndrome. We found no clear features in our cases that might predict the future occurrence of PSC after the initial diagnosis of AIH. However, some points should be noted. First, all patients had at least a 2-fold elevation of ALP level at presentation, and 2 had more than 3-fold ALP level elevations. As mentioned previously, although 81% of AIH patients have an abnormal ALP level, only 33% have 2-fold elevations and only 10% have 4-fold increases.]" Second, 3 of the 6 patients in our group had ulcerative colitis. It has been shown recently that patients with AIH who have ulcerative colitis (about 16%) are at increased risk for PSC. In fact, a study has reported that up to 42% of patients with AIH and ulcerative colitis will have an abnormal ERCP.23 These patients are clinically and biochemically indistinguishable from patients with autoimmune hepatitis, ulcerative colitis, and normal ERCP. Among our cases, 3 of the 6 did not have underlying ulcerative colitis, and did not develop colitis on follow-up evaluation, either pre- or posttransplantation. oreover, neither of the 2 other patients with concomitant ulcerative colitis and AIH, from the entire 91-patient AIH group, developed PSC during ongoing follow-up evaluation. Therefore, this evolutionary syndrome, though perhaps more frequent in those with underlying ulcerative colitis, is by no means confined to those with colitis. Angulo et al.:i4 recently reported the long-term follow-up evaluation of 18 patients who had typical features of PSC on liver biopsy but normal ERCP. Three of these patients with small-duct PSC developed classic PSC documented by ERCP after variable periods of follow-up evaluation. In our patients, this entity (small-duct PSC) is unlikely because of the absence of any biliary changes on the initial liver biopsy and by the dramatic initial response to immunosuppressive therapy. Finally, because initial ERCP was not performed in 3 of our patients, it is impossible to definitively exclude the possibility that they may have had occult PSC at presentation. However, their initial liver biopsies showed classic appearances of AIH and, more importantly, no biliary lesions, and the subsequent clinical course including responsiveness to immunosuppressive treatment that was lost with the later emergence of cholesrasis were identical to the 3 ERCP-proven cases. Therefore, we are confident that these 3 cases, despite the absence of ERCP, represent true cases of evolution. In conclusion, we have described 6 adult patients who developed PSC after years of well-established and wellcontrolled AIH. This clearly shows that some patients with initially typical AIH can evolve to a typical PSC disease pattern. In patients with AIH who become resistant to immunosuppression or develop a cholestatic liver enzyme pattern, PSC should be ruled out by ERCP. Hopefully, with increased awareness of this phenomenon and more report:;, this unique subset of patients will be fully characterized., References 1. Gregorio GV, Porcmann B, Karani J, Harrison P, Donaldson PT, Vergani D. ieli-vergani (2. Autoimmune hepatitislaclerosing cholangitis overlap syndrome in childhood: a 16-year prospective study. HEPATOLOGY 2001; 33: Koskinas J, Raptis I, anika Z, Hadziyannis S. Overlapping syndrome of autoimmune hepatitis and primary sclerosing cholangitis associated with pyoderma gangreriosum and ulcerative colitis. Eur J Gastroenterol Hepatol 1999;11: Gohlke F, Lohse,4W, Dienes HP, Lohr H, arker-hermann E, Gerken G, eyer zum Buschenfelde KH. Evidence of an overlap syndrome of autoimmune heparitis and primary sclerosing cholangitis. J Hepatol 1996; 24: cnair AN, crloney, Portniann BC, Williams R, cfarlane lg. Autoimmune hepatitis overlapping with primary sclerosing cholangitis in five cases. Am J Gastroenterol 1998;93: Lukeric V, Gomez D, Sanyal A, Shiffnian. An atypical presentation for primary sclerosing cholangitis. Dig Dis Sci 1997;42: Ben-Ari 1, Dhillon AP, Sherlock S. Autoimmune cholangiopathy: part of the spectrum of autoimmune chronic active hepatitis. HEPAIOLOGY 1993; l8: ichieletti P, W;inless IR, Katz A, Scheuer PJ, Yeanian SJ, Bassendine F, Hcathcote J. Antimitochondrial antibody negative primary biliary cirrhosis: a distinct syndrome of autoimmune cholangitis. Gut 1994:35: Alvarez F, Berg PA, Bianchi FB, Bianchi L, Burroughs AK, Cancado EL, Chapman RW, ct al. International Autoiinrnune Hepatitis Group Report: review of criterid for diagnosis of autoimmune hepatitis. J Hepatol 1999: Roberts EA. Primary sclerosing cholangitis in children. Gastroenterol Hepatol 1999:1,4: Czaja AJ. Varianlt forms ofautoimmune hepatitis. Curr Gastroenterol Rep 1999; , 11. Chapman RWG, Arborgh BA, Rhoes J. Primary sclerosing cholangitis: a review of its clinical features, cholangiography, and hepatic histology. Gut 1980;: Zauli D, Schrunipf E, Crespi C. An autoantibody profile in primaiy sclerosing cho1angii:is. J Hepatol 1987;5: Czaja AJ, Santrach PJ. Breanndan S. Shared genetic risk factors in autoimmune liver disease. Dig Dis Sci 2001;46: Chapman RWG. Role of immune factors in the pathogenesis of primary sclerosing cholangitis. Semin Liver Dis 1991:11: Johnson PJ, cfarlane IG, Eddleston ALWF. The natural course and heterogeneity of autoimmune type chronic active hepatiris. Semin Liver Dis 1991;11: Bodenheimer I-IC, LaRusso NA, Thaycr WR. Elevated circulating immune complexes in primary sclerosing cholangitis. HEPATOLOGY 1983;3:

7 HEPATOLOGY, Vol. 36, No. 6, 2002 ABDO ET AL Whiteside TL, Lasky S, Si L, Van Thiel DH. Immunologic analyses of mononuclear cells in liver tissues and blood of patients with primary sclerosing cholangitis. HEPATOLOGY 1985;5: Lindor KD, Wiesner RH, Katzmann JA. Lymphocyte subsets in primary sclerosing cholangitis. Dig Dis Sci 1987;32: Boberg K, Aadland E, Jahnsen J, Raknerud N, Stiris, Bell H. Incidence and prevalence of primary biliary cirrhosis, primary sclerosing cholangitis, and autoimmune hepatitis in a Norwegian population. Scand J Gastroenterol 1998;33: Kaya, Angulo P, Lindor KD. Overlap of autoimmune hepatitis and primary sclerosing cholangitis: an evaluation of a modified scoring system. J Hepatol2000;33: van Buuren HR, van Hoogstraten HJE, Terkivatan T, Schalm SW, Vleggaar FP. High prevalence of autoimmune hepatitis among patients with primary sclerosing cholangitis. J Hepatol 2000;33: Czaja A, Carpenter H. Autoimmune hepatitis with incidental histologic features of bile duct injury. HEPATOLOGY 2001;34: Perdigoto R, Carpenter H, Czaja A. Frequency and significance of chronic ulcerative colitis in severe corticosteroid-treated autoimmune hepatitis. J Heparol 1992;14: Angulo P, Yaakov -K, Lindor KD. Small-duct primary sclerosing cholangitis: a long-term follow-up study. HEPATOLOGY 2002;35: