What Is Known
- There is limited literature regarding the clinical spectrum and utility of available diagnostic scores for pediatric autoimmune liver disease in Indian population.
- New (year 2017) pediatric autoimmune liver disease proposed score remains untested till date.
What Is New
- Despite advanced disease presentation due to delayed diagnosis, treatment outcomes are still favorable in the local Indian cohort.
- There is no difference in the predictive value of the 3 available diagnostic scores for pediatric autoimmune liver disease.
- Any diagnostic score, based on the ease of use, may be used to support diagnosis of autoimmune liver disease.
Autoimmune liver disease (AILD), encompassing both autoimmune hepatitis (AIH) and autoimmune sclerosing cholangitis (ASC), is one of the few treatable causes of pediatric liver disease. Existing literature has provided important insights regarding its overall spectrum in the western populations (1–4). Case series from the Asia-Pacific region are still limited, thus limiting the formulation of population specific guidelines. Also, diagnostic criteria for pediatric AILD (including the new proposed score by European Society of Pediatric Gastroenterology, Hepatology and Nutrition or ESPGHAN), which were extrapolated from adult studies (5,6), require further validation in Asian populations (7,8). This work was thus aimed to study the profile of pediatric AILD subjects and validate the available diagnostic criteria in the local Indian population.
A review of the prospectively (year 2011–2017) collected data of all pediatric cases (<18 years age at presentation) presenting to a single tertiary care institute was done. Consecutive cases of Autoimmune Liver Disease (AILD) were selected for review. Diagnostic criteria for AIH included abnormal hepatic function tests and classical histological criteria (variable degree/combination of interface hepatitis, lymphoplasmacytic portal infiltrate, and hepatic pseudorosette formation) with or without presence of autoantibodies and/or elevated serum total Immunoglobulin G levels (9,10). In seronegative cases, diagnosis was re-confirmed by the excellent remission rates after immunosuppressive therapy. Diagnostic scores including original revised criteria (year 1999), simplified criteria (year 2008) and new proposed (year 2017) score were calculated at index presentation (5,6,8). ASC was diagnosed based on features of AIH along with either abnormal cholangiographic study (multiple strictures and/or dilatations of intra or extrahepatic biliary tree) or evidence of significant bile ductal injury (including fibro-obliterative cholangitis, bile duct loss, and/or periductal fibrosis) on hepatic histopathology (2,11–13). These cases were divided into either large duct disease (with abnormal cholangiograms) or small duct disease (with normal cholangiograms but classical histopathological picture). For diagnostic score(s) validation, we also included consecutive pediatric non-AILD subjects as a control group (in 1:1 ratio): Wilson disease (n = 22), non-cirrhotic portal fibrosis (n = 18) (22), drug induced liver injury (n = 12), celiac hepatitis (n = 9), primary sclerosing cholangitis (n = 7), chronic hepatitis B infection (n = 7), progressive familial intrahepatic cholestasis type 3 (n = 4), Budd-Chiari syndrome (n = 3), congenital hepatic fibrosis (n = 2), and idiopathic paucity of interlobular bile ducts (n = 1).
Data, including demographic features, clinical presentation/outcome, laboratory tests, information on extra-hepatic autoimmune disorders (EHADs), hepatic histopathology, and imaging (abdominal ultrasonography and/or magnetic resonance imaging or MRI) reports, was analyzed, after ethical committee approval. Clinical presentation was classified into acute liver failure (14,15), acute hepatitis (acute index presentation without advanced hepatic fibrosis (ie, <F4 fibrosis (16)), chronic hepatitis (prolonged or recurrent symptoms without advanced hepatic fibrosis), chronic liver disease (advanced hepatic fibrosis, with/out decompensation defined as presence of variceal bleed, ascites and/or hepatic encephalopathy), and acute on chronic liver failure (17,18). Esophageal endoscopy was done for screening for varices; findings were graded as per standard definitions (19,20). Autoantibodies tested in all subjects included antinuclear antibody (ANA), anti smooth muscle antibody (ASMA), antiliver and kidney type 1 (anti LKM-1), p-antinuclear cytoplasmic antibody (p-ANCA), and anti-soluble liver antigen (anti SLA). Liver biopsy was performed in all cases, by either percutaneous or transjugular method. Hepatic histopathology was verified independently by 2 trained hepatopathologists (A.R. and C.B.). AILD cases were classified into type 1 AIH (positive ANA and/or ASMA); type 2 AIH (positive anti LKM-1 antibody); seronegative (negative ANA/ASMA/anti LKM-1 antibody, on at least 2 occasions during course); and unclassified (varying combinations of antibodies, eg, ANA + anti LKM-1 etc) (1,3). All patients with AIH, received conventional treatment with prednisolone, initiated alone or in combination with azathioprine (based on serum total bilirubin levels), with maintenance therapy with azathioprine with or without low-dose prednisolone. ASC cases were treated with AIH regimen along with ursodeoxycholic acid and biliary interventions if needed. Further details are available in Supplementary Digital Content (https://links.lww.com/MPG/B400).
Normally distributed continuous variables were expressed as mean (SD) and the continuous variables with skewed distribution were expressed as median (range). The mean differences between the groups were tested by independent sample t test. The chi-square (or Fisher's exact) test was used to compare differences between the groups for categorical variables. Results were considered statistically significant for P < 0.05. For diagnostic score(s) validation, comparison (including area under receiver operating characteristic [AUROC] curves) was done and agreement analysis was performed (including intraclass correlation coefficient [ICC] estimates based on 2-way mixed-effects model) (21). Data were analyzed by using SPSS version 22 (IBM SPSS Statistics, IBM Corporation, Armonk, NY).
A total of 85 subjects were diagnosed as having AILD (70 AIH and 15 ASC subjects, Table 1). Disease onset was seen in second half of first decade in both entities (8.2 and 9 years in AIH and ASC, respectively), with a younger age at diagnosis for AIH cases (9.6 and 14 years, respectively). There was a slight female preponderance in AIH subjects, compared to male preponderance in ASC group. Clinical presentation in AIH subjects included acute presentation in almost one-third of the cases (35.7%), as against predominant chronic presentation in vast majority (86.7) % of ASC cases (Fig. 1). More than one-third cases in both groups had evidence of decompensation of liver disease, while vast majority (70%–100%) had evidence of portal hypertension.
Patients of both groups had evidence of poor hepatic synthetic functions suggestive of advanced disease at presentation (Table 2). On the basis of autoantibodies, subjects could be classified into predominant type 1 AIH (45.7% and 53.3%, respectively), while there was only 1 patient with type 2 AIH in ASC group. Seronegative AILD was seen in about one-fourth of the cases (23.5%). TPMT (Thiopurine S-Methyltransferase) mutation testing (in 17 subjects) did not reveal any positive results. Esophageal endoscopy was done in 68 (80%) subjects; findings included no varices (26, 30.6%), small varices (8, 9.4%), large varices (34, 40%), gastric varices (6, 7%); and portal hypertensive gastropathy (29, 34.2%).
Magnetic Resonance Imaging
MR cholangiopancreatopgraphy (MRCP) was available in all cases of ASC and 50% (n = 35) cases of AIH. ASC cases showed large duct disease and small duct disease in 8 and 7 subjects, respectively; findings included stricture/dilatation/irregularity of biliary tree (involving intrahepatic ducts only in 6 subjects and both extrahepatic and intrahepatic duct involvement in 2 subjects). No evidence of calculous disease was noted. In contrast, 11 (15.7%) AIH subjects also showed mild biliary changes; 2 with mild common bile duct (CBD) indentations; 5 with subtle prominence and irregularity of intra-hepatic biliary radicals (IHBRs); 3 with asymptomatic cholelithiasis; and 1 with chronic symptomatic calculous cholecystic changes.
Liver biopsy was available in all 85 subjects with majority (80%–86.7%) being obtained before starting medical therapy (Table 2). Interface hepatitis and lymphoplasmacytic inflammation were present in majority (>90%) in both groups. Emperipolesis, regarded a typical sign of AIH, could be demonstrated only in 20% cases of AIH (SDC-2, https://links.lww.com/MPG/B400). Classical signs of significant biliary damage, characterized the ASC group; with ductopenia in 40% cases and periductal concentric fibrosis in 26.7% cases. Although typically seen in ASC group, signs of mild biliary injury were also seen in a significant number of AIH cases; including lymphocytic nondestructive cholangiolitis (12.9%).
Extra-hepatic Autoimmune Disorders
Overall, 38 (44.7%) subjects had evidence of EHADs with the most common EHADs being autoimmune hemolytic anemia (AIHA, n = 30, 35.3% subjects) and celiac disease (CD, n = 10, 11.8% subjects); further details are available in SDC-3 and 4, https://links.lww.com/MPG/B400. Comparison of AILD cases with and without associated EHADs, revealed that those with EHADs had higher proportion of decompensated liver disease at presentation (odds ratio 2.6, 95% confidence interval [CI] 1.1 to 6.4, P = 0.046), more severe anemia (mean difference −1.78, 95% CI −2.66 to −0.89, P < 0.001), and hypoalbuminemia (mean difference −0.38, 95% CI −0.72 to −0.04, P = 0.027) (SDC-5, https://links.lww.com/MPG/B400).
Of the AIH subjects (n = 70), at a median follow-up of 16 months (interquartile range or IQR 7 to 38.5 months, total range 1–80 months), 55 children (78.6%) achieved remission, while 7 (10%) did not achieve remission but improved medically; 5 children were lost to follow-up after initial diagnosis and 2 children underwent liver transplantation (LT) at index presentation. Six children died over the study period, including 4 children who achieved initial remission and 1 child who died after LT. Two patients were later listed for LT, in view of worsening hepatic dysfunction (despite initial remission) due to repeated disease flares secondary to poor drug compliance. Thus, good outcome (survival with native liver with medically controllable disease), at last follow-up, was seen in 56 subjects (80%) while poor outcome (death and/or, underwent or listed for LT) was seen in 9 (12.9%) subjects (Table 1, Fig. 1). In ASC (n = 15) subjects, at a median follow-up of 9 months (IQR 3.5–24 months, total range 2–57 months), 9 children (60%) achieved remission, while 3 (20%) did not achieve remission but improved medically; 1 subject died due to sepsis while on immunosuppression; and 2 subjects were listed for LT in view of worsening hepatic functions despite treatment. Thus, good outcome, at last follow-up, was seen in 12 subjects (80%) while poor outcome was seen in 3 (20%) subjects (Table 1, Fig. 1). Further details are available in SDC-6, https://links.lww.com/MPG/B400.
AIH cases were characterized by younger age at diagnosis, higher values of platelet count/aspartate aminotransferase/alanine aminotransferase, higher diagnostic scores, higher histopathologic incidence of bridging necrosis/pseudorosetting/parenchymal collapse and lower incidence of signs of biliary damage, compared to ASC subjects (SDC-7, https://links.lww.com/MPG/B400). Comparison between seronegative (n = 16) and seropositive AIH cases (n = 54), did not reveal any difference (SDC-8, https://links.lww.com/MPG/B400).
Validation of Diagnostic Scores
AILD group showed lesser male preponderance, and higher levels of AST and ALT, compared to non-AILD group (SDC-9, https://links.lww.com/MPG/B400). For total cohort, all 3 scores had AUROC curves exceeding 0.9 suggestive of excellent discrimination of AILD (Original criteria: 0.970, 95% CI 0.950 to 0.990, simplified criteria: 0.974, 95% CI 0.956 to 0.993; and new score: 0.959, 95% CI 0.931 to 0.986) (Table 3, SDC-10, https://links.lww.com/MPG/B400). Similar excellent results were obtained on other subgroup analysis also: only AIH subjects (SDC-11 and 12, https://links.lww.com/MPG/B400); only seropositive AIH subjects; only seronegative AIH subjects; and only ASC subjects (data not shown). When the AUROC curves of all 3 scores were compared, there was no statistical difference between them (P value between original vs simplified criteria: 0.323; P value between original criteria vs new score: 0.441; and P value between simplified criteria vs new score: 0.821). Thus, the prediction value for AILD was similar for all the 3 scores. Similar results were obtained on other subgroup analysis also as done earlier (data not shown).
Agreement analysis revealed that the overall ICC score (between all the 3 scores) was 74.5% (95% CI 67.1% to 80.5%, P < 0.001), and Cronbach's alpha was 74.5% suggestive of good reliability. The ICC between new score and simplified criteria was 81.1% (95% CI 72.3% to 86.8%, P < 0.001), and Cronbach's alpha was 82.6%; suggestive of good reliability. Similar analysis between new score and original criteria showed an ICC of 53.5% (95% CI −3.7 5 to 76%, P < 0.001), and Cronbach's alpha of 68.9%; suggestive of moderate reliability. The analysis between simplified criteria and original criteria showed an ICC of 40.3% (95% CI −10.5% to 65.9%, P < 0.001) and Cronbach's alpha of 58.1%; suggestive of poor reliability.
The present study highlights the favorable outcomes in pediatric AILDs in the local population despite having presented with advanced disease at baseline. Similar outcomes in treated cases have been documented across the world in various studies (1,3,4,23). This amply justifies the statement that all cases of pediatric AILD may deserve a trial of immunosuppressive therapy (whenever feasible), irrespective of the disease status at index presentation.
Seronegative AILD deserves special mention since they constituted almost one-fourths (23.5%) of the cohort in present study. This percentage is higher compared to other studies where this proportion varied from 7% to 16% (3,23). Probable reasons for this higher frequency may include testing of anti LC-1 antibody in only limited number of study subjects and use of Western blot method for identification of anti LKM-1, instead of the more preferred indirect immunofluorescence (IF) technique (24). But this seronegativity did not alter the disease outcomes, as was seen in a previous study also (23). Though not confirmed in the present study, such cases may have higher incidence of hematological complications (25).
Cholangiographic techniques are considered as the cornerstone to diagnose ASC (12). In the landmark study, almost 50% of total AILD cases were diagnosed as ASC based on suggestive ERCP findings (stricture/mucosal irregularity of the common bile duct/biliary radicals among others) (12). In the present study, 15.7% of AIH subjects showed some biliary changes on MRCP, similar to that seen in a previous study (3); these were deemed to be secondary to underlying cirrhotic liver as per opinion of expert hepatobiliary radiologists. This was further corroborated by histological findings where none of these patients had any significant biliary changes. But since only 50% of AIH cases underwent MRCP, the total number of ASC subjects could be an underestimation. Similar caution in over-interpretation of MRCP has also been elegantly described in an adult study done by Lewin et al (26), where mild IHBR changes were seen in 24% of AIH cases without any convincing evidence of sclerosing cholangitis on MRCP or biopsy. This study thus concluded, that mild abnormalities in IHBRs result from hepatic fibrosis (and subsequent remodeling) and portosystemic collaterals, and in the absence of a cholestatic presentation or suggestive histology, MRCP screening may not be required.
Similarly, mild biliary histological changes in AIH subjects, as seen in the present study, should not be considered as a negative pointer to AIH diagnosis. Such changes have also been seen in previous pediatric studies also, varying from 20% to 40% (1,3). In a landmark adult AIH study also, one-fourth (24%) of subjects had biliary histological changes; treatment outcomes were similar to those without biliary changes (27). They thus concluded that, biliary changes can occur in classical AIH; in the absence of a clinical cholestasis, should not alter management strategies. But limited number of ASC subjects and short follow-up duration in the present study limits definite conclusions.
Though EHADs are a common occurrence in AILD, only limited literature is available from Indian subcontinent. The present report found incidence of EHADs to be around half (45.0%) of all cases of AILD; most common being AIHA (35.3%) and CD (11.8%). Prevalence of EHADs in pediatric AILD varies from 18% to 39% (3,4,12,28), while it is between 30% and 40% in adult series (29–32). Other pediatric series have reported lower incidence of AIHA (around 2%) and CD (2%) while reporting higher incidence of IBD (17%) and joint disease (4%–7%) (3,4). CD requires special attention since it may help correct underlying nutritional deficiencies and growth failure, and also in achieving higher disease remission rates and preventing disease relapse after immunosuppression withdrawal in AIH subjects (33–35). This varying distribution between different studies may be attributed to differences in the genetic background and environmental influences.
Use of scoring systems in AILD diagnosis is another important area of discussion. It has been clearly emphasized in the literature that the gold standard for AILD diagnosis is “clinical judgment” and any scoring system, whatsoever, can never supersede treating clinician's experience. Thus, overreliance on the results of any scoring system should be avoided at all costs (36,37). Thus, as has been our protocol, we did not use scoring systems for inclusion or exclusion of AILD cases. Multiple previous studies have confirmed the diagnostic ability of the scoring systems (original and revised criteria) for pediatric AILD (38–41). In the present study, all the 3 scores did equally well in diagnosing AILD. Thus, any of scores, based on the ease of use, may be utilized for AILD diagnosis. The reliability analysis between different diagnostic scores for pediatric AILD has not been attempted before. The overall reliability for all scores was good, but that of the original score with other 2 scores was only moderate-poor; this is likely due to the difference in assigning a specific value to a particular component, and also, different number of components used in the scores. The component of “Cholangiography” in the new proposed score requires further clarification (8). Subtraction of points for AIH subjects (if abnormal study) and ASC subjects (if normal study) may decrease both the scores leading to underdiagnosis. This may happen because, as described above, mild (but non specific) cholangiographic biliary changes may be seen in 15% to 24% of AIH cases (3,26). Also, due to this fact, this score missed all the small duct ASC cases (42), and thus, could diagnose only 50% of the ASC cases in the present cohort (Table 2). Thus, we propose that this component may be modified for future studies.
The present report is limited by its single center nature, restricted number of study subjects for subgroup comparisons, limited follow-up duration, and lack of definitive testing to assess autoimmune nature of EHADs.
To conclude, pediatric AILDs usually present with advanced hepatic disease in local cohort but may still have a good outcome if timely therapy can be instituted. Patients should be carefully screened for associated EHADs, especially AIHA and CD. There is no difference in the predictive value of the 3 available diagnostic scores. Larger multicenter studies are needed to provide definitive evidence.
1. Gregorio GV, Portmann B, Reid F, et al. Autoimmune hepatitis in childhood: a 20-year experience. Hepatology
2. Deneau M, Jensen MK, Holmen J, et al. Primary sclerosing cholangitis, autoimmune hepatitis, and overlap in Utah children: epidemiology and natural history. Hepatology
3. Jiménez-Rivera C, Ling SC, Ahmed N, et al. Incidence and characteristics of autoimmune hepatitis. Pediatrics
4. Rodrigues AT, Liu PM, Fagundes ED, et al. Clinical characteristics and prognosis in children and adolescents with autoimmune hepatitis and overlap syndrome. J Pediatr Gastroenterol Nutr
5. Alvarez F, Berg PA, Bianchi FB, et al. International autoimmune hepatitis group report: review of criteria for diagnosis of autoimmune hepatitis. J Hepatol
6. Hennes EM, Zeniya M, Czaja AJ, et al. Simplified criteria for the diagnosis of autoimmune hepatitis. Hepatology
7. Hiejima E, Komatsu H, Sogo T, et al. Utility of simplified criteria for the diagnosis of autoimmunehepatitis in children. J Pediatr Gastroenterol Nutr
8. Mieli-Vergani G, Vergani D, Baumann U, et al. Diagnosisand management of paediatric autoimmune liver disease: ESPGHAN Hepatology Committee Position Statement. J Pediatr Gastroenterol Nutr
9. Manns MP, Czaja AJ, Gorham JD, et al. Diagnosis and management of autoimmune hepatitis. Hepatology
10. Guindi M. Histology of autoimmune hepatitis and its variants. Clin Liver Dis
11. Czaja AJ. Frequency and nature of the variant syndromes of autoimmune liver disease. Hepatology
12. Gregorio GV, Portmann B, Karani J, et al. Autoimmune hepatitis/sclerosing cholangitis overlap syndrome in childhood: a 16-year prospective study. Hepatology
13. Deneau MR, El-Matary W, Valentino PL, et al. The natural history of primary sclerosing cholangitis in 781 children: a multicenter international collaboration. Hepatology
14. Squires RH Jr, Shneider BL, Bucuvalas J, et al. Acute liver failure in children: the first 348 patients in the pediatric acute liver failure study group. J Pediatr
15. Alam S, Khanna R, Sood V, et al. Profile and outcome of first 109 cases of paediatric acute liver failure at a specialized paediatric liver unit in India. Liver Int
16. Ishak K, Baptista A, Bianchi L, et al. Histological grading and staging of chronic hepatitis. J Hepatol
17. Sarin SK, Kedarisetty CK, Abbas Z, et al. APASL ACLF Working Party. Acute-on-chronic liver failure: consensus recommendations of the Asian Pacific Association for the Study of the Liver (APASL) 2014. Hepatol Int
18. Alam S, Lal BB, Sood V, et al. Pediatric acute-on-chronic liver failure in a specialized liver unit: Prevalence, profile, outcome and predictive factors. J Pediatr Gastroenterol Nutr
19. Garcia-Tsao G, Sanyal AJ, Grace ND, et al. Practice Guidelines Committee of the American Association for the Study of Liver Diseases; Practice Parameters Committee of the American College of Gastroenterology. Prevention and management of gastroesophageal varices and variceal hemorrhage in cirrhosis. Hepatology
20. Sarin SK, Lahoti D, Saxena SP, et al. Prevalence, classification and natural history of gastric varices: a long-term follow-up study in 568 portal hypertension patients. Hepatology
21. Koo TK, Li MY. A guideline of selecting and reportingintraclass correlation coefficients for reliability research. J Chiropr Med
22. Sood V, Lal BB, Khanna R, et al. Noncirrhotic portal fibrosis in pediatric population. J Pediatr Gastroenterol Nutr
23. Jain V, Srivastava A, Yachha SK, et al. Autoimmuneacute liver failure and seronegativeautoimmune liver disease in children: Are they different from classical disease? Eur J Gastroenterol Hepatol
24. Terziroli Beretta-Piccoli B, Mieli-Vergani G, Vergani D. Serology in autoimmunehepatitis: a clinical-practice approach. Eur J Intern Med
25. Maggiore G, Socie G, Sciveres M, et al. Seronegative autoimmune hepatitis in children: spectrum of disorders. Dig Liver Dis
26. Lewin M, Vilgrain V, Ozenne V, et al. Prevalence of sclerosing cholangitis in adults with autoimmune hepatitis: a prospective magnetic resonanceimaging and histological study. Hepatology
27. Czaja AJ, Carpenter HA. Autoimmunehepatitis with incidental histologicfeatures of bile duct injury. Hepatology
2001; 34 (4 pt 1):659–665.
28. Junge N, Tiedau M, Verboom M, et al. Human leucocyte antigens and pediatric autoimmune liver disease: diagnosis and prognosis. Eur J Pediatr
29. Muratori P, Fabbri A, Lalanne C, et al. Autoimmune liver disease and concomitant extrahepatic autoimmune disease. Eur J Gastroenterol Hepatol
30. Wong GW, Heneghan MA. Association of extrahepatic manifestations with autoimmune hepatitis. Dig Dis
2015; 33 (Suppl 2):25–35.
31. Teufel A, Weinmann A, Kahaly GJ, et al. Concurrent autoimmune diseases in patients with autoimmune hepatitis. J Clin Gastroenterol
32. Choudhuri G, Somani SK, Baba CS, et al. Autoimmune hepatitis in India: profile of an uncommon disease. BMC Gastroenterol
33. Rubio-Tapia A, Murray JA. The liver in celiac disease. Hepatology
34. Nastasio S, Sciveres M, Riva S, et al. Celiac disease-associated autoimmune hepatitis in childhood: long-term response to treatment. J Pediatr Gastroenterol Nutr
35. Colecchia A, Di Biase AR, Scaioli E, et al. Coeliac disease and autoimmune hepatitis: gluten-free diet can influence liver disease outcome. Dig Liver Dis
36. Czaja AJ. Diagnosis and management of autoimmune hepatitis: current status and future directions. Gut Liver
37. Czaja AJ. Performance parameters of the diagnostic scoring systems for autoimmune hepatitis. Hepatology
38. Mileti E, Rosenthal P, Peters MG. Validation and modification of simplified diagnostic criteria for autoimmune hepatitis in children. Clin Gastroenterol Hepatol
2012; 10:417–421. e1-2.
39. Hiejima E, Komatsu H, Sogo T, et al. Utility of simplified criteria for the diagnosis of autoimmune hepatitis in children. J Pediatr Gastroenterol Nutr
40. Ebbeson RL, Schreiber RA. Diagnosing autoimmune hepatitis in children: is the International Autoimmune Hepatitis Group scoring system useful? Clin Gastroenterol Hepatol
41. Ferri PM, Ferreira AR, Miranda DM, et al. Diagnostic criteria for autoimmune hepatitis in children: a challenge for pediatric hepatologists. World J Gastroenterol
42. Deneau MR, El-Matary W, Valentino PL, et al. The natural history of primary sclerosing cholangitis in 781 children: a multicenter, internationalcollaboration. Hepatology