自身免疫性肝炎的诊断和治疗标准(AASLD)

AASLD PRACTICE GUIDELINES Diagnosis and Treatment of Autoimmune Hepatitis Albert J. Czaja1 and Deborah K. Freese1,2 Preamble These guidelines provide a data-supported approach to the diagnosis and management of patients with autoim- mune hepatitis. They are based on the following: (1) a formal review and analysis of the published world litera- ture on autoimmune hepatitis (914 articles) (Medline Search from 1966-2002; the search term was autoim- mune hepatitis); (2) recommendations developed and published by the International Autoimmune Hepatitis Group; (3) concepts developed at the AASLD Single Topic Conference on autoimmune hepatitis in Septem- ber 1999; and (4) 40 years of combined experience by both authors in the clinical and laboratory investigation and care for patients with this disease. The guidelines, intended for use by physicians, are meant to be flexible, in contrast to “standards of care,” which are inflexible poli- cies to be followed in almost every case. They have been developed in a manner consistent with the American As- sociation for the Study of Liver Diseases’ Policy Statement on Development and Use of Practice Guidelines. Specific recommendations are based on relevant pub- lished information. In an attempt to standardize recom- mendations, the Practice Guidelines Committee of the American Association for the Study of Liver Diseases modified the categories of the Infectious Diseases Society of America’s Quality Standards. These categories are re- ported with each recommendation, using the Roman nu- merals I through IV to determine the quality of evidence upon which the recommendations are based. The catego- ries are as follows: I, evidence from multiple well-designed randomized controlled trials, each involving a number of participants to be of sufficient statistical power; II, evi- dence from at least one large, well-designed clinical trial with or without randomization, from cohort or case-con- trol analytic studies, or from well-designed meta-analysis; III, evidence based on clinical experience, descriptive studies, or reports of expert committees; and IV, not rated. Background Autoimmune hepatitis (AIH) is an unresolving inflam- mation of the liver of unknown cause.1 It is characterized by the presence of interface hepatitis and portal plasma cell infiltration on histologic examination, hypergamma- globulinemia, and autoantibodies.1,2 Autoimmune hepa- titis reflects a complex interaction between triggering factors, autoantigens, genetic predispositions, and immu- noregulatory networks.3,4 The mean annual incidence of AIH among white Northern Europeans is 1.9 per 100,000, and its point prevalence is 16.9 per 100,000.5 It accounts for 2.6% of the transplantations in Europe6 and 5.9% in the United States.7 Women are affected more than men (gender ratio, 3.6:1),8 and all ages9,10 and ethnic groups10-14 are susceptible. A prospective study has indicated that as many as 40% of patients with untreated severe disease die within 6 months of diagnosis.15 Cirrhosis develops in at least 40% of survivors16; 54% develop esophageal varices within 2 years after cirrhosis17; and 20% of individuals with esoph- ageal varices die from hemorrhage.17 Sustained serum aminotransferase levels of more than 10-fold normal or more than 5-fold normal in conjunction with serum �-globulin concentrations at least 2-fold normal identify patients with early mortality.15 Bridging necrosis or mul- tiacinar necrosis on histologic examination progresses to cirrhosis in 82% of patients within 5 years, and mortality is 45%.18 Patients with less severe laboratory and histo- logic findings fare better, but cirrhosis still develops in 49% within 15 years and death from hepatic failure oc- curs in 10%.19 An acute onset of illness is common (40%),20-23 and a fulminant presentation, characterized by hepatic encephalopathy within 8 weeks of disease on- set, is possible.24 Three randomized, controlled treatment trials pub- lished between 1971 and 1974 have established that pred- nisone alone or in combination with azathioprine improves symptoms, laboratory tests, histologic findings, and immediate survival.15,17,25 Liver transplantation has been associated with 5-year patient and graft survivals that Abbreviations: AIH, autoimmune hepatitis; ANA, antinuclear antibodies; SMA, smooth muscle antibodies; anti-LKM1; antibodies to liver/kidney microsome type 1; anti-ASGPR, antibodies to asialoglycoprotein receptor; anti-LC1, antibodies to liver-specific cytosol antigen type 1; anti-SLA/LP, antibodies to soluble liver antigen/liver pancreas; pANCA, perinuclear anti-neutrophil cytoplasmic antibod- ies; APECED, autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy. From the 1Divisions of Gastroenterology and Hepatology and 2Pediatric Gastro- enterology and Hepatology, Mayo Clinic and Mayo Foundation, Rochester, MN. Address reprint requests to: Albert J. Czaja, M.D., Mayo Clinic, 200 First Street S.W., Rochester, MN 55905. E-mail: czaja.albert@mayo.edu;fax: 507-284-0538. Copyright © 2002 by the American Association for the Study of Liver Diseases. 0270-9139/02/3602-0028$35.00/0 doi:10.1053/jhep.2002.34944 479 exceed 80%,26,27 and recurrent disease after transplanta- tion has been usually mild and manageable.6,28-35 In chil- dren, recurrence after transplantation occurs more frequently and may be more difficult to treat.35 Diagnostic Criteria Diagnosis requires the presence of characteristic fea- tures and the exclusion of other conditions that resemble AIH.36,37 Interface hepatitis (Fig. 1) is the histologic hall- mark of the syndrome,2 and portal plasma cell infiltration (Fig. 2) typifies the disorder.2,36-38 Neither histologic finding is disease specific, and the absence of portal plasma cells does not preclude the diagnosis. All patients suspected of AIH must be evaluated for hereditary (Wil- son disease, �1 antitrypsin deficiency, and genetic hemo- chromatosis), infectious (hepatitis A, B, and C infection), and drug-induced (minocycline, nitrofurantoin, isonia- zid, propylthiouracil, and �-methyldopa) liver injury, some of which may have autoimmune features. The con- ditions most likely to be confused with AIH are Wilson disease, drug-induced hepatitis, and chronic viral hepati- tis, especially chronic hepatitis C.39-41 Liver biopsy examination is essential to establish the diagnosis and evaluate disease severity to determine the need for treatment. Serum aminotransferase and �-glob- ulin levels do not predict the histologic pattern of injury or the presence or absence of cirrhosis.42 Histologic changes, such as ductopenia or destructive cholangitis, may indicate a variant syndrome of AIH and primary sclerosing cholangitis, AIH and primary biliary cirrhosis, or autoimmune cholangitis,43-46 and the findings of ste- atosis or iron overload may suggest alternative diagnoses, such as nonalcoholic fatty liver disease, Wilson disease, chronic hepatitis C, drug toxicity, or genetic hemochro- matosis.47 Autoantibodies must be present, and the con- ventional serologic markers of AIH are antinuclear antibodies (ANA), smooth muscle antibodies (SMA), and antibodies to liver/kidney microsome type 1 (anti- LKM1).48,49 Diagnostic criteria have been codified and updated by an international panel (Table 1).36,37 Differences between a definite and probable diagnosis of AIH relate mainly to the degree of serum �-globulin or immunoglobulin G elevation, levels of ANA, SMA, or anti-LKM1, and expo- sures to alcohol, medications, or infections that could cause liver injury. There is no time requirement to estab- lish chronicity, and cholestatic clinical, laboratory, and histologic changes preclude the diagnosis. The presence of antibodies to asialoglycoprotein receptor (anti-ASGPR), liver-specific cytosol antigen type 1 (anti-LC1), soluble liver antigen/liver pancreas (anti-SLA/LP), actin (anti-ac- tin), and/or perinuclear anti-neutrophil cytoplasmic anti- bodies (pANCA) support a probable diagnosis if the other conventional markers are absent. A scoring system has been proposed to assess the strength of the diagnosis (Table 2).36,37 By weighing each component of the syndrome, discrepant features can be accommodated (normal serum �-globulin level50) and bi- ases associated with isolated inconsistencies (destructive cholangitis45) can be avoided. Autoimmune hepatitis typ- ically enters remission during corticosteroid therapy and frequently relapses after drug withdrawal. These charac- teristic post-treatment responses have also been incorpo- rated into the scoring system. The score based on pretreatment features can be upgraded or downgraded by the response to treatment, and inconsistent findings that Fig. 2. Plasma cell infiltrate. Plasma cells are identified by their eccentric, clock-face nucleus and pale perinuclear cytoplasmic crescent. They are characteristic of autoimmune hepatitis, but neither pathogno- monic of the disease or required for its diagnosis. Staining by hematox- ylin-eosin; original magnification �400. Fig. 1. Interface hepatitis. The portal tract is expanded by a mono- nuclear infiltrate; the limiting plate is disrupted; and the inflammatory process extends into the acinus. Staining by hematoxylin-eosin; original magnification �200. 480 CZAJA AND FREESE HEPATOLOGY, August 2002 Table 1. Diagnostic Criteria for Autoimmune Hepatitis Requisites Diagnostic Criteria Definite Probable No genetic liver disease Normal �1 antitrypsin phenotype Partial �1 antitrypsin deficiency Normal serum ceruloplasmin, iron, and ferritin levels Nonspecific serum copper, ceruloplasmin, iron, and/or ferritin abnormalities No active viral infection No markers of current infection with hepatitis A, B, and C viruses No markers of current infection with hepatitis A, B, and C viruses No toxic or alcohol injury Daily alcohol � 25 g/d and no recent use of hepatotoxic drugs Daily alcohol � 50 g/d and no recent use of hepatotoxic drugs Laboratory features Predominant serum aminotransferase abnormality Predominant serum aminotransferase abnormality Globulin, �-globulin or immunoglobulin G level � 1.5 times normal Hypergammaglobulinemia of any degree Autoantibodies ANA, SMA, or anti-LKM1 � 1:80 in adults and �1:20 in children; no AMA ANA, SMA, or anti-LKM1 � 1:40 in adults or other autoantibodies* Histologic findings Interface hepatitis Interface hepatitis No biliary lesions, granulomas, or prominent changes suggestive of another disease No biliary lesions, granulomas, or prominent changes suggestive of another disease Abbreviation: AMA, antimitochondrial antibodies. *Includes perinuclear anti-neutrophil cytoplasmic antibodies and the not generally available antibodies to soluble liver antigen/liver pancreas, actin, liver cytosol type 1, and asialoglycoprotein receptor. Based on recommendations of the International Autoimmune Hepatitis Group (J Hepatol 1999;31:929-938). Table 2. Diagnostic Scoring System for Atypical Autoimmune Hepatitis in Adults Category Factor Score Category Factor Score Gender Female �2 Concurrent immune disease Any nonhepatic disease of an immune nature �2 Alk Phos:AST (or ALT) ratio �3 �1.5 �2 �2 Other autoantibodies* Anti-SLA/LP, actin, LC1, pANCA �2 �-globulin or IgG (times above upper limit of normal) �2.0 1.5-2.0 1.0-1.5 �1.0 �3 �2 �1 0 Histologic features Interface hepatitis Plasma cells Rosettes None of above Biliary changes† Atypical features‡ �3 �1 �1 �5 �3 �3 ANA, SMA, or anti-LKM1 titers �1:80 1:80 1:40 �1:40 �3 �2 �1 0 HLA DR3 or DR4 �1 AMA Positive �4 Treatment response Remission alone Remission with relapse �2 �3 Viral markers of active infection Positive Negative �3 �3 Hepatotoxic drugs Yes No �4 �1 Pretreatment score Definite diagnosis Probable diagnosis �15 10-15 Alcohol �25 g/d �60 g/d �2 �2 Posttreatment score Definite diagnosis Probable diagnosis �17 12-17 Abbreviations: Alk phos, serum alkaline phosphatase level; AST, serum aspartate aminotransferase level; ALT, serum alanine aminotransferase level; IgG, serum immunoglobulin G level; AMA, antimitochondrial antibodies; HLA, human leukocyte antigen. *Unconventional or generally unavailable antibodies associated with liver disease include perinuclear anti-neutrophil cytoplasmic antibodies (pANCA) and antibodies to actin, soluble liver antigen/liver pancreas (anti-SLA/LP), asialoglycoprotein receptor (ASGPR), and liver cytosol type 1 (LC1). †Includes destructive cholangitis, nondestructive cholangitis, or ductopenia. ‡Includes steatosis, iron overload consistent with genetic hemochromatosis, alcohol-induced hepatitis, viral features (ground-glass hepatocytes), or inclusions (cytomegalovirus, herpes simplex). Based on recommendations of the International Autoimmune Hepatitis Group (J Hepatol 1999;31:929-938). HEPATOLOGY, Vol. 36, No. 2, 2002 CZAJA AND FREESE 481 do not affect treatment response may not alter the diag- nosis. The definite diagnosis prior to corticosteroid treat- ment requires a score greater than 15, whereas the definite diagnosis after corticosteroid treatment requires a score greater than 17 (Table 2). The original scoring system has been validated against AIH and other chronic liver diseases. The sensitivity of the scoring system for AIH ranges from 97% to 100%,50-52 and its specificity for excluding AIH in pa- tients with chronic hepatitis C ranges from 66% to 92%.52-54 In most instances, the scoring system is unnec- essary for the diagnosis of AIH since the clinical, labora- tory, and histologic components of the syndrome are usually well defined.50 The major value of the scoring system may be in the objective assessment of variant or atypical syndromes that resemble the classical di- sease.43-46,55 Its major weakness has been in excluding cholestatic syndromes with autoimmune features. Among these disorders, the ability of the scoring system to exclude AIH has ranged from 45% to 65%.50,55 This weakness has justified revision of the scoring system to further down- grade cholestatic findings.37 Preliminary assessments of the revised system by retrospective analysis of prospec- tively acquired patient data have indicated a better perfor- mance in excluding biliary diseases.56 The diagnostic criteria for children are different from those of adults.36,37 Autoantibody titers tend to be lower in children, and the presence of autoantibodies in any titer, in combination with other requisite elements, is suf- ficient to support a definite diagnosis (Table 1). Autoan- tibodies are neither pathogenic nor disease specific,49 and their expression can vary during the course of AIH.57 A single low autoantibody titer should never exclude the diagnosis of AIH in an adult or child, nor should high titers establish the diagnosis in the absence of other sup- portive findings. Seronegative individuals may be classi- fied at presentation as having cryptogenic chronic hepatitis until conventional markers appear later in the course57-59 or until autoantibodies that are not generally available are tested.60,61 Autoantibody titers reflect the strength of the immune response, and they are useful only in diagnostic schemes to complement other features that support the diagnosis of AIH. Autoantibodies do not cause the disease nor do their levels reflect response to treatment. Accordingly, they do not need to be moni- tored.36,37,41,46 A unique form of sclerosing cholangitis in children (sometimes termed “autoimmune sclerosing cholangitis”) has been described in a single center.62,63 This disease may mimic classic AIH, and cholangiography is often needed to distinguish the disorders. The small number of re- ported cases, the lack of general agreement regarding the nature of the process, and the absence of a treatment consequence do not compel the routine performance of cholangiography in the initial evaluation. Conventional Repertoire of Autoantibodies ANA, SMA, and anti-LKM1 should be determined in all patients with clinical, laboratory, and/or histologic fea- tures that suggest the diagnosis of AIH, and they consti- tute the conventional repertoire of autoantibodies for this condition.36,37 ANA are the traditional markers of AIH, and they are present alone (13%) or with SMA (54%) in 67% of pa- tients with the disease.57 Nuclear reactivity can be assessed by indirect immunofluorescence on Hep-2 cell lines64 or by an enzyme immunoassay using microtiter plates with adsorbed recombinant or highly purified antigens.65 The nuclear targets of ANA in AIH are uncertain, and many ANA in AIH are nonreactive to the major recombinant nuclear antigens.64 Some medical centers, therefore, pre- fer to assess ANA by indirect immunofluorescence until the performance parameters of enzyme immunoassay in AIH are fully defined. ANA in AIH react against diverse recombinant nuclear antigens, including centromere, ribonucleoproteins, and ribonucleoprotein complexes.64 None of these reactivities has been associated with a specific pattern of indirect im- munofluorescence or prognostic importance.64 Further- more, the patterns of indirect immunofluorescence (homogenous versus speckled) have had no clinical signif- icance.66 ANA can be found in primary biliary cirrho- sis,67,68 primary sclerosing cholangitis,69,70 chronic viral hepatitis,71,72 drug-related hepatitis,73,74 nonalcoholic ste- atohepatitis,47,75 and alcohol-induced liver disease,76,77 and their expression can be variable in the same patient.57 SMA are directed against actin and nonactin compo- nents, including tubulin, vimentin, desmin, and skeletin, and they are also standard markers of AIH.48,49,78-80 SMA are present in 87% of patients with AIH, either as the sole marker of the disease (33%) or in conjunction with ANA (54%).57 Three types have been described using cultured fibroblasts treated with vinblastine. These are antibodies to actin, tubulin, and intermediate filaments.78-80 SMA are present in a variety of liver and nonliver diseases, and their utility as diagnostic markers depends on the clinical syndrome.69-72,80 Like ANA, SMA have a variable expres- sion in individual patients.57 Typically, SMA are dem- onstrated in the clinical laboratory by indirect immunofluorescence on murine stomach and kid- ney.49,79-81 Anti-LKM1 typically occur in the absence of SMA and ANA.82,83 Seropositivity requires reactivity against the proximal tubules of the murine kidney and the hepato- 482 CZAJA AND FREESE HEPATOLOGY, August 2002 cytes of the murine liver by indirect immunofluores- cence.49,83 Antibodies-LKM1 react with high specificity to a short linear sequence of the recombinant antigen, cytochrome mono-oxygenase CYP2D6 (P450 IID6), and they also inhibit CYP2D6 activity in vitro.84,85 These findings together with evidence that liver-infiltrating lym- phocytes have specific reactivity to CYP2D6 have impli- cated this cytochrome as an autoantigen in AIH.86 Homologies exist between CYP2D6 and the genome of the hepatitis C virus,84 and occasionally anti-LKM1 can be found in this infection.87-90 The absence of anti-LKM1 among patients with chronic hepatitis C in the United States91 may relate to environmental factors, host genetic predispositions, or regional differences in the viral geno- type associated with anti-LKM1.92,93 The anti-LKM1 found in chronic hepatitis C in Europe react to different epitopes on the recombinant CYP2D6 molecule than an- ti-LKM1 associated with AIH, and these diverse reactiv- ities distinguish the antibodies.88,94 Antibodies to LKM1 are rare in the United States, occurring in only 4% of adults with AIH.83 They have been described mainly in pediatric patients in Europe, but 20% of patients with anti-LKM1 in France and Germany are adults.82 The reasons for these regional differences in the occurrence of anti-LKM1 in AIH are unknown, but they may reflect variable host expression of CYP2D6, ge- netic differences in the immune response to the ta