INTRODUCTION — Autoimmune hepatitis is a chronic hepatitis characterized by immunologic and autoimmune features, generally including the presence of circulating autoantibodies and a high serum gamma globulin concentration .
Type 1, or classic autoimmune hepatitis, is characterized by circulating antinuclear antibodies and/or anti-smooth muscle antibodies. It is the predominant form of autoimmune hepatitis seen in adults and adolescents (table 1) . Type 2 autoimmune hepatitis is defined by the presence of antibodies to liver/kidney microsomes (ALKM-1) and/or to a liver cytosol antigen (ALC-1). It is typically diagnosed in infants, girls, and young women. (See "Autoimmune hepatitis: Disease classification", section on 'Classification' and "Autoimmune hepatitis: Clinical manifestations and diagnosis".)
This topic will review the treatment of autoimmune hepatitis. The pathogenesis, clinical manifestations, and diagnosis of autoimmune hepatitis, as well as the treatment of variants of autoimmune hepatitis, are discussed separately.
The discussion that follows represents our approach to the treatment of autoimmune hepatitis. Treatment recommendations have also been outlined in 2010 guidelines from the American Association for the Study of Liver Diseases  and in 2011 guidelines from the British Society of Gastroenterology . Our approach and the approaches in the two guidelines are similar, but they do vary in some areas such as what level of aminotransferase elevation is required to initiate treatment and the specific approaches to treatment (eg, whether to initiate treatment with a glucocorticoid alone or in combination with azathioprine and when to recommend maintenance therapy).
REFERRAL TO A SPECIALIST — Many patients with autoimmune hepatitis are under the care of a hepatologist when the diagnosis is made, and we generally encourage management by a hepatologist. However, treatment by a primary care clinician who is familiar with the treatment of autoimmune hepatitis is reasonable. Referral to a hepatologist should be considered for patients who fail to go into remission with glucocorticoid therapy, who worsen despite therapy, or who have cirrhosis at the time of diagnosis. Patients presenting with acute liver failure should immediately be admitted to a hospital, preferably one with a liver transplantation program. (See "Acute liver failure in adults: Etiology, clinical manifestations, and diagnosis" and "Acute liver failure in adults: Management and prognosis".)
INDICATIONS FOR TREATMENT — The decision to treat a patient with autoimmune hepatitis is based on the severity of symptoms, the magnitude of the serum aminotransferase and gamma globulin elevations, histologic findings, and the potential for side effects.
We treat patients who fulfill any of the following criteria:
●Serum aminotransferase levels greater than 10-fold the upper limit of normal
●Serum gamma globulin level greater than twice the upper limit of normal
●Serum aminotransferase levels greater than twice the upper limit of normal along with:
•An elevated gamma globulin level, even if less than twice the upper limit of normal
•An elevated conjugated bilirubin level
•Interface hepatitis on biopsy
●Histologic features of bridging necrosis or multiacinar necrosis
●Cirrhosis with any degree of inflammation on biopsy (see 'Cirrhosis' below)
●Children (see 'Children' below)
Our approach is generally consistent with guidelines from American Association for the Study of Liver Diseases (AASLD) and the British Society of Gastroenterology (BSG). However, it does differ in a few ways. The AASLD guideline only recommends treatment for patients with gamma globulin levels greater than twice the upper limit of normal if the aminotransferases are at least fivefold the upper limit of normal, and does not recommend treatment for patients with gamma globulin levels less than twice the upper limit of normal unless the aminotransferases are greater than 10-fold the upper limit of normal. Our concern with the AASLD approach is that the level of serum aminotransferase or gamma globulin elevation does not correlate perfectly with the degree of histologic injury. In many cases, we have found that therapy may be required when serum aminotransferases or gamma globulin levels are elevated to a lesser degree than suggested by the AASLD guideline. Our approach also differs from the BSG guideline, which recommends treating if the serum aminotransferase levels are greater than fivefold the upper limit of normal, regardless of other criteria for treatment.
Treatment may not be required in asymptomatic patients with normal or near-normal serum aminotransferase and gamma globulin levels who have minimal necroinflammatory activity on liver biopsy. Such patients are at a relatively low risk of disease progression. However, it is reasonable to offer treatment to a patient with histologic evidence of interface hepatitis without bridging necrosis or multiacinar necrosis, particularly if the patient is young (age less than 50 years) and is unlikely to have severe side effects related to therapy. Patients who are not treated should be monitored for signs of disease progression. If symptoms develop, if the aminotransferase or gamma globulin levels increase, or if histologic evidence of active disease is seen on follow-up biopsies, treatment should be reconsidered using the above criteria. (See 'Glucocorticoid monotherapy' below and 'Monitoring and follow-up' below.)
Treatment is not recommended for patients with cirrhosis and inactive disease (characterized by the absence of inflammatory cells on liver biopsy and normal or near-normal serum aminotransferases) [3,4]. Such patients may be at increased risk for the development of glucocorticoid-related side effects, and the benefit of treatment is uncertain. On the other hand, treatment generally is recommended for patients with cirrhosis and active inflammation. (See 'Cirrhosis' below.)
GENERAL APPROACH TO TREATMENT — Our approach is to begin treatment with glucocorticoid monotherapy (prednisone or prednisolone 60 mg per day followed by a taper to a maintenance dose) in most patients (table 2 and algorithm 1 and algorithm 2). For patients with mild disease (eg, asymptomatic patients with aminotransferase levels <10 times the upper limit of normal), lower-dose prednisone monotherapy (20 mg per day) is an alternative.
In patients with moderate to severe disease who are at increased risk for side effects from glucocorticoids (eg, those with brittle diabetes, osteoporosis, emotional lability, a history of psychosis, or poorly controlled hypertension), we will use a combination of lower-dose prednisone (30 mg per day) and azathioprine (50 mg per day) for initial treatment. (See 'Induction therapy' below.)
The American Association for the Study of Liver Diseases recommends initial therapy with either glucocorticoid monotherapy or a combination of a glucocorticoid and azathioprine , whereas the British Society of Gastroenterology and European Association for the Study of the Liver recommend initial treatment with a glucocorticoid and azathioprine rather than glucocorticoid monotherapy [4,5]. For patients at high risk for glucocorticoid side effects, combination therapy permits administration of lower doses of prednisone. Budesonide may be an alternative to prednisone in these patients, though long-term data on safety and efficacy are lacking. (See 'Glucocorticoid plus azathioprine' below.)
Azathioprine is a prodrug of 6-mercaptopurine, and both have been associated with aplastic anemia in patients with no thiopurine methyltransferase (TPMT) enzyme activity (which leads to the preferential production of 6-thioguanine, which results in aplastic anemia) (figure 1). If azathioprine is used, TPMT phenotyping should be obtained prior to initiating treatment. (See "6-mercaptopurine (6-MP) metabolite monitoring and TPMT testing in patients with inflammatory bowel disease", section on 'TPMT phenotyping'.)
Subsequent management will depend on how the patient responds to the initial treatment (remission, incomplete response, failed treatment, drug intolerance) and whether the patient relapses if treatment is withdrawn. In some cases, treatment can be withdrawn completely whereas other patients require maintenance therapy. (See 'Subsequent management' below and 'Probability of responding to treatment' below.)
TESTING FOR TPMT ENZYME ACTIVITY AND DRUG METABOLITE LEVELS — Azathioprine is a prodrug of 6-mercaptopurine, and both have been associated with aplastic anemia in patients with no thiopurine methyltransferase (TPMT) enzyme activity (which leads to the preferential production of 6-thioguanine, which results in aplastic anemia) (figure 1). We obtain TPMT phenotyping prior to initiating treatment. This approach is consistent with recommendations from the US Food and Drug Administration for patients being treated with these drugs. Other adverse effects of azathioprine include gastrointestinal disturbances, lesser degrees of bone marrow suppression (in those with decreased TPMT activity), infection, pancreatitis, and malignancy. The side effects of these drugs are discussed in detail elsewhere. (See "6-mercaptopurine (6-MP) metabolite monitoring and TPMT testing in patients with inflammatory bowel disease", section on 'TPMT phenotyping' and "Pharmacology and side effects of azathioprine when used in rheumatic diseases", section on 'Adverse effects'.)
Drug metabolite levels (6-thioguanine and 6-methylmercaptopurine) can also be measured if there is concern about drug toxicity or to confirm that a patient is adherent to therapy. (See "6-mercaptopurine (6-MP) metabolite monitoring and TPMT testing in patients with inflammatory bowel disease", section on 'Following 6-TG and 6-MMP levels to predict toxicity AZA or 6-MP' and "6-mercaptopurine (6-MP) metabolite monitoring and TPMT testing in patients with inflammatory bowel disease", section on 'Demonstration of AZA/6-MP non-compliance or resistance'.)
Toxicity to azathioprine can occur independently of TPMT activity, and some patients intolerant of azathioprine can take 6-mercaptopurine without side effects . In patients with mild adverse effects (eg, gastrointestinal disturbances), we may substitute 6-mercaptopurine at one-half of the azathioprine dose.
Allopurinol, a xanthine oxidase inhibitor, can lead to the increased production of active metabolites and should generally be avoided, though it may have a role in patients who do not respond to or are intolerant of thiopurines. (See 'Patients who deteriorate despite treatment' below and "Pharmacology and side effects of azathioprine when used in rheumatic diseases", section on 'Metabolism'.)
INDUCTION THERAPY — Initial treatment for autoimmune hepatitis is typically with a glucocorticoid, with or without azathioprine or 6-mercaptopurine (table 2 and algorithm 1). Prior to initiating treatment with azathioprine or 6-mercaptopurine, patients should undergo thiopurine methyltransferase (TPMT) phenotyping. Testing for latent tuberculosis should also be considered for patients who are likely to need long-term glucocorticoid therapy. Patients should also be vaccinated against hepatitis A and B if not already immune (preferably before starting immunosuppressive therapy). Patients with fulminant hepatitis and acute liver failure require hospitalization for management of their liver failure (which may include liver transplantation). (See 'Testing for TPMT enzyme activity and drug metabolite levels' above and 'Fulminant hepatitis and acute liver failure' below and "Acute liver failure in adults: Etiology, clinical manifestations, and diagnosis" and "Acute liver failure in adults: Management and prognosis" and "Approach to diagnosis of latent tuberculosis infection (tuberculosis screening) in adults".)
While glucocorticoid monotherapy has not been directly compared with combination therapy (a glucocorticoid plus azathioprine) in large trials with long-term follow-up, limited data and clinical experience suggest that the two approaches have similar efficacy as induction therapy . Azathioprine monotherapy can be used for maintenance therapy, but is not effective for induction therapy .
In the United States, the glucocorticoid used is typically prednisone, whereas in Europe it is prednisolone (which is given at the same dose as prednisone). If 6-mercaptopurine, is used, it is given at half the dose recommended for treatment with azathioprine.
Glucocorticoid monotherapy — We prefer prednisone monotherapy (60 mg daily) for the initial treatment of most patients with autoimmune hepatitis because doing so avoids potential confusion if there are subsequent elevations in the serum aminotransferases or if additional symptoms develop after starting treatment (table 2 and algorithm 1). For patients with mild disease (eg, asymptomatic patients with aminotransferase levels <10 times the upper limit of normal), lower dose prednisone monotherapy (20 mg per day) is a reasonable alternative. In patients with moderate to severe disease who are at increased risk for glucocorticoid side effects, we will use a lower dose of prednisone (30 mg daily), with or without azathioprine (50 mg daily). A possible alternative to prednisone is budesonide (9 mg per day) in combination with azathioprine, particularly for patients at increased risk for glucocorticoid side effects, though its role in treatment requires additional study. In addition, budesonide should not be used in patients with cirrhosis because it undergoes extensive first-pass hepatic metabolism. (See 'Glucocorticoid plus azathioprine' below and 'Investigational treatments' below.)
After a week of treatment, we typically start a four-week taper of the prednisone to a maintenance dose (20 mg daily if being used as monotherapy and 10 mg daily if being used with azathioprine) (algorithm 2). The dose of azathioprine (if used) is held constant. Serum aminotransferase, bilirubin, or gamma globulin levels are checked weekly during the taper. If the patient develops recurrent symptoms or worsening laboratory values during the prednisone taper, we maintain the patient at the lowest dose that controls symptoms and is associated with improvements in laboratory values. This may require increasing the dose of prednisone again to 40 to 60 mg daily, and tapering according to the clinical response and side effects. Further treatment decisions will depend on the patient's response to initial treatment. (See 'Subsequent management' below.)
Approximately 90 percent of patients have improvement in serum aminotransferases, bilirubin, and gamma globulin within two weeks. However, histologic improvement lags behind biochemical improvement by three to eight months. Importantly, in adults, serum levels of autoantibodies do not appear to parallel disease activity and should not be used to monitor activity .
The side effects of glucocorticoids include the development of Cushingoid features, diabetes, fatty liver, hypertension, psychiatric problems, cataracts, glaucoma, osteoporosis, and compression fractures. The risk of glucocorticoid-related side effects is increased in patients with brittle diabetes, osteoporosis, emotional lability, a history of psychosis, or poorly controlled hypertension. Patients with these conditions require special precautions and/or monitoring. In such patients, consideration should be given to the use of combination therapy as part of the initial treatment regimen in order to use a lower dose of prednisone. Budesonide plus azathioprine may also be an option in such patients, though the role of budesonide is not yet clear. (See 'Glucocorticoid plus azathioprine' below and 'Investigational treatments' below.)
Glucocorticoid plus azathioprine — The addition of an immunomodulatory drug (azathioprine or 6-mercaptopurine) permits the use of lower glucocorticoid doses and may reduce glucocorticoid-related side effects. When using combination therapy, we typically give prednisone (30 mg daily) and azathioprine (50 mg daily) (algorithm 1). The dose of azathioprine can be increased to 2 mg/kg daily (maximum dose of 200 mg daily) if needed to achieve remission. (See 'Patients with an incomplete response to treatment' below.)
In Europe, azathioprine at a dose of 1 to 2 mg/kg daily is often used. If 6-mercaptopurine is used, the dose is half of the azathioprine dose. Glucocorticoid monotherapy may be preferred in settings in which there is concern related to exposing the patient to azathioprine such as in patients with pre-existing cytopenias, malignancy, or deficiency of the enzyme responsible for the metabolism of azathioprine and 6-mercaptopurine, TPMT. 6-mercaptopurine can be tried as an alternative for patients who are intolerant of azathioprine (though not for patients who fail to respond to azathioprine) . (See 'Glucocorticoid monotherapy' above and 'Testing for TPMT enzyme activity and drug metabolite levels' above.)
Combination therapy with budesonide (9 mg per day) and azathioprine may be an option for patients at high risk for glucocorticoid side effects or who do not tolerate prednisone, provided they do not have cirrhosis (since budesonide undergoes extensive first-pass hepatic metabolism) . However, long-term data on the safety and efficacy of budesonide in this setting are lacking. (See 'Investigational treatments' below.)
Patients can respond to induction therapy in several ways:
●Remission – Approximately 65 to 80 percent
●Incomplete response to therapy – Approximately 13 percent
●Failure to respond to treatment – Approximately 10 percent
In addition, some patients (approximately 10 percent) develop side effects from treatment that require discontinuation of a particular therapy.
Remission — Approximately 65 and 80 percent of patients achieve remission by 18 months and three years, respectively (average time to remission of 18 to 24 months) . The probability of achieving remission decreases after two years. Once remission has been achieved, attempts can be made to withdraw immunosuppressive therapy or patients can be transitioned to maintenance therapy. (See 'Patients in remission' below.)
Remission is defined by :
●Resolution of symptoms
●Normalization of serum aminotransferase levels
●Normalization of serum bilirubin and gamma globulin levels
●Improvement in liver histology to normal or only mild portal hepatitis (or minimal to no activity in patients with cirrhosis)
Patients with normalization of their aminotransferase levels are more likely to achieve histologic remission (absence of interface hepatitis), though histologic remission lags behind biochemical remission by several months . Normalization of laboratory tests and improvement in histology to normal or near normal is generally not observed before 12 months . (See 'Monitoring and follow-up' below and 'Patients in remission' below.)
However, the presence of normal or near-normal serum aminotransferases does not necessarily indicate histologic normalization. In one study, 20 percent of patients with normal aminotransferases had interface hepatitis on biopsy when treatment was stopped . In addition, problems with reliance on aminotransferase levels include sex-based differences that may not be reflected in a laboratory's normal values and differences in "normal" levels among laboratories. (See "Approach to the patient with abnormal liver biochemical and function tests", section on 'Aminotransferases'.)
Arthralgias frequently persist despite remission of autoimmune hepatitis, and aminotransferase levels may remain elevated in the presence of coexisting nonalcoholic fatty liver disease. Also, in some patients, gamma globulin levels remain elevated despite remission by other criteria. This may be secondary to other chronic inflammatory and/or autoimmune diseases.
Incomplete response to treatment — An incomplete response to treatment is seen in approximately 13 percent of patients . (See 'Patients with an incomplete response to treatment' below.)
An incomplete response is defined by:
●Some or no improvement in clinical, laboratory, and histologic features despite compliance with treatment for two to three years
●No worsening of the condition
Treatment failure — Approximately 10 percent of patients have clinical and laboratory deterioration despite adherence to conventional treatment . (See 'Patients who deteriorate despite treatment' below.)
Treatment failure is characterized by:
●Sustained biochemical and histologic activity leading to the development or worsening of cirrhosis with eventual complications and death
●The need for orthotopic liver transplantation
Wilson disease and the overlap syndrome of autoimmune hepatitis/primary sclerosing cholangitis (autoimmune sclerosing cholangitis) should be considered in young patients with treatment failure since the histologic findings are similar to autoimmune hepatitis. (See "Wilson disease: Diagnostic tests" and "Autoimmune hepatitis variants: Definitions and treatment", section on 'Autoimmune hepatitis-PSC overlaps'.)
Treatment failure is more frequent in three groups of patients [11,12]:
●Those with established cirrhosis
●Those who develop disease at a younger age or have had a longer duration of disease before treatment
●Those who possess the human leukocyte antigens (HLA)-B8 and/or HLA-DR3 phenotypes
The frequency of HLA-B8 has been noted to be higher in patients who are refractory to treatment and in those referred for liver transplantation [11,13]. As a result, the HLA phenotypes HLA-B8 and/or HLA-DR3 may serve as markers for severity of disease rather than merely a predisposition to treatment failure. However, sex and age may be confounding variables . (See "Autoimmune hepatitis: Pathogenesis".)
Other factors have been suggested as possible predictors of treatment failure. One study identified black race (especially among men) as a predictor of more aggressive disease and a reduced response to conventional immunosuppression . In another study, the prognosis in men was generally better than in women, although race was not specified . In a population-based study, predictors of liver-related death or the need for liver transplantation included age ≤20 years or >60 years at diagnosis, incomplete alanine aminotransferase normalization by six months, and low serum albumin concentration at diagnosis .
SUBSEQUENT MANAGEMENT — Patients may respond to treatment in multiple ways; they may go into a sustained remission, they may achieve remission and then relapse, they may have an incomplete response to treatment, they may have progressive disease despite treatment (treatment failure), or they may not tolerate treatment due to side effects. The management following induction therapy will depend on the patient's response (algorithm 1 and algorithm 2). (See 'Response to induction therapy' above.)
Patients in remission — Once remission has been established and maintained for at least 18 months, drug withdrawal can be attempted (algorithm 2).
The decision to use maintenance azathioprine or to wait and treat if there is a relapse depends on the estimated likelihood of relapse, severity of liver disease, and anticipated side effects. Some clinicians attempt to withdraw treatment at least once when a sustained remission has been achieved on treatment. We will typically attempt to withdraw treatment if a patient is in remission, but we first discuss the advantages and disadvantages of both approaches with our patients prior to making a treatment recommendation. In particular, we inform patients that 50 to 90 percent of patients relapse within 12 months of drug withdrawal . (See 'Remission' above.)
Guidelines from the American Association for the Study of Liver Diseases and the European Association for the Study of the Liver recommend an attempt at withdrawing therapy for patients in remission for at least 24 months [3,5]. The British Society of Gastroenterology guidelines recommend routine maintenance therapy in younger patients and in patients who are liver/kidney microsomes (ALKM-1)-positive or soluble liver antigen-positive . (See 'Conventional maintenance therapy' below.)
A liver biopsy may help predict which patients are likely to have successful drug withdrawal, but it is not required prior to withdrawing treatment. However, our preference is to obtain a biopsy prior to withdrawing treatment and to continue treatment if there is significant interface hepatitis or other histologic evidence of active disease. The importance of histologic findings was examined in a study with 120 patients who were in remission (defined by normal serum ALT and globulin) . Patients with persistent histologic activity on liver biopsy (Ishak histologic activity index [HAI] ≥4) were less likely than patients with histologic remission (HAI <4) to have regression of fibrosis (32 versus 60 percent) and had excess mortality (standardized mortality ratio 1.4 versus 0.7). Persistent histologic activity was also associated with all-cause death/transplantation (hazard ratio 3.1, 95% CI 1.2 to 8.1).
Alanine aminotransferase (ALT) levels prior to drug withdrawal may also help predict which patients are more likely to relapse following drug withdrawal. This was examined in a series of 28 patients who had been in remission for at least 24 months (defined as repeatedly normal serum aminotransferase levels and normal immunoglobin G [IgG] levels) prior to withdrawal of immunosuppression . After a median of 28 months follow-up, 15 patients (54 percent) were in remission and 13 patients (46 percent) had relapsed. Of the patients in long-term remission, all had ALT levels at the time of drug withdrawal that were <2 times the upper limit of normal and IgG levels <12 g/L. Among the patients who relapsed, only three had ALT levels <2 times the upper limit of normal and IgG levels <12 g/L. On multivariable analysis, higher levels of ALT were associated with the risk of relapse (hazard ratio 4.4; 95% CI 1.1-17).
If therapy is being withdrawn, the first step is typically tapering the glucocorticoid. The dose of prednisone can be decreased by 10 mg/day every week until a dose of 20 mg/day is reached. Tapering should then be by increments of 5 mg/day each week until a dose of 10 mg/day is reached. At that point, tapering should be done in increments of 2.5 mg/day every week until the drug is withdrawn. We monitor the serum aminotransferases, total bilirubin, and gamma globulin levels every three weeks during withdrawal and for three months after withdrawal. We then monitor the levels every six months for one year, and yearly thereafter.
For patients who are also being treated with azathioprine, we generally withdraw it after the glucocorticoid has been discontinued. For patients taking 50 mg daily, we simply stop the azathioprine. For patients taking higher doses, we reduce the dose by 50 mg/day every three months with serologic monitoring every three months. In some cases, azathioprine cannot be successfully withdrawn and patients may need maintenance therapy. (See 'Conventional maintenance therapy' below.)
Patients who relapse during drug withdrawal
Likelihood of relapse — In practice, 50 to 90 percent of patients who achieve remission will relapse within 12 months of drug withdrawal [4,17]. The risk of relapse can be predicted based on histologic findings prior to withdrawing treatment. Patients who achieve histologic remission have a 20 to 30 percent chance of relapse once treatment is withdrawn whereas those who have evidence of interface hepatitis have a 75 to 90 percent chance of relapse [20,21].
One report suggested that the probability of sustained remission without treatment was higher in patients who had received continuous immunosuppressive therapy for four years prior to drug withdrawal compared with those who had received continuous treatment for two, or one to two years (67 versus 17 and 10 percent, respectively) . Relapse was also more common in those with a human leukocyte antigen (HLA)-A1, HLA-B8, or HLA-DR3 phenotype. In another report, the only variable that predicted relapse after initial treatment was a higher gamma globulin level (3.3 versus 2.8 g/dL in those who relapsed versus those who did not, respectively) .
With retreatment, over 80 percent of patients will again achieve remission. Thirteen percent of patients will achieve a sustained remission if drug withdrawal is again attempted, but the majority will require maintenance therapy.
Signs of relapse — Relapse may be heralded by the development of fatigue, arthralgias, and anorexia, accompanied by a rise in serum aminotransferase levels and/or an increase in serum gamma globulin levels. A rise in serum aminotransferases to more than three times the upper limit of normal or a rise in serum gamma globulins to more than 2 g/dL correlates strongly with the presence of histologic deterioration .
Treatment of relapses — A reasonable approach following the first relapse is to resume the treatment that initially led to remission. Treatment should be initiated at induction doses, followed by glucocorticoid tapering (table 2). Once clinical remission is reestablished, attempts can be made to withdraw the drug(s) since approximately 12 to 13 percent of patients will achieve a sustained remission despite multiple relapses . Patients who continue to relapse require maintenance therapy. (See 'Induction therapy' above and 'Patients in remission' above.)
Our approach is to attempt drug withdrawal on one occasion, provided there is no interface hepatitis seen on biopsy. However, if relapse occurs, the patient is counseled that there is a likelihood of life-long therapy and that any further attempts to withdraw immunosuppression would again require liver biopsy to confirm there is no significant interface hepatitis or other histologic evidence of active disease. Moreover, in clinical practice, any further attempt to completely withdraw immunosuppression takes place over a longer weaning period.
Conventional maintenance therapy — Patients who require maintenance therapy should be treated with the lowest dose of either azathioprine or prednisone that controls symptoms and laboratory tests. Both drugs are effective in controlling the disease in the majority of patients [3,24,25]. The decision to use azathioprine either as a steroid-sparing agent or as monotherapy involves weighing the risks of long-term steroid-induced side effects (eg, osteoporosis) against those of azathioprine (eg, a low risk of malignancy). (See 'Glucocorticoid monotherapy' above and 'Glucocorticoid plus azathioprine' above.)
An alternative to conventional prednisone is budesonide, though data supporting its use for maintenance therapy are lacking. The lowest maintenance dose we use is 3 mg per day. For patients who require lower maintenance doses of glucocorticoids, it may be easier to use prednisone, because it can be adjusted in 1 mg increments. In addition, some patients with prednisone-responsive extrahepatic disease will have exacerbations when switched from prednisone to budesonide .
For patients with multiple relapses who cannot be maintained on azathioprine alone, an alternative is partial suppression. Partial suppression permits the use of lower doses of prednisone than are required for conventional maintenance therapy by aiming for serum aminotransferases that are mildly or moderately elevated rather than normal. (See 'Partial suppression' below.)
We use the following approach when transitioning patients to conventional maintenance therapy (algorithm 2):
●All patients – Regardless of whether the patient is receiving glucocorticoid monotherapy or combination therapy, the first step in switching to maintenance therapy is to taper the prednisone. The dose is reduced to the lowest level that is associated with symptom resolution and stability of the serum aminotransferases. The dose of prednisone can be decreased by 10 mg/day each week until a dose of 20 mg/day is reached. Tapering should then be by increments of 5 mg/day each week until a dose of 10 mg/day is reached. At that point, tapering should be done in increments of 2.5 mg/day each week down to a dose of 5 mg daily. If laboratory tests worsen or symptoms return, the taper should stop and the previous dose resumed.
●Patients receiving glucocorticoid monotherapy – In patients taking only prednisone in whom a prednisone-alone strategy is desired, prednisone is continued at the lowest dose that results in symptom resolution and stability of the serum aminotransferases. A typical maintenance dose of prednisone is less than 10 mg daily.
In patients taking prednisone alone in whom prednisone-free maintenance therapy with azathioprine is desired, azathioprine (2 mg/kg daily; maximum 200 mg daily) can be added once the prednisone has been tapered. After the patient has been on azathioprine for a month, the dose of prednisone can be reduced by 2.5 mg/day each month until complete withdrawal. We then taper the azathioprine by 50 mg/day every three months to the lowest dose needed to maintain remission. However, factors that predict which patients may be able to maintain remission with lower doses of azathioprine (ie, <1.5 to 2 mg/kg/day) have yet to be identified.
●Patients receiving combination therapy – If maintenance therapy with prednisone alone is desired, once the prednisone has been tapered, the azathioprine dose is reduced by 50 mg every three months with an increase in the dose of prednisone if the serum aminotransferases start to increase. A typical maintenance dose of prednisone is less than 10 mg daily.
If maintenance therapy with azathioprine alone is desired, once the prednisone has been tapered, the azathioprine dose is increased to 2 mg/kg daily (maximum 200 mg daily). After a month, the dose of prednisone can be decreased by 2.5 mg/day each month until complete withdrawal. We then taper the azathioprine by 50 mg/day every three months to the lowest dose needed to maintain remission.
If a patient develops a hepatitis flare while on a stable dose of maintenance therapy, we resume the treatment that initially resulted in remission. Once in remission, the patient can again be transitioned to maintenance therapy.
Partial suppression — Because of potentially severe steroid side effects, "partial suppression" may be preferable in patients who suffer multiple relapses and cannot be maintained on azathioprine alone. Partial suppression refers to treating patients to alleviate symptoms and to maintain serum aminotransferase levels that are mildly to moderately elevated. Partial suppression can be achieved with low doses of prednisone, with or without azathioprine. It may be particularly appropriate for postmenopausal women at significant risk for osteoporosis and in diabetic patients, who may experience significant glucose intolerance from glucocorticoid treatment. (See "Prevention and treatment of glucocorticoid-induced osteoporosis".)
This approach can lead to a significant reduction in side effects, and some data suggest it does not increase the risk of cirrhosis or death compared with conventional maintenance therapy. One report evaluated 22 patients who had had a mean of 3.4 relapses . The patients were treated with the lowest dose of medication that relieved symptoms and maintained the serum aspartate aminotransferase concentration below five times the upper limit of normal. The mean prednisone dose was 7.5 mg/day, with or without azathioprine. Their long-term outcomes (mean follow-up four years) were similar to those seen in 31 patients receiving conventional maintenance therapy. Only 1 of the 22 patients entered sustained remission allowing for drug discontinuation, while 16 continued treatment. Morbidity was less than in conventionally treated patients since steroid-induced side effects usually resolved with low-dose prednisone.
Patients with an incomplete response to treatment — The first step in assessing a patient with an incomplete response to treatment is to confirm adherence to the therapeutic regimen. Patients with an incomplete response to treatment despite adherence require long-term maintenance therapy. In patients on glucocorticoid monotherapy, azathioprine (50 mg daily) can be added. In patients taking combination therapy with glucocorticoids and azathioprine, the dose of azathioprine can be increased to 2 mg/kg daily (to a maximum of 200 mg daily). These patients are subsequently managed similarly to those who relapse following remission. (See 'Patients who relapse during drug withdrawal' above.)
Patients who deteriorate despite treatment — The optimal treatment of resistant disease is not well established. We generally treat with high doses of azathioprine and prednisone. In addition, several investigational therapies are available, though their role in treatment has yet to be established. For patients who progress to cirrhosis, liver transplantation may be required. (See 'Investigational treatments' below and 'Need for liver transplantation' below.)
Management with glucocorticoids and azathioprine — Our initial approach to treating patients who fail induction therapy is to add azathioprine or increase the dose (2mg/kg daily up to a dose of 200 mg daily) while increasing the prednisone to 40 to 60 mg/day if the patient was on a lower dose.
In patients already on azathioprine, measurement of thiopurine metabolite levels can be helpful . Low levels of 6-thioguanine (an active metabolite) may account for a poor treatment response. In such patients, addition of allopurinol may increase the levels of 6-thioguanine (figure 1). In a study of eight patients who received allopurinol in conjunction with a thiopurine, biochemical improvements were seen in all eight, with a sustained response in seven . However, if allopurinol (or another xanthine oxidase inhibitor) is used in conjunction with a thiopurine, it is important to monitor thiopurine metabolite levels. (See "6-mercaptopurine (6-MP) metabolite monitoring and TPMT testing in patients with inflammatory bowel disease", section on 'Monitoring'.)
In patients who respond to treatment with high doses of a glucocorticoid and azathioprine (with or without allopurinol), we taper the dose of prednisone by 10 mg/day each month until conventional maintenance doses are reached (10 mg/day) or to the lowest dose associated with clinical improvement (minimum of 10 mg/day). We continue the azathioprine for up to one year before considering a dose reduction. After one year of clinical improvement, we taper the dose of azathioprine by 50 mg/day each month until a dose of 50 mg/day is reached. For patients who enter remission with this regimen, we will consider withdrawing treatment. However, patients often require long-term maintenance therapy. (See 'Patients in remission' above and 'Treatment of relapses' above.)
An alternative regimen suggested in the guidelines from the American Association for the Study of Liver Diseases (AASLD) includes either monotherapy with prednisone (60 mg daily) or combination therapy with prednisone (30 mg daily) and azathioprine (150 mg daily) . The regimen is continued for at least one month, after which the dose of prednisone is reduced by 10 mg/day and the dose of azathioprine reduced by 50 mg/day after each month of clinical improvement. Dose reduction is continued until conventional maintenance doses are reached.
Patients who do not tolerate treatment — Side effects require treatment to be discontinued in approximately 10 percent of patients. A reasonable approach in this setting is to treat with either glucocorticoids or azathioprine alone (whichever was tolerated) at a dose that controls disease activity. If there is disease progression on the single agent, alternative therapies such as mycophenolate mofetil can be considered. (See 'Investigational treatments' below.)
INVESTIGATIONAL TREATMENTS — Alternative treatments such as mycophenolate mofetil, budesonide, cyclosporine, tacrolimus, and sirolimus have been studied for the treatment of patients who fail to respond to conventional treatment or to provide treatment options that minimize or avoid glucocorticoid side effects. Of these agents, our general preference, as well as that of AASLD, has been to try mycophenolate mofetil (1 g twice daily) .
●Mycophenolate mofetil – Mycophenolate mofetil (MMF) acts by blocking de novo purine synthesis. It has been used in patients with autoimmune hepatitis who are refractory to conventional treatment or who are unable to tolerate azathioprine. It generally has been given at a dose of 1 g twice daily. While studies have demonstrated biochemical and histologic improvement with MMF [29-33], side effects have led to discontinuation of the drug in up to one-third of patients studied [34-36]. In general, the results are more favorable in patients who have been intolerant of, rather than refractory to, standard treatments. (See "Mycophenolate: Overview of use and adverse effects in the treatment of rheumatic diseases", section on 'Adverse effects'.)
In one series, 59 treatment-naïve patients were treated with prednisolone plus 1.5 to 2 g/day of MMF . After a median treatment duration of 26 months, 88 percent of patients had responded clinically and biochemically, the majority within three months. A partial response was noted in the remaining patients. Overall, remission was achieved in 35 patients (59 percent), and 22 (37 percent) were able to be tapered off of prednisolone while being maintained on MMF. Nine patients had follow-up liver biopsies. Inflammation improved in all nine, though it did not resolve in any patient. In addition, there was improvement in fibrosis in three patients and stable fibrosis in six.
MMF appeared to be effective as a rescue therapy in a series of 26 children with autoimmune hepatitis who either did not respond to standard treatment or who had significant side effects from standard treatment . MMF was started at a dose of 20 mg/kg daily and increased to a dose of 40 mg/kg daily. A response to treatment (defined by a decrease in aspartate aminotransferase levels [AST]) was seen in 18 patients (69 percent). In 14 patients (54 percent) the AST normalized. After a median follow-up of 61 months, the AST remained normal in 12 patients (46 percent).
Larger studies that include routine histologic evaluation of patients before and after treatment are needed.
●Budesonide – The role of budesonide in treating autoimmune hepatitis has not been extensively studied. Case series and a couple of randomized trials have suggested that, in patients without cirrhosis, it is effective and has fewer side effects than prednisone [38-44]. Side effects are decreased because it undergoes extensive first-pass hepatic metabolism (as a result, it should not be used in patients with cirrhosis). Budesonide is not effective in patients who have failed to respond to glucocorticoids .
One randomized trial included 208 patients with autoimmune hepatitis without cirrhosis who were randomly assigned to budesonide (3 mg two or three times daily) or prednisone (40 mg daily tapered to 10 mg daily) for six months . Both groups also received azathioprine (1 to 2 mg/kg daily). Complete biochemical remission was achieved significantly more often in the budesonide group (47 versus 18 percent). On subgroup analysis, complete remission was seen more often with budesonide than prednisone in adults (60 versus 39 percent) but not in children (16 versus 15 percent). Patients receiving budesonide also had significantly fewer glucocorticoid-related side effects. Patients who were initially treated with prednisone were crossed over to budesonide and observed for an additional six months. The response in these patients was not significantly different from that seen in the patients who initially received budesonide.
A second randomized trial compared budesonide with prednisone for the treatment of autoimmune hepatitis in 46 children (both given in combination with azathioprine) . There was no statistically significant difference between the budesonide and prednisone groups with regard to achieving biochemical remission (32 versus 33 percent) or in the occurrence of steroid-specific side effects (47 versus 63 percent). At the end of the trial, 42 patients received open-label budesonide treatment for an additional six months, and 19 (46 percent) achieved complete remission.
Budesonide as second line therapy (alone or in combination with other immunosuppressive agents) was studied in patients with autoimmune hepatitis who either failed or developed side effects to conventional treatment with prednisone with or without azathioprine [40,44]. For these patients, biochemical response rates ranged from 55 to 78 percent after at least six months of budesonide therapy.
●Cyclosporine — Cyclosporine is a calcineurin inhibitor that has shown promise in small studies of both children and adults. However, the use of cyclosporine may be limited by concern for side effects such as hypertension, renal insufficiency, hyperlipidemia, infection, and malignancy. (See "Pharmacology of cyclosporine and tacrolimus".)
It may be a reasonable initial alternative in children in whom compliance with glucocorticoids can be a particular problem because of unacceptable side effects. A multicenter trial included 32 children who were considered to have definite autoimmune hepatitis . Cyclosporine alone was administered for six months, initially at a dose of 4 mg/kg daily, with increases every two to three days to achieve a whole blood concentration of 250 +/- 50 mg/mL for the first three months. In patients with aminotransferases less than three times the upper limit of normal, the dose was reduced to achieve whole blood concentrations of 200 +/- 50 mg/mL for an additional three months. Low-dose prednisone (0.3 to 0.5 mg/kg daily) and azathioprine (1.5 mg/kg daily) were combined with cyclosporine during the seventh month, after which cyclosporine was discontinued over a period of 15 days, with the continuation of prednisone and azathioprine.
Two patients withdrew (one because of noncompliance and the other for liver failure that did not improve). Normal aminotransferases were achieved in 25 of the remaining 30 patients by six months and in all patients by one year. Adverse events were mild and resolved during weaning of cyclosporine. Similar benefits were described in another report that included 15 children and adolescents with type 2 autoimmune hepatitis . (See "Autoimmune hepatitis: Disease classification".)
Cyclosporine has also been studied in adults [48-51]. In an open label study with 19 patients (9 treatment naïve), reductions in serum aminotransferases and histologic activity were noted after six months of treatment .
●Tacrolimus — Tacrolimus, like cyclosporine, is a calcineurin inhibitor. It, too, has been shown in small series to be effective for treating autoimmune hepatitis, including in patients who were refractory to glucocorticoids [52-56]. As an example, in an open label study with 21 patients, improvement in biochemical parameters was seen after three months, though increases in serum urea and creatinine level were noted . In a study of 93 patients with autoimmune hepatitis who were refractory to standard therapy with prednisone and/or azathioprine, patients who received tacrolimus had higher rates of response to therapy compared with patients given mycophenolate mofetil (MMF) (57 percent versus 34 percent). The rates of liver-related deaths or liver transplantation were similar in both treatment groups .
●Sirolimus — Sirolimus is structurally similar to tacrolimus and binds the same target (FK-binding protein), but does not inhibit calcineurin. Instead, it blocks the transduction signal from the interleukin-2 receptor, thus inhibiting T and B cell proliferation. It was examined in a case series with five patients who had failed treatment with glucocorticoids and azathioprine . Three patients had also failed treatment with MMF. Sirolimus was given in addition to the patients' current immunosuppressive regimen. A >50 percent decrease in alanine aminotransferase (ALT) was achieved in four patients, and sustained normalization of the ALT was achieved in two. All of the patients were able to decrease their prednisone dose (from a mean of 40 mg daily at baseline to 17 mg daily at three months).
A series of 11 patients with autoimmune hepatitis found that treatment with infliximab led to reduction of inflammation (mean AST prior treatment 475 unit/L±466, mean AST during treatment 43 unit/L±32) as well as a reduction in immunoglobulin G (IgG) levels (pretreatment mean IgG 24.8 mg/dL±10.1, mean IgG during treatment 17.38 mg/dL±6) . Initiation of therapy was based on a mix of indications including steroid failure, azathioprine toxicity or intolerance, and having reached maximal dosing of cyclophosphamide. Infectious complications occurred in 7 out of 11 patients, and close monitoring was necessary.
SPECIAL CIRCUMSTANCES AND POPULATIONS — While the treatment approach outlined above applies to the majority of adult patients, the approach to treatment varies in some settings, including in:
●Patients with fulminant hepatitis and acute liver failure
●Women who are pregnant
●Patients with cirrhosis
●Patients with coexisting hepatitis C virus infection
●Patients with variant forms of autoimmune hepatitis (eg, overlapping autoimmune hepatitis and primary biliary cirrhosis)
Fulminant hepatitis and acute liver failure — Acute liver failure refers to the rapid development of severe acute liver injury with impaired synthetic function and encephalopathy in a person who previously had a normal liver or had well-compensated liver disease. It is crucial to identify patients with acute liver failure so that they are managed appropriately.
Patients with fulminant hepatitis and acute liver failure often require liver transplantation . Whether administration of glucocorticoids can prevent such patients from requiring a liver transplantation is unsettled, and there is concern over septic complications in patients receiving glucocorticoids. Our approach to managing such patients is to transfer all patients with fulminant hepatitis to a liver transplantation center. A trial of glucocorticoids (for two weeks or less) can then be given while performing the transplantation evaluation and closely monitoring the patient's clinical status and Model for End-stage Liver Disease (MELD) score , reserving liver transplantation for patients who do not improve. The European Association for the Study of the Liver suggests treating patients with acute severe autoimmune hepatitis with >1 mg/kg of prednisone/prednisolone daily . (See "Model for End-stage Liver Disease (MELD)" and "Acute liver failure in adults: Management and prognosis".)
Studies looking at the use of glucocorticoids for fulminant hepatitis in patients with autoimmune hepatitis have had variable results:
●A retrospective series that examined 32 patients with severe autoimmune hepatitis compared outcomes among those treated with glucocorticoids (23 patients) and those who were not treated with glucocorticoids (9 patients) . Liver transplantation was required for 10 of the patients who received glucocorticoids (48 percent) and in all of the patients who did not (p = 0.01). Death rates between those who received glucocorticoids and those who did not were similar (17 versus 22 percent).
●An observational study of 16 patients with severe or fulminant disease (including 63 percent with encephalopathy) compared outcomes in 12 patients who received glucocorticoids with four who did not . Thirteen patients underwent liver transplantation, 10 of whom had received glucocorticoids. Of the two patients treated with glucocorticoids who did not undergo liver transplantation, only one improved. Severe septic complications occurred in three patients receiving glucocorticoids. Of the four patients who did not receive glucocorticoids, three survived and one died after liver transplantation. The authors concluded that glucocorticoids are of little benefit in fulminant autoimmune hepatitis and may favor septic complications.
●In other reports, the response to glucocorticoids with severe disease has ranged from 36 to 100 percent [68-73].
One problem with comparing studies is that the definition of severe disease used can be variable and in some cases is unclear.
Studies have attempted to identify factors that predict which patients with fulminant hepatitis are more likely to respond to glucocorticoids. Unfortunately, the ability to predict who might benefit from glucocorticoids is imprecise. One study in patients with acute, severe autoimmune hepatitis (though not necessarily acute liver failure) found that failure to respond to glucocorticoids was predicted by higher initial MELD, MELD-Na, and United Kingdom End-stage Liver Disease (UKELD) scores . The median MELD score among patients who responded to treatment was 20, compared with 26 for those who did not. Similarly, for the MELD-Na score, the medians were 22 and 27, respectively, and for the UKELD score, the medians were 57 and 59, respectively.
Children — Autoimmune hepatitis tends to be more severe in children compared with adults. Therefore, treatment is usually recommended at the time of diagnosis for all children, regardless of the degree of laboratory test or biopsy abnormalities [3,4,37,75].
Treatment of autoimmune hepatitis (both type 1 and type 2) in children typically consists of a glucocorticoid (prednisone or prednisolone; 1 to 2 mg/kg daily, up to a maximum of 60 mg daily), with or without azathioprine (0.5 to 2 mg/kg daily, up to a maximum of 200 mg daily) [2,3]. Some groups recommend treating with azathioprine from the onset, whereas others reserve it for patients with severe side effects from glucocorticoids or in whom serum liver tests fail to normalize with glucocorticoid monotherapy. Our approach is to treat initially with prednisone monotherapy, with early introduction of azathioprine to abrogate against the more severe consequences of glucocorticoid use in children (bone disease, linear growth stunting, acne, and impacts on puberty).
The prednisone is tapered over four to eight weeks to a maintenance dose of 2.5 to 5 mg per day. The goal is to achieve an 80 percent reduction in the aminotransferase levels during the first eight weeks. Normalization of the serum liver tests is seen in 75 to 90 percent of children after six to nine months of treatment . Initially, serum liver tests should be checked weekly to guide reductions in the dose of prednisone (eg, the dose should not be decreased if the liver tests are not improving). Unlike in adults, autoantibody levels can be used to monitor treatment responses in children .
Azathioprine is started at a dose of 0.5 mg/kg daily and can be gradually increased to a dose of 2 mg/kg daily if needed, to a maximum of 200 mg daily (provided signs of toxicity do not develop). If azathioprine is used as part of maintenance therapy, it is typically given as 1 to 1.5 mg/kg daily (maximum 200 mg daily) in conjunction with low-dose prednisone.
Attempts can be made to gradually discontinue treatment under close serologic monitoring if the patient has two to three years of biochemical remission and histologic resolution [2,3]. However, it has been recommended that attempts to stop treatment not occur during the pubertal period or during the first three years after diagnosis because relapses are more frequent at these times. Immunosuppressive therapy can be permanently stopped without disease recurrence in about 20 to 40 percent of children with type 1 autoimmune hepatitis but rarely in children with type 2 autoimmune hepatitis [76,77]. Overall, the outcomes for children are similar regardless of whether they have type 1 or type 2 autoimmune hepatitis.
Potential alternatives to prednisone that have been studied in children include budesonide  and mycophenolate mofetil , but their role in the treatment of autoimmune hepatitis in children is not yet clear. (See 'Investigational treatments' above.)
In a meta-analysis of 15 studies of children with refractory autoimmune hepatitis, the six month overall response rates for cyclosporine (15 patients), tacrolimus (4 patients) and mycophenolate mofetil (MMF) (34 patients) were 83, 50 and 36 percent, respectively . Adverse effects were most frequently reported in cyclosporine-treated patients (64 percent experiencing at least one adverse effect) followed by tacrolimus (54 percent) and MMF (48 percent).
Pregnancy — Women appropriately treated for autoimmune hepatitis can have successful pregnancies. Usual treatment consists of glucocorticoids and/or azathioprine, both of which appear to be safe during pregnancy, though azathioprine is listed as pregnancy class D. Mycophenolate mofetil, an agent sometimes used for patients who do not respond to azathioprine, is not safe in pregnancy and should be withdrawn at least three months prior to conception and replaced with an alternative agent. Cessation of treatment during pregnancy has been associated with relapse of the disease [79,80]. (See "Safety of antiinflammatory and immunosuppressive drugs in rheumatic diseases during pregnancy and lactation" and 'Investigational treatments' above and "Mycophenolate: Overview of use and adverse effects in the treatment of rheumatic diseases", section on 'Pregnancy'.)
Pregnancy in women with autoimmune hepatitis has been associated with an increased risk of prematurity, low birth weight, and fetal loss [81,82]. The presence of antibodies to soluble liver/liver pancreas antigen (SLA/LP) and Ro/SSA (Sjögren's syndrome A) antigen was associated with adverse pregnancy outcomes in one report. Patients need to be monitored carefully during pregnancy and several months post-partum because of the risk of flares in disease activity [79,80].
Cirrhosis — Cirrhosis is common in patients with autoimmune hepatitis, with some patients requiring liver transplantation for end-stage liver disease [83,84]. Treatment is generally not recommended for patients with cirrhosis and inactive disease. On the other hand, treatment is appropriate for patients with active disease and should not be withheld from patients with compensated or decompensated cirrhosis based solely on the presence of cirrhosis. Treatment is the same as that used in patients who do not have cirrhosis. With treatment, many patients do well. (See 'Induction therapy' above and 'Prognosis' below.)
The response to treatment may be excellent, even in those who have already experienced bleeding from esophageal varices or who have significant ascites. The 10-year survival for treated patients, including those with cirrhosis, exceeds 90 percent (figure 2) .
In a series of 97 patients, reversibility of fibrosis was demonstrated in 8 of 14 patients who had evaluable biopsies both before and after treatment . An additional 25 patients with cirrhosis had successful medical treatment but did not have biopsies performed after treatment.
The general management of patients with cirrhosis due to autoimmune hepatitis is the same as that for patients with cirrhosis due to other causes and includes screening for esophageal varices and treating complications of cirrhosis. Patients who develop decompensated liver disease may require liver transplantation. (See "Cirrhosis in adults: Overview of complications, general management, and prognosis".)
While a survival benefit for surveillance in patients with autoimmune hepatitis and cirrhosis has not been proven, we agree with the suggestion of the American Association for the Study of Liver Diseases that patients with cirrhosis undergo surveillance for hepatocellular carcinoma . (See "Prevention of hepatocellular carcinoma and recommendations for surveillance in adults with chronic liver disease".)
Coexisting hepatitis C virus infection — In rare instances autoimmune hepatitis is accompanied by chronic hepatitis C virus (HCV) infection. Treatment for HCV infection does not need to be delayed if patients are going to be receiving interferon-free regimens using direct-acting antivirals. However, in clinical situations where interferon will be used, treatment should first be directed toward autoimmune hepatitis because of the danger of exacerbating autoimmune hepatitis with interferon-based treatment . Although this approach may result in increased viral levels of HCV, it is the safer initial strategy. (See "Overview of the management of chronic hepatitis C virus infection", section on 'Regimen selection' and "Principles of interferon therapy in liver disease and the induction of autoimmunity", section on 'Liver disease'.)
Variants of autoimmune hepatitis — The management of patients with variant forms of autoimmune hepatitis, such as autoimmune hepatitis that overlaps with primary biliary cirrhosis or primary sclerosing cholangitis, is discussed elsewhere. (See "Autoimmune hepatitis variants: Definitions and treatment".)
MONITORING AND FOLLOW-UP — Patients require monitoring during treatment to assess for a treatment response and to monitor for side effects of treatment.
Patients receiving treatment — For patients on combination therapy, monitoring includes weekly liver tests, blood sugar, and blood cell count for four weeks and then every one to three months, along with an immunoglobulin level . For patients being treated with glucocorticoid monotherapy, the monitoring is the same, except that blood cell counts do not need to be followed as frequently.
For patients who go into remission, we obtain a liver biopsy prior to withdrawing therapy. Patients are then followed with serologic testing during drug tapering and periodically thereafter. (See 'Patients in remission' below.)
For patients receiving maintenance therapy, once on a stable dose of medication, we monitor the serum aminotransferases, bilirubin, and gamma globulin levels every six months. In addition, we obtain repeat liver biopsies to assess for disease progression every two years or prior to withdrawing treatment.
In patients on long-term glucocorticoid therapy, it is important to monitor for and attempt to prevent glucocorticoid side effects. Patients should have bone density testing at the beginning of glucocorticoid therapy and then every one to two years; adjunctive therapies for preventing or treating bone disease include a regular weight-bearing exercise program, supplementation with vitamin D and calcium, and where appropriate, bone active agents such as bisphosphonates. (See "Screening for osteoporosis", section on 'Bone density' and "Prevention and treatment of glucocorticoid-induced osteoporosis".)
After a year on glucocorticoids, patients should also undergo screening for glaucoma and cataracts.
Patients in remission — For patients who stop treatment after attaining remission, we monitor the serum aminotransferases, total bilirubin, and gamma globulin levels every three weeks during glucocorticoid withdrawal and for three months after withdrawal. If patients are also being tapered off of azathioprine, which is typically done after tapering the glucocorticoid, we monitor the levels every three months during the azathioprine withdrawal. Once off all medications, we then monitor the levels every six months for one year, and yearly thereafter. Treatment should be reinstituted if a relapse occurs. (See 'Treatment of relapses' above.)
Patients not receiving treatment — In patients who do not have an indication for treatment, we monitor for signs of disease progression, with evaluation of the aminotransferase, total bilirubin, and gamma globulin levels at three- to six-month intervals. We generally rebiopsy patients who are not receiving treatment to look for histologic evidence of disease progression every two years.
If symptoms develop, if the aminotransferase or gamma globulin levels increase, or if histologic evidence of active disease is seen on follow-up biopsies, treatment should be reconsidered. (See 'Indications for treatment' above.)
PROGNOSIS — Overall, the prognosis in patients who are treated for autoimmune hepatitis is good, despite the fact that many have cirrhosis at the time of diagnosis or develop cirrhosis during follow-up.
Probability of responding to treatment — The goal of therapy is to achieve a sustained remission without the need for medications. However, this endpoint is achievable in only 10 to 40 percent of patients based on our experience and the published literature .
Overall, it is estimated that among patients receiving treatment:
●Up to 90 percent of patients with moderate to severe autoimmune hepatitis will respond with a decrease in serum aminotransferases along with symptom improvement within two weeks [3,4]. In the majority of these patients, the serum aminotransferases will fall into the normal range, generally after 12 or more months of treatment [22,91-93].
●Of those who achieve remission, 50 to 90 percent will relapse within 12 months of drug withdrawal [4,17]. With retreatment, 80 percent will again achieve remission. (See 'Patients who relapse during drug withdrawal' above.)
●Clinical, laboratory, and histologic parameters improve but fail to normalize (incomplete response) in 13 percent .
●Clinical, laboratory, and histologic parameters worsen in 10 percent (treatment failure) .
Risk of cirrhosis — Up to 30 percent of adult patients and almost half of pediatric patients will have cirrhosis at the time of diagnosis, and cirrhosis develops in another 30 to 50 percent of patients during follow-up [89,94]. Patients who suffer from multiple relapses are at increased risk for cirrhosis compared with patients with sustained remission after initial treatment (38 versus 4 percent) .
Risk of hepatocellular carcinoma — Patients with autoimmune hepatitis are at increased risk for hepatocellular carcinoma, particularly if they have cirrhosis. In a meta-analysis of 25 studies, including over 6500 patients with autoimmune hepatitis who were followed for a median of eight years, the incidence rate of hepatocellular carcinoma was 3.1 (95% CI 2.2-4.2) per 1000 person-years . In sixteen studies of patients with autoimmune hepatitis and cirrhosis, the incidence rate of hepatocellular carcinoma was 10.1 (95% CI 6.9-14.7) per 1000 patient-years, and we recommend surveillance for these patients. (See "Prevention of hepatocellular carcinoma and recommendations for surveillance in adults with chronic liver disease".)
Need for liver transplantation — Liver transplantation may be required for patients with acute liver failure due to autoimmune hepatitis, for patients with decompensated cirrhosis, and for patients with hepatocellular carcinoma. Ultimately, 10 to 20 percent of patients with autoimmune hepatitis will require liver transplantation . Among patients who undergo transplantation, 20 to 30 percent will have a recurrence of autoimmune hepatitis [95-98].
We refer patients for a transplantation evaluation if they have:
●Acute liver failure.
●Hepatocellular carcinoma meeting criteria for transplantation (see "Liver transplantation for hepatocellular carcinoma", section on 'Indications for transplantation' and "Model for End-stage Liver Disease (MELD)", section on 'MELDNa').
This approach is consistent with that recommended by the American Association for the Study of Liver Diseases . The British Society of Gastroenterology recommendations are similar though they also suggest referral for patients with severe disease and serum aminotransferases that show no or a very slow response to treatment .
Mortality — The prognosis in patients treated for autoimmune hepatitis is good, although they do have a lower overall transplant-free survival than the general population, with an estimated standardized mortality ratio of 1.6 to 2.1 [91,99]. High mortality rates have been noted in symptomatic patients who are untreated. Symptomatic patients from the control arms of early treatment trials had mortality rates of approximately 50 percent at five years [9,100-102]. Survival rates without treatment among patients who are asymptomatic are less clear. Some studies suggest that untreated patients have a lower survival rate than treated patients whereas others suggest that survival is similar [103-105].
The 10-year survival rate for patients with autoimmune hepatitis who receive treatment is approximately 90 percent . Among patients with cirrhosis who are treated, the 10-year survival rate is similar to that seen in patients with autoimmune hepatitis without cirrhosis; however, survival is lower after 10 to 20 years (figure 2) .
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Beyond the Basics topics (see "Patient education: Autoimmune hepatitis (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●The decision to treat a patient with autoimmune hepatitis should be individualized and based on the severity of symptoms, the degree of elevation of serum aminotransferases and gamma globulin levels, histologic findings, and the potential for side effects. (See 'Indications for treatment' above.)
●Without treatment, the survival rate in patients with symptomatic autoimmune hepatitis at five years is approximately 50 percent. However, with treatment, the 10-year survival rate is approximately 90 percent. (See 'Prognosis' above.)
●In adult patients with autoimmune hepatitis, we suggest treatment, rather than expectant management, if any of the following criteria are fulfilled (Grade 2B) (see 'Indications for treatment' above):
•Serum aminotransferase levels greater than 10-fold the upper limit of normal
•Serum gamma globulin level greater than twice the upper limit of normal
•Histologic features of bridging necrosis or multiacinar necrosis
In adult patients who lack the above indications for treatment, we suggest treatment if the serum aminotransferase level is greater than twice the upper limit of normal along with (Grade 2C):
•An elevated gamma globulin level, even if less than twice the upper limit of normal
•An elevated conjugated bilirubin
•Interface hepatitis on biopsy
In our experience, many of these patients will have disease progression if not treated. Finally, it is reasonable to offer treatment to a patient with histologic evidence of interface hepatitis who does not meet other criteria, particularly if the patient is young (age less than 50 years) and is unlikely to have severe side effects related to therapy.
●In adults, we suggest initial treatment with glucocorticoid monotherapy rather than combination therapy with a glucocorticoid and azathioprine (table 2 and algorithm 1) (Grade 2C). In the United States, the preferred glucocorticoid is often prednisone, whereas in Europe it is prednisolone. (See 'Glucocorticoid monotherapy' above.)
We prefer glucocorticoid monotherapy for the initial treatment of most patients because doing so avoids potential confusion if there are elevations in the serum aminotransferases or if additional symptoms develop after starting treatment. When used as monotherapy, prednisone or prednisolone is usually given at a dose of 60 mg, followed by a taper. For patients with mild disease (eg, asymptomatic patients with aminotransferase levels <10 times the upper limit of normal), lower dose prednisone monotherapy (20 mg per day) is an alternative.
In patients with moderate or severe disease who are at increased risk for side effects from glucocorticoids (eg, those with brittle diabetes, osteoporosis, emotional lability, a history of psychosis, or poorly controlled hypertension), using a lower-dose glucocorticoid (eg, prednisone 30 mg per day) for monotherapy or in combination with azathioprine (50 mg per day) for initial treatment is a reasonable alternative. An alternative to azathioprine is 6-mercaptopurine, which is given at half the dose of azathioprine. Treatment with budesonide and azathioprine may also be an alternative in these patients, provided the patients do not have cirrhosis, though the long-term safety and efficacy of this approach is unclear. (See 'Glucocorticoid plus azathioprine' above.)
●We recommend treatment for all children with autoimmune hepatitis, rather than reserving treatment for children who meet the above criteria (Grade 1B). Autoimmune hepatitis tends to be more severe in children compared with adults possibly because of a delay in diagnosis, and up to one-half have cirrhosis at the time of diagnosis. (See 'Children' above.)
In children, we suggest initial treatment with a glucocorticoid (Grade 2C). Some groups recommend treating with azathioprine from the onset. We introduce azathioprine early to abrogate against the more severe consequences of glucocorticoid use in children (bone disease, linear growth stunting, acne, and impacts on puberty). Prednisone or prednisolone is given as 1 to 2 mg/kg daily (maximum of 60 mg daily) and is then tapered. Azathioprine is given as 0.5 to 2 mg/kg daily (maximum 200 mg daily). (See 'Children' above.)
●Thiopurine methyltransferase (TPMT) enzyme activity should be determined before starting azathioprine or 6-mercaptopurine. (See "6-mercaptopurine (6-MP) metabolite monitoring and TPMT testing in patients with inflammatory bowel disease", section on 'TPMT enzyme determination'.)
●Subsequent management in both adults and children will depend on how the patient responds to the initial treatment (remission, incomplete response, failed treatment, drug intolerance) and whether the patient relapses if treatment is withdrawn (algorithm 1 and algorithm 2). In some cases, treatment can be withdrawn completely whereas other patients require maintenance therapy. (See 'Response to induction therapy' above and 'Subsequent management' above and 'Children' above.)
●While the treatment approach outlined above applies to the majority of patients, the approach to treatment varies in some settings including in (see 'Special circumstances and populations' above):
•Patients with fulminant hepatitis and acute liver failure (see 'Fulminant hepatitis and acute liver failure' above)
•Women who are pregnant (see 'Pregnancy' above)
•Patients with cirrhosis (see 'Cirrhosis' above)
•Patients with coexisting hepatitis C virus infection (see 'Coexisting hepatitis C virus infection' above)
•Patients with variant forms of autoimmune hepatitis (eg, overlapping autoimmune hepatitis and primary biliary cirrhosis) (see "Autoimmune hepatitis variants: Definitions and treatment", section on 'Treatment')
●Guidelines for the management of autoimmune hepatitis were issued in 2010 by the American Association for the Study of Liver Diseases and in 2011 by the British Society of Gastroenterology.