Autoimmune Hepatitis

Autoimmune hepatitis (AIH) is a chronic, immune-mediated inflammatory liver disease of unknown trigger, characterized by elevated aminotransferases, circulating autoantibodies, elevated serum immunoglobulin G (IgG), and a characteristic histologic pattern — interface hepatitis — on liver biopsy. Unlike some other liver diseases covered elsewhere on this platform, biopsy remains a genuinely active part of both confirming the diagnosis and staging severity, not merely an optional confirmatory step.^[1]

AIH affects women disproportionately (roughly 70-80% of cases) and can present at any age. Two broad serologic/clinical subtypes are recognized: type 1 — positive antinuclear antibody (ANA) and/or anti-smooth-muscle antibody (ASMA), the large majority of cases, any age — and type 2 — positive anti-liver-kidney microsomal type 1 antibody (anti-LKM1), less common, more often affecting children and young adults, and generally following a more severe, relapsing course.^[1][2]

Clinically, AIH presents as one of two genuinely distinct phenotypes that this chapter treats as a central organizing concept: a more common indolent chronic presentation — often discovered incidentally on routine liver enzymes, sometimes only after cirrhosis has already developed — and a less common but clinically urgent acute severe presentation that can closely mimic, and without prompt recognition and treatment can evolve directly into, acute liver failure. The Acute Liver Failure chapter's King's College Criteria, hepatic encephalopathy grading, and general supportive-care principles all apply directly to a deteriorating acute severe AIH presentation and are cross-referenced rather than repeated here (see Red Flags & Refer and Choose Initial Treatment).

Chronic, indolent presentation — the more common pattern

Many patients have no symptoms at all, or only nonspecific fatigue, mild right-upper-quadrant discomfort, arthralgias, or pruritus, with the diagnosis first suspected because of incidentally elevated transaminases on routine bloodwork. Because this presentation is so easy to miss, a meaningful proportion of patients — historically cited at around a quarter to a third — already have cirrhosis at the time AIH is first diagnosed, underscoring why unexplained transaminase elevation deserves a genuine diagnostic work-up rather than being dismissed as incidental.^[1] If cirrhosis has already developed, the usual stigmata of chronic liver disease may be present — see the Cirrhosis chapter for the full differential of spider angiomata, palmar erythema, splenomegaly, and related findings, not repeated here.

Acute severe presentation — the clinically urgent pattern

A meaningful minority of patients instead present acutely: rapid-onset jaundice, marked malaise, and, in the most severe cases, evolving coagulopathy and hepatic encephalopathy indistinguishable at the bedside from acute liver failure of any other cause. This is commonly described as an INR ≥1.5 without hepatic encephalopathy ("acute severe AIH") versus INR ≥1.5 with any degree of encephalopathy (frank acute liver failure) — see Red Flags & Refer for how urgently each requires escalation.^[1] Notably, biopsy in an apparently "acute" presentation not infrequently reveals pre-existing fibrosis, suggesting many of these cases are really an acute exacerbation of previously unrecognized chronic disease rather than a genuinely new process — a distinction that does not change acute management but is worth knowing.

Extrahepatic autoimmune associations

AIH commonly coexists with other autoimmune conditions — autoimmune thyroiditis, celiac disease, type 1 diabetes, rheumatoid arthritis, and inflammatory bowel disease among the most frequently reported — and can overlap with the other autoimmune liver diseases, primary biliary cholangitis and primary sclerosing cholangitis (see Special Populations).^[2]

• Female sex — women make up roughly 70-80% of cases, across both type 1 and type 2 disease.
• A personal or family history of other autoimmune disease — autoimmune thyroiditis, type 1 diabetes, celiac disease, and others — increases suspicion when unexplained transaminase elevation is found.
• Genetic susceptibility — specific HLA class II haplotypes (HLA-DRB1*03:01 and *04:01 in type 1 disease in North American and European populations; HLA-DRB1*07 more often in pediatric type 2 disease) are associated with increased risk, though genetic testing does not currently change management.
• Drug- or checkpoint-inhibitor-triggered autoimmune-like hepatitis — nitrofurantoin and minocycline are the most consistently implicated conventional drugs; immune checkpoint inhibitors used in oncology can trigger a clinically and histologically similar immune-mediated hepatitis. These phenotypes often improve after stopping the causative agent, sometimes without needing the years-long immunosuppression that de novo AIH requires — see Special Populations, and the Acute Liver Failure chapter for a fuller drug-induced liver injury differential not repeated here.
• Viral infection has long been postulated as a possible trigger in a genetically susceptible person, though no single virus has been established as causal.
• Unlike alcohol-associated or metabolic liver disease, AIH is not driven by alcohol quantity, obesity, or metabolic syndrome — its risk factors are immunologic and genetic rather than exposure-based.

Core laboratory pattern

Aminotransferases (ALT, AST) are elevated, sometimes markedly so, and unlike alcoholic hepatitis there is no characteristic AST:ALT ratio pattern to lean on — ALT is often the more elevated of the two, as in most other non-alcohol-related hepatitis. Serum IgG (or total gamma-globulin where IgG is not directly measured) is elevated in the great majority of patients and, together with autoantibodies, is one of the core diagnostic and remission-monitoring markers used throughout this chapter.^[1]

Autoantibody profile and the two recognized subtypes

A positive anti-soluble liver antigen/liver-pancreas antibody (anti-SLA/LP), found in a minority of type 1 patients, is highly specific for AIH and is associated with a more severe, relapsing course requiring closer long-term follow-up.^[2] Antimitochondrial antibody (AMA) should also be checked at baseline specifically to screen for primary biliary cholangitis overlap (see Special Populations); a low-titer ANA or ASMA alone, without the rest of the diagnostic picture, is not sufficient to diagnose AIH, since these autoantibodies are also found at low titer in the general population and in other liver diseases.^[1]

Liver biopsy

Liver biopsy remains an important, active part of confirming the diagnosis, not merely staging fibrosis — the histologic hallmark is interface hepatitis: a dense lymphoplasmacytic infiltrate crossing the limiting plate into the periportal parenchyma, often with hepatocyte "rosette" formation and, in some cases, emperipolesis. Biopsy also stages fibrosis at diagnosis (a meaningful minority already have bridging fibrosis or cirrhosis) and is the tool used later to help confirm histologic remission before attempting drug withdrawal (see Choose Initial Treatment and Monitoring & Follow-Up).^[1][2]

Simplified diagnostic criteria

The International Autoimmune Hepatitis Group's simplified scoring system combines four readily available elements into a point-based diagnostic aid; it does not replace clinical judgment but is a useful teaching and documentation tool.^[3]

A total score of 6 or more indicates "probable" AIH; a score of 7 or more indicates "definite" AIH.^[3]

Differential diagnosis

• Viral hepatitis (A, B, C, and, where relevant, E) — check serologies in every new presentation before attributing hepatitis to AIH.
• Drug-induced liver injury, including checkpoint-inhibitor hepatitis — a careful, specific medication/supplement history covering the preceding weeks to months is essential (see the Acute Liver Failure chapter for a fuller DILI differential, not repeated here).
• Wilson disease — particularly in patients under 40; a low alkaline phosphatase relative to bilirubin and a Coombs-negative hemolytic anemia are supportive clues (see the Acute Liver Failure chapter).
• MASLD/MASH and alcohol-associated liver disease — can coexist with AIH and confound the biochemical picture; consider especially when the response to immunosuppression is only partial.
• Primary biliary cholangitis and primary sclerosing cholangitis, alone or as an overlap syndrome with AIH (see Special Populations).

Baseline evaluation before starting treatment

• CBC, comprehensive metabolic panel, and INR at baseline, before the first dose of any therapy.
• TPMT (and NUDT15, where available) activity or genotype testing before starting azathioprine, where accessible (see Choose Initial Treatment and Drug Tables).^[4]
• Bone density (DEXA) at baseline if a substantial corticosteroid course is anticipated, particularly with other osteoporosis risk factors present.
• Baseline glucose/HbA1c, given the hyperglycemic effect of induction corticosteroid therapy.
• Pregnancy test in any patient of childbearing potential before starting or changing therapy.

• Achieve and confirm biochemical remission — normalization of both aminotransferases and IgG, not aminotransferases alone.
• Work toward histologic remission (resolution of interface hepatitis) as the deeper treatment goal, since biochemical normalization can occur while histologic activity persists.
• Minimize cumulative corticosteroid exposure by transitioning to steroid-sparing azathioprine (or an alternative) as early as clinically appropriate.
• Recognize and treat an acute severe presentation promptly enough to avoid progression to frank acute liver failure and transplantation.
• Only attempt drug withdrawal after genuinely sustained remission, understanding and counselling on the high relapse risk.
• Prevent progression to cirrhosis, and manage established cirrhosis and its complications per the Cirrhosis chapter once it develops.
• Protect bone health and monitor metabolic effects throughout corticosteroid therapy.
• Maintain the patient's engagement with regular monitoring — AIH is frequently asymptomatic even during a flare, so lab-based surveillance carries real diagnostic weight, not just confirmatory weight.

• Bone health — calcium and vitamin D supplementation and weight-bearing exercise for the duration of meaningful corticosteroid exposure; baseline and periodic bone density (DEXA) monitoring given the cumulative steroid dose over what is often a multi-year course.
• Glycemic and metabolic monitoring during induction corticosteroid therapy — proactive rather than reactive glucose management, the same principle applied in the Alcoholic Hepatitis chapter.
• Vaccination review: hepatitis A and B (if non-immune) and standard age-based adult vaccines; avoid live vaccines during significant immunosuppression (high-dose corticosteroid, azathioprine, mycophenolate mofetil, or a calcineurin inhibitor) and resume once the regimen is tapered to a level generally considered non-significantly-immunosuppressive.
• Avoid unregulated herbal and dietary supplement products and uncapped acetaminophen use — the same standing caution as every other chapter in this liver-disease group, given an already-inflamed liver.
• Alcohol moderation/avoidance — AIH is not alcohol-driven, but ongoing heavy alcohol use adds an independent source of hepatic injury on top of the autoimmune process.
• Nutrition, and hepatocellular carcinoma/variceal surveillance once cirrhosis has developed, follow the Cirrhosis chapter in full and are not repeated here.
• Patient education on relapse recognition is a genuinely high-value counselling point, since AIH flares are frequently silent until relatively advanced — reinforce that stopping medication early or missing scheduled follow-up labs because "I feel fine" is one of the most common preventable causes of relapse.
• Pregnancy planning counselling for anyone of childbearing potential on treatment — discussed in full in Special Populations.

Pregnancy

Azathioprine is generally considered compatible with pregnancy — a genuinely useful counselling point, since it stands apart from many other immunosuppressants used in autoimmune disease: current guidance supports continuing it through pregnancy in a patient who needs it for disease control, rather than stopping or switching preemptively.^[1][2] Mycophenolate mofetil, in contrast, is clearly teratogenic and should be stopped and switched (typically to azathioprine) well before conception is attempted, with reliable contraception used while on it. Corticosteroids (prednisone/prednisolone) are also generally continued through pregnancy at the effective dose. Postpartum flares are a recognized risk — monitor liver enzymes closely in the months after delivery rather than assuming the disease will remain quiet simply because pregnancy itself is often associated with relative immunologic quiescence. Breastfeeding is generally considered compatible with prednisone and azathioprine at usual therapeutic doses per current guidance. Manage jointly with maternal-fetal medicine and hepatology.^[2]

Cirrhosis at diagnosis or developing during treatment

Budesonide is contraindicated once cirrhosis is present (see Choose Initial Treatment); prednisone or prednisolone combined with azathioprine remains the induction approach. Manage cirrhosis complications, HCC/variceal surveillance, and transplant-referral triggers per the Cirrhosis chapter, not repeated here. Azathioprine-related cytopenias can be harder to distinguish from portal-hypertension-related cytopenias (splenic sequestration) in this population — a falling platelet count alone does not necessarily mean azathioprine toxicity, and both possibilities need to be considered.

Overlap syndromes

AIH-primary biliary cholangitis overlap — add ursodeoxycholic acid (ursodiol) under specialist direction, alongside AIH-directed immunosuppression. AIH-primary sclerosing cholangitis overlap — generally managed primarily as PSC with adjunctive immunosuppression, under specialist direction. Suspect either when the biochemical or serologic picture is atypical for straightforward AIH — a significant cholestatic pattern, or an unexpectedly positive AMA — see Assessment & Diagnosis.

Drug- and checkpoint-inhibitor-induced autoimmune-like hepatitis

Distinguish from de novo AIH where possible — stopping the causative drug is the first step, and many of these cases improve without needing the years-long immunosuppression that established AIH requires. Involve oncology and hepatology jointly when a checkpoint inhibitor is implicated, since resuming cancer therapy is a genuine competing consideration not present in de novo AIH.

Older adults

AIH in older adults can present, and often responds, similarly to younger patients, but corticosteroid-related risks (osteoporosis, diabetes, infection, falls) carry disproportionate consequences — monitor bone health and glucose more closely and favour an earlier transition to steroid-sparing therapy where tolerated.

• ALT, AST, and bilirubin every 1-2 weeks during induction; space to monthly once responding, then every 3-6 months once in confirmed sustained remission.
• IgG alongside every significant liver-enzyme check — normalization of IgG, not aminotransferases alone, is required to call biochemical remission (see Treatment Targets).
• CBC regularly on azathioprine — commonly weekly to every 2 weeks for the first 4-8 weeks, then spacing to every 1-3 months once a stable dose is established; more frequent monitoring if TPMT/NUDT15 status is unknown or indicates reduced activity.^[4]
• Bone density (DEXA) periodically during a prolonged corticosteroid course.
• Glucose monitoring during induction-dose corticosteroid therapy.
• Consider repeat liver biopsy specifically before attempting drug withdrawal, to confirm histologic (not just biochemical) remission (see Choose Initial Treatment).
• HCC surveillance (ultrasound ± AFP every 6 months) and variceal screening once cirrhosis has developed, per the Cirrhosis chapter, continued indefinitely regardless of how well-controlled the AIH itself appears.
• Close monitoring — frequent liver enzymes — for at least the first 12 months after any significant dose reduction or a drug-withdrawal attempt, since most relapses happen early.

1. Incomplete biochemical response after roughly 3-6 months of adequately dosed first-line therapy — confirm adherence first (a common and easily missed cause of apparent non-response, since AIH is often asymptomatic), then confirm azathioprine has actually reached its target weight-based dose before concluding true non-response.
2. Confirmed non-response despite adherent, adequately dosed prednisone plus azathioprine — switch to or add mycophenolate mofetil.^[2]
3. Azathioprine intolerance (significant GI upset, hepatotoxicity, pancreatitis, or cytopenias not attributable to another cause) — switch to mycophenolate mofetil rather than pushing through an intolerable regimen.
4. Refractory to both azathioprine and mycophenolate mofetil — refer to hepatology for a calcineurin inhibitor (tacrolimus or cyclosporine) or another specialist-directed option; this is not an ambulatory-initiated decision.
5. Relapse after a drug-withdrawal attempt — restart the previously effective induction regimen; most patients respond again, but a relapse after withdrawal generally establishes the need for indefinite maintenance therapy rather than a near-term repeat withdrawal attempt.^[1]
6. Progression to decompensated cirrhosis despite treatment — refer for transplant evaluation using the Cirrhosis chapter's triggers (a first decompensation event, or MELD ≥15), not a fixed AIH-specific threshold.
7. Deteriorating acute severe presentation despite prompt high-dose corticosteroids — do not wait for a clearly failed steroid trial; treat as evolving acute liver failure and contact the transplant centre in parallel with continuing medical therapy (see the Acute Liver Failure chapter).

• Diagnosing AIH from a positive autoantibody or an elevated IgG alone, without biopsy confirmation and a reasonable exclusion of other causes — autoantibodies are also found, often at low titer, in other liver diseases and in the general population.^[1]
• Choosing budesonide for a patient who already has cirrhosis — portosystemic shunting removes its first-pass-metabolism advantage and produces unpredictable systemic corticosteroid exposure.^[1]
• Starting azathioprine without any consideration of TPMT/NUDT15 status where testing is accessible, then discovering severe myelosuppression at what should have been a standard dose.^[4]
• Treating a single normal set of liver enzymes as "cured" and stopping therapy without confirming sustained (and ideally histologic) remission over a genuinely prolonged period.^[2]
• Abruptly stopping corticosteroids or azathioprine — for a mild side effect, a missed refill, or the patient's own decision once feeling well — rather than tapering and monitoring; abrupt discontinuation risks both adrenal insufficiency (corticosteroids) and a rebound flare.
• Missing an evolving acute severe presentation by treating new jaundice and malaise as a routine flare rather than actively checking INR and mental status for signs approaching acute liver failure.
• Prescribing or continuing mycophenolate mofetil in a patient who is pregnant or actively planning conception, rather than using azathioprine, which is generally considered the compatible option.
• Not checking AMA or otherwise screening for an overlap syndrome when the biochemical or serologic picture is atypical for straightforward AIH.
• Discontinuing HCC/variceal surveillance once cirrhosis has developed just because the AIH itself is now biochemically quiet — surveillance is time-based, not activity-based, the same point made in the Cirrhosis chapter.

How to use this table: Doses are adult doses and require patient-specific confirmation (renal/hepatic function, weight, allergy history). Pediatric dosing is weight-based per the current product monograph unless stated. Drug pricing is intentionally omitted. Select any drug name to open its complete Pharmasist Drug Database page.

Corticosteroids — induction

Combine with azathioprine from the outset (or within about 2 weeks) in most patients able to tolerate it, to reduce cumulative steroid dose.

Budesonide — non-cirrhotic induction alternative

Azathioprine — steroid-sparing induction and maintenance

Second- and third-line steroid-sparing options

Reserved for azathioprine intolerance or non-response; calcineurin inhibitors require specialist direction and therapeutic drug monitoring.

Overlap syndrome — adjunctive therapy

Medications and exposures requiring caution