Autoimmune hepatitis (AIH) is an inflammatory lesion of the liver parenchyma mediated by an abnormal autoimmune response, mostly in women, characterized by hypergammaglobulinemia, positive serum autoantibodies and response to immunosuppressive therapy. Drug-induced liver injury (DILI) is a drug-related liver injury that also occurs in women, sometimes with positive autoantibodies and in severe cases requiring immunosuppressive therapy. The relationship between the two and how to identify and treat them appropriately at the time of initial diagnosis is an issue that clinicians need to pay attention to. I. Relationship between AIH and DILI In clinical work, there is a phenomenon that AIH-like manifestations can occur in patients with DILI who recover from DILI and then use the same or similar drugs again. The structure of the drugs used before and after the two liver injuries was similar, and the target of action was similar in two cases. In addition, 4 patients (4/9) were diagnosed or possibly diagnosed with AIH according to the traditional and simplified AIH diagnostic score criteria.In general, the diagnosis of AIH requires the exclusion of a history of drug use, whereas the opposite is true for DILI, and in practice it is often difficult to determine the cause-and-effect relationship between drugs and liver injury, coupled with the fact that drugs, in addition to their own direct adverse effects, can also cause liver injury through immune-mediated inflammatory responses The relationship between AIH and DILI is complicated by factors such as liver injury. Recently, Weiler-Normann and Schramm divided the relationship between AIH and DILI into three categories: (1) immune-mediated DILI: a drug-induced liver injury disease with AIH manifestations (e.g., autoantibody positivity, immunoglobulin elevation, interfacial hepatitis, etc.), which Czaja called drug-mediated autoimmune hepatitis, and not really AIH. (2) Drug-mediated AIH: i.e., patients who do not have or have mild AIH themselves and are diagnosed after the use of certain drugs that induce or manifest significantly. (3) AIH with DILI: Patients with clearly diagnosed AIH who have concurrent DILI and often show fibrosis on liver histology. In the author’s opinion, essentially, the disease attacks in patients with immune-mediated DILI depend on the continuous use of drugs, i.e., they can manifest as disease attacks or even have certain clinical features of AIH when the drugs are used, and the disease usually resolves on its own once the drugs are discontinued, therefore, they still belong to the category of drug-induced liver disease. On the contrary, the course of drug-induced AIH is no longer dependent on the use of drugs, but the patient has a chronic disease course, and the use of drugs is only a trigger, in line with the immunological principle of “Hit and Rlm”. The diagnosis of drug-induced AIH is often difficult because of the long time lapse between drug use and disease onset and the difficulty of establishing causality. Recently, Bjomsson et al. reported 24 cases of drug-induced AIH, 11 of which were caused by dimethylaminetetracycline and the other 11 by furantoin, in which the patients were still using the relevant drugs at the time of diagnosis, and the subsequent hormonal treatment was effective, with no relapse after discontinuation of the drugs. In the author’s opinion, these patients were more appropriately diagnosed as immune-mediated DILI rather than drug-induced AIH in the true sense. Reasons for confusion between AIH and DILI The reasons why DILI and AIH are easily confused are similar clinical manifestations, poor specificity of autoantibodies, and difficulty in determining the causal relationship between drugs and identification of liver injury. AIH is usually a chronic liver disease, while DILI often presents acutely. However, about 10% of AIH can have an acute onset, and very few cases even show a fulminant course; similarly, chronic DILI can be seen in patients who have been taking certain drugs (e.g., anti-tuberculosis drugs, chemotherapeutic oncology drugs, proprietary Chinese medicines, etc.) for a long time. AIH and some DILI (especially immune-mediated DILI) are hepatocellular, with AST and ALT exceeding 5-20 times the upper limit of normal (ULN), while cholestasis indicators such as alkaline phosphatase are not significantly elevated, and immunoglobulin G (IgG) may also be elevated. Autoantibodies are important indicators for the diagnosis of AIH, including antinuclear antibodies (ANA), anti-smooth muscle antibodies (SMA) and antibodies to liver kidney microsome type 1 (anti-smooth muscle antibodies). ANA is a widespread autoantibody found in type I AIH, while SMA, whose main target antigen is F-actin, is often found in type I AIH together with ANA, but the specificity of both is poor. Anti-LKM1 is the hallmark antibody of type II AIH. the most common antibodies in DILI are ANA and SMA, which are associated with the administration of dimethylamine-tetracycline or furantoin, which can easily cause confusion with type I AIH, and anti-LKM2, which is associated with tinidic acid. Dimethylaminetetracycline is an antibiotic used to treat acne, and liver injury usually occurs within 2 years of dosing, with elevated ANA and IgG. Furantoin is a drug used to treat urinary tract infections, and acute liver injury usually occurs within 6 weeks of dosing. Tinidate is a diuretic used to treat hypertension and has potential hepatotoxicity. Other drugs that cause autoantibody positivity are dihydropyridazine, trifluorochloroethane, tumor necrosis factor alpha and interferon beta. The diagnosis of AIH requires the exclusion of drug factors, whereas the diagnosis of DILI requires drugs as the etiology. the problem is that it is extremely difficult to clarify the causal relationship between drugs and the onset of the disease. the latency period (i.e., the time after drug treatment when liver damage occurs) in patients with DILI may be 1 to 8 weeks or up to 12 months. Attempting to reappear with the same drug with corresponding symptoms is the most reliable basis for diagnosis, but doing so may be fatal for the patient. There are currently two methods for determining the causal relationship between drugs and liver injury: the Roussel Uclaf causality rating method and clinical diagnostic criteria. The former includes seven factors: drug latency, disease course characteristics, risk factors (age, alcohol consumption, pregnancy), concomitant medication, non-drug factors, known drug hepatotoxicity, and re-drug response, while the latter includes only five factors: latency, exclusion of other causes, extra-hepatic immune response, intentional or unintentional re-dosing, and known drug hepatotoxicity, with the main difference between the two being extra-hepatic immune manifestations (fever, arthralgia, erythema, eosinophilia). Both methods are too complex to be promoted in clinical practice, while clinicians’ empirical judgment is the best way to make the diagnosis. A thorough medical history is a prerequisite for a clear diagnosis. History taking should include the time of occurrence and number of liver function abnormalities, whether the liver function abnormalities have completely returned to normal between episodes, the course of medications including herbal medicines, supplements and nutritional products, and the relationship between liver function abnormalities and medication use. III. Histological identification of AIH and DILI AIH and DILI have certain histological similarities, including interfacial hepatitis, infiltration of plasma cells, lymphocytes and eosinophils in the portal area and necrosis in the central 3 bands of the lobules. However, the results of Suzuki et al. showed that there are still differences in liver histology between the two. They randomly and double-blindly assigned 35 cases of clinically definite diagnosis of DILI (19 hepatocellular and 16 biliary or mixed) and 28 cases of AIH to 4 liver pathologists for liver biopsy based on Ishak score, type of inflammatory cells in the hilar region and lobules, presence of The results showed that interfacial hepatitis, focal necrosis and inflammation in the portal area were both present, but AIH was more severe than DILI, and histological manifestations specific to AIH included plasma cell infiltration, rosettes and invasive phenomena, while diffuse infiltration of neutrophils in the confluent area and intrahepatic cholestasis were mostly seen in DILI; in addition, there were 7 cases of immune In addition, there were seven cases of immune-mediated DILI in the study, and there was no significant liver fibrosis in them compared with AIH. However, some authors are skeptical because of the low concordance rate (46%) among the four pathologists, the low percentage of DILI patients who underwent liver biopsy, and the fact that the acute phase of liver biopsy had already passed by the time of the biopsy, resulting in poor understanding of the histologic features by the pathologists. In general, eosinophil infiltration is more common in DILI, but some studies have shown that eosinophil infiltration cannot be used to differentiate AIH from DILI. Treatment and prognosis Classical AIH treatment includes hormonal and immunosuppressive therapy. 2010 US guidelines for the management of AIH recommend prednisone (starting at 30 mg/d and tapering to 10 mg/d over 4 weeks) Combined with azathioprine (50 mg/d) or high-dose prednisone alone (start at 40-60 mg/d and taper to 20 mg/d over 4 weeks), the combination regimen should be recommended in preference to a total course of at least 2 years. The principles of DILI treatment are to discontinue the suspected drug in question, to promote drug metabolism in the body, and to monitor changes in the disease. In general, the majority of DILI patients gradually recover liver function within 1 to 3 months after appropriate treatment. Hormones may be used in severe cases or when immune features (e.g., immune-related DILI) cannot be distinguished from AIH. In a study by Bjomsson et al. on 24 cases of immune-related DILI (mainly caused by dimethylaminetetracycline or furantoin) and AIH, hormones showed a remission effect on both, with the difference being The former had no recurrence at a mean follow-up of 3 years after discontinuation of the drug, while the latter had 65% recurrence. Hormone therapy can be used as a way to differentiate between the two, as the initial dosage of hormones in DILI patients is smaller and their onset of action is faster compared to AIH. In addition, the different manifestations after hormone discontinuation are also helpful to differentiate the two, as AIH will relapse after hormone discontinuation while DILI will not. Two concepts need to be clarified here: (1) remission refers to the subsidence of symptoms, normalization of serum transaminase, bilirubin and IgG levels, and improvement of liver histology; (2) relapse is the increase of serum transaminase levels to more than 3 times ULN after discontinuation, which mostly occurs within 15-20 months of discontinuation of immunosuppressive drugs. Discontinuation of the drug is required after improvement of clinical symptoms, biochemical parameters and liver histology, especially liver histology. Since liver histology improves later than clinical symptoms and biochemistry, discontinuation based on biochemistry alone may lead to disease progression due to residual inflammation in liver histology, which is often confused with relapse and thus compromises the judgment, the most appropriate approach is to confirm with liver biopsy before discontinuation. The prognosis of AIH is related to the degree of liver inflammation and the presence or absence of cirrhosis. Patients with serum transaminases higher than 5-10 times ULN, with Y-globulin levels higher than 2 times ULN, with histological bridging necrosis or multilobular necrosis, and with cirrhosis before and after treatment have a poor prognosis, and response to treatment is also an important factor in prognosis.