Autoimmune and Cholestatic Liver Disease: Immune Liver Disease: Autoimmune Hepatitis

0:00Welcome to Board Pearls. This is episode one of three of the Autoimmune and Cholestatic Liver Disease chapter, in the Liver Disease module. This episode is autoimmune hepatitis, from diagnosis to the question of stopping treatment. A middle-aged woman with a hepatocellular pattern and autoantibodies shows how the three immune liver diseases differ at the first visit, what causes autoimmune hepatitis, why a mild-looking case can already be cirrhotic, what separates type one from type two, why the simplified score usually needs a biopsy, and why the drug history comes before the diagnosis. A man with fatty liver and a weak antibody shows a positive test that isn't the disease. A young man starting treatment shows the induction regimen, what to check before azathioprine, when budesonide is wrong, what counts as remission, and why a normal alanine aminotransferase, ALT, isn't a reason to stop. Then three more patients show azathioprine failure, acute severe disease, and pregnancy.

0:57Here's the first patient. A forty-seven-year-old woman has fatigue and joint aches. Her ALT is four hundred eighty, her aspartate aminotransferase, AST, three hundred ninety, her alkaline phosphatase normal, and her bilirubin one point six. Her antinuclear antibody, ANA, is positive at one to one hundred sixty, her smooth muscle antibody at one to eighty, and her immunoglobulin G, IgG, is twenty-four hundred, well over one point one times the upper limit. Her viral tests are negative, and she takes no new drugs or supplements.

1:33A hepatocellular pattern with ANA, smooth muscle antibody, and a high IgG in a middle-aged woman suggests autoimmune hepatitis. But it's one of three immune liver diseases, managed very differently, so it's placed among them first. All three come from a loss of tolerance to a self-antigen, and they differ in the cell attacked. Autoimmune hepatitis attacks hepatocytes. Primary biliary cholangitis attacks the small bile duct cells inside the lobule. Primary sclerosing cholangitis attacks the larger bile ducts, inside and outside the liver. Three kinds of information separate them: the biochemical pattern, the demographics, and the antibodies.

2:13The pattern follows the target. Autoimmune hepatitis gives a hepatocellular pattern, with a high ALT and AST and a relatively preserved alkaline phosphatase. Primary biliary cholangitis gives a cholestatic pattern, with a high alkaline phosphatase and aminotransferases usually under five times normal. Primary sclerosing cholangitis is also cholestatic, with the alkaline phosphatase the dominant abnormality, but that value fluctuates and is normal in forty to fifty percent of patients at any given time, so a normal value doesn't exclude it. Mixed patterns are common in sclerosing cholangitis and in the overlap syndromes, so the labs alone aren't enough. Hers is hepatocellular: an ALT near five hundred with a normal alkaline phosphatase.

3:01The demographics differ as sharply. Autoimmune hepatitis affects about four women for every man, with two age peaks, the late teens to early twenties and the forties and fifties. Primary biliary cholangitis is even more female, nine or ten to one, with a mean age at diagnosis of forty to fifty. Primary sclerosing cholangitis is the reverse. About sixty-five to seventy-five percent of patients are men, with a mean age of forty, and seventy to eighty percent have inflammatory bowel disease, mostly ulcerative colitis, though only two to five percent of patients with ulcerative colitis develop it. Chapter fourteen covers that colitis. She's a woman in her forties, in the second peak.

3:43The antibodies complete the distinction. Autoimmune hepatitis is the disease of ANA and smooth muscle antibody, with anti-liver-kidney microsomal antibody type one, anti-LKM-1, and anti-liver cytosol antibody type one, anti-LC1, marking the rarer type two. Primary biliary cholangitis is the disease of the anti-mitochondrial antibody, AMA, which is positive in about ninety-five percent, with the M2 subtype the most specific. Primary sclerosing cholangitis has no diagnostic antibody. Atypical perinuclear antineutrophil cytoplasmic antibody, p-ANCA, is positive in seventy to eighty percent but appears across inflammatory bowel disease, autoimmune hepatitis, and other autoimmune diseases, so sclerosing cholangitis is an imaging diagnosis.

4:32The immunoglobulins add one clue each. A polyclonal rise in IgG goes with autoimmune hepatitis and is built into its diagnostic score, a polyclonal rise in immunoglobulin M, IgM, goes with primary biliary cholangitis, and a high immunoglobulin G4, IgG4, points toward IgG4-related cholangitis, covered in episode three.

4:57Put together, the three pieces usually name the disease at the first visit. A middle-aged woman with a hepatocellular pattern, ANA and smooth muscle antibody, and a high IgG has autoimmune hepatitis until proven otherwise, and the next step is biopsy and treatment. A woman with a dominant alkaline phosphatase, fatigue, itching, and a positive AMA has primary biliary cholangitis and starts ursodeoxycholic acid without a biopsy, which is episode two. A man with ulcerative colitis and a cholestatic pattern needs a magnetic resonance cholangiopancreatography, MRCP, which is episode three.

5:33Her disease comes from a T-cell attack, which is why steroids work. Activated T cells attack hepatocytes in response to self-antigens that have escaped tolerance. The autoantibodies are made by B cells, but they're serologic markers, not the cause of the damage. The disease is genetic at base, with the human leukocyte antigen types HLA-DR3 and HLA-DR4 conferring susceptibility in adults, and HLA-DR7 in some children with type two disease, and a viral, drug, or environmental trigger starts the chronic process in that primed host.

6:09How it presents explains why it's considered early. It can present acutely, with aminotransferases of about five hundred to one thousand and variable jaundice, or chronically, as an incidental aminotransferase rise in someone who feels well, or as cirrhosis of unknown cause found late. About twenty-five to thirty percent already have cirrhosis at diagnosis, so a modest ALT doesn't mean early disease, and autoimmune hepatitis is considered even when the labs look mild.

6:37Her antibodies set her type. Type one is the classic adult disease, marked by ANA, smooth muscle antibody, and less often anti-soluble liver antigen, with about four women to one man, and overlap with primary biliary cholangitis and sclerosing cholangitis is documented in type one. Type two is less common in North America and occurs mainly in children and young adults, marked by anti-LKM-1 and anti-LC1, with about ten women to one man. It runs more aggressively, is more often linked to autoimmune polyglandular syndrome, and is more likely to present as acute liver failure, so a child with acute liver failure and anti-LKM-1 has type two until proven otherwise. She has ANA and smooth muscle antibody and she's an adult, so she has type one.

7:28A negative antibody panel doesn't end the question. About ten to thirty-five percent of patients are seronegative for the standard antibodies, and in them the diagnosis rests on histology, a high IgG, and exclusion of competing causes. So a hepatocellular pattern with a high IgG still calls for a biopsy when the ANA and smooth muscle antibody are negative.

7:50How certain her diagnosis is comes from the simplified scoring system of the international autoimmune hepatitis group. Autoantibodies score one point for ANA or smooth muscle antibody at one to forty or higher, and two points at one to eighty or higher, or for anti-LKM-1 at one to forty or higher, or for a positive anti-soluble liver antigen. IgG scores one point above the upper limit of normal and two points above one point one times the upper limit. Histology scores one point if it's compatible and two if it's typical. The absence of viral hepatitis scores two points. A total of six is probable autoimmune hepatitis, and seven or more is definite.

8:33Before her biopsy, her antibodies at one to eighty and higher score two, her IgG over one point one times the upper limit scores two, and the absence of viral hepatitis scores two, for a total of six, probable only. In one retrospective cohort, twenty-three percent of patients with confirmed autoimmune hepatitis scored six or less without histology, so stopping at serology leaves nearly a quarter of real cases short of a definite diagnosis. That's why a biopsy is recommended in most suspected cases when it's safe and feasible.

9:06Her biopsy completes the score. The hallmark is interface hepatitis: the inflammatory infiltrate breaches the limiting plate where the portal tract meets the lobule and produces piecemeal necrosis of hepatocytes. The infiltrate is rich in plasma cells in about two-thirds of cases, with lobular hepatitis in nearly half. Two ancillary features add specificity: emperipolesis, one live cell engulfing another, in about sixty-five percent, and hepatocyte rosettes, clusters of reactive hepatocytes ringed by inflammatory cells, in about a third. Her biopsy shows interface hepatitis with plasma cells, emperipolesis, and rosettes, which is typical histology worth two more points, for a score of eight: definite type one autoimmune hepatitis.

9:50Before she's committed to years of treatment, her drug history is checked again, because the microscope can't settle it. Eosinophils may appear on the biopsy, but they don't by themselves point to drug injury, because autoimmune hepatitis and drug-induced autoimmune-like injury look the same under the microscope. Nitrofurantoin, minocycline, methyldopa, hydralazine, atorvastatin, and the checkpoint inhibitors ipilimumab, nivolumab, and pembrolizumab can all produce that picture. Drug-induced injury remits when the drug is withdrawn, with a steroid taper, and doesn't need long-term immunosuppression. In a patient with a plausible drug trigger, a taper without a flare confirms drug injury, and a flare on tapering points to idiopathic disease and long-term maintenance. She takes no implicated drugs, so her disease is idiopathic.

10:39The second patient shows a positive antibody that isn't the disease. A fifty-four-year-old man with obesity, diabetes, and a fatty liver on ultrasound has an ALT of seventy and a smooth muscle antibody positive at one to forty. His antibody scores one point and his normal IgG scores nothing. A low-titer smooth muscle antibody is common in fatty liver disease, and his whole picture fits fatty liver disease, covered in chapter twenty-one. Steroids would be the mistake.

11:09Once the diagnosis is definite, treatment begins, and the third patient shows how it's chosen. A twenty-three-year-old man has newly diagnosed type one autoimmune hepatitis, confirmed by biopsy, with an ALT of six hundred and a high IgG. Neither his biopsy nor his elastography shows cirrhosis. His bilirubin is one point two and his international normalized ratio, INR, is normal. He's worried about steroids, because a friend gained weight and got acne on them.

11:39His treatment has two phases, a steroid to induce remission and a steroid-sparing drug to maintain it. In adults without cirrhosis, induction is prednisone thirty to sixty milligrams a day, with some protocols opening at twenty to forty, and azathioprine is added at fifty to one hundred fifty milligrams a day. The steroid controls the inflammation quickly, and azathioprine lets the steroid come down and stay down for years.

12:05Before his azathioprine starts, one enzyme is checked: thiopurine methyltransferase, TPMT. Patients deficient in TPMT metabolize azathioprine into toxic levels of six-thioguanine and develop severe bone marrow suppression, so the TPMT result comes before the first dose, not after the first blood count. With deficiency, standard-dose azathioprine isn't started. His TPMT is normal.

12:34His worry about steroids can be met, because without cirrhosis he can avoid prednisone. Budesonide, nine milligrams a day, is an alternative to prednisone for induction in patients without cirrhosis. The liver clears about ninety percent of it on the first pass, so less reaches the rest of the body and the systemic side effects are lower. He starts budesonide with azathioprine.

12:56That choice depended on his staging, because budesonide must not be used in cirrhosis. Portosystemic shunting lets blood bypass the liver, so the first-pass effect is lost: the drug works less well in the liver and more of it reaches the systemic circulation. With cirrhosis on elastography, he'd get prednisone with azathioprine regardless of his worries. Because a quarter to a third of patients are cirrhotic at diagnosis, the liver is staged before the steroid is chosen.

13:23His response is followed in the blood and later in the tissue. A biochemical response usually appears at four to eight weeks. On a prednisone regimen, the steroid is then tapered over about six months to a maintenance dose of five to ten milligrams a day, with azathioprine taking over maintenance. The goal is remission in two parts: biochemical remission, meaning a normal ALT, AST, and IgG, and histologic remission, a biopsy without active inflammation, which lags the biochemical remission by about eight months. A normal ALT means the blood tests have recovered, not the liver. His ALT and IgG are normal at three months, he feels well, and he asks to stop. The answer is no.

14:09Stopping is where most mistakes happen. The dominant error in autoimmune hepatitis is premature withdrawal. Sixty to eighty-eight percent of patients relapse within five years of stopping immunosuppression, and recurrent relapse drives cirrhosis and liver-related death. The standard is biochemical remission for at least twenty-four months before withdrawal is considered, and many experts repeat the biopsy first, because patients with normal histology relapse less often than those with persistent portal inflammation.

14:39At two and a half years he's still in biochemical remission, but his repeat biopsy shows persistent portal inflammation, so he stays on azathioprine. With a normal biopsy, withdrawal would have been reasonable, with close monitoring of his ALT and IgG and a plan made with the relapse rate in mind. For a man of twenty-three, it's repeated relapse, not years of low-dose azathioprine, that leads to cirrhosis.

15:04Some patients can't take azathioprine or don't respond to it. Mycophenolate mofetil is the preferred second-line agent for azathioprine intolerance or non-response, and stopping treatment instead would simply be premature withdrawal. Calcineurin inhibitors and rituximab are used in refractory disease, in a center experienced with them. Liver transplant is indicated for fulminant disease that doesn't respond to steroids and for decompensated cirrhosis. After transplant, acute rejection is more common in the first year, and autoimmune hepatitis recurs in the new liver in a minority. Chapter twenty-four covers transplant.

15:40The fourth patient shows that acute severe disease follows different rules. A thirty-eight-year-old woman whose workup points to autoimmune hepatitis is admitted with an ALT of one thousand four hundred, deep jaundice, and a rising INR. She has acute severe autoimmune hepatitis, which changes three things. She gets prednisone sixty milligrams a day or intravenous methylprednisolone, she gets no azathioprine in the acute phase, and she gets no budesonide. Her response is assessed every seven to fourteen days, and without a response in that window the next step is transplant evaluation, not a higher steroid dose, which only delays the referral. She's referred to the transplant center on day one so the evaluation is ready.

16:25Her disease can cross into liver failure. Fulminant autoimmune hepatitis is one of three exceptions that count as acute liver failure despite underlying chronic disease, so if she becomes encephalopathic with an INR over one and a half, the acute liver failure care from chapter nineteen applies, including early transfer. It's one of the few kinds of liver failure in which steroids belong, because in drug-induced liver failure they don't improve survival and may raise the risk of infection. Her bilirubin and INR begin to fall in the second week, and azathioprine is added once she's out of the acute phase.

17:00The last patient shows pregnancy, the other setting where patients want to stop. A twenty-nine-year-old woman with autoimmune hepatitis in remission on low-dose prednisone and azathioprine becomes pregnant and asks whether to stop both drugs to protect the baby. She continues them, because the risk of a flare on withdrawal is greater than the risk these drugs pose to the fetus. Flares after delivery occur in up to half of patients, so her liver tests are followed closely in the months after she gives birth. Chapter thirty-five covers pregnancy.

17:31Here are the traps the boards set in these patients. Negative serology doesn't rule out autoimmune hepatitis. Serology and IgG alone don't make it definite without a biopsy. A low-titer smooth muscle antibody in fatty liver disease isn't autoimmune hepatitis. Jaundice isn't needed before it's considered, because a quarter to a third already have cirrhosis. A child with anti-LKM-1 and liver failure has type two. Eosinophils on biopsy don't diagnose drug injury. And a normal alkaline phosphatase doesn't rule out sclerosing cholangitis, which no antibody can diagnose.

18:08The treatment traps follow. Azathioprine isn't started without checking TPMT. Budesonide isn't used in cirrhosis. Azathioprine isn't added in acute severe disease. When there's no response in one to two weeks, the answer is transplant referral, not more steroid. Treatment doesn't stop when the ALT normalizes, or before two years of remission, and a repeat biopsy often comes before withdrawal. Immunosuppression isn't stopped in pregnancy, and the flare after delivery is watched for. And a patient with drug-induced autoimmune-like injury isn't committed to lifelong treatment.

18:44So, start to finish. Separate the three immune liver diseases by pattern, demographics, and antibodies. A hepatocellular pattern with a high IgG and ANA or smooth muscle antibody is autoimmune hepatitis. A cholestatic pattern with AMA and a high IgM is primary biliary cholangitis. A cholestatic pattern in a man with ulcerative colitis is primary sclerosing cholangitis, an imaging diagnosis.

19:12Confirm autoimmune hepatitis with serology, IgG, and a biopsy showing interface hepatitis with plasma cells, because many patients score only probable without histology, and some are seronegative. Check the drug history, because drug-induced injury looks the same and remits when the drug stops. A low-titer smooth muscle antibody alone, common in fatty liver disease, isn't autoimmune hepatitis.

19:39Treat with prednisone plus azathioprine after TPMT testing, or with budesonide instead of prednisone only when there's no cirrhosis. Expect a response at four to eight weeks, taper the steroid, and let azathioprine carry the maintenance. In acute severe disease, give high-dose steroids alone and refer for transplant if there's no response in one to two weeks.

20:03Aim for a normal ALT, AST, and IgG, and treat for at least two years before any withdrawal, often guided by a repeat biopsy. Switch to mycophenolate for azathioprine failure, and continue treatment in pregnancy.

20:17Episode two is primary biliary cholangitis: diagnosis without a biopsy, the AMA-negative patient, ursodeoxycholic acid and its response criteria, second-line therapy, and pruritus.

20:31For the full chapter, the practice vignettes, and the topic-tagged question bank, head to board pearls dot com. You'll find the rest of the series on Apple Podcasts, Spotify, or wherever you listen to podcasts. That brings us to the end of episode one of three of chapter twenty, and I'll see you in the next one.