Autoimmune and Cholestatic Liver Disease: Sorting the Immune Liver Diseases and Autoimmune Hepatitis

0:00Welcome to Board Pearls. This is episode one of three of the Autoimmune and Cholestatic Liver Diseases chapter, in the Liver Disease module. This episode is the sorting framework and autoimmune hepatitis: distinguishing the three immune liver diseases by antibody pattern and biliary involvement, autoimmune hepatitis as autoantibody-positive interface hepatitis with its type one and type two split, and treatment with corticosteroids plus a steroid-sparing agent.

0:26The three classical immune liver diseases share one frame and split on one variable. The frame is loss of tolerance to a self-antigen on a particular cell; the variable is which cell the immune system attacks. Autoimmune hepatitis is the attack on hepatocytes, primary biliary cholangitis the attack on the small intralobular bile duct epithelium, and primary sclerosing cholangitis the attack on the larger intrahepatic and extrahepatic ducts. Once the target is named, the biochemistry is predictable: hepatocyte injury gives a hepatocellular pattern with the transaminases elevated and the alkaline phosphatase relatively preserved; small-bile-duct injury gives a cholestatic pattern with the alkaline phosphatase elevated and the transaminases only modestly raised; and large-duct injury gives the same cholestatic pattern with one trap, that in sclerosing cholangitis the alkaline phosphatase fluctuates and is often normal at any single check, so a normal value doesn't exclude it.

1:21The demographics close the second discriminator. Autoimmune hepatitis is female-predominant, roughly four to one, with a bimodal age distribution in the late teens to twenties and again in the forties and fifties. Primary biliary cholangitis is even more female-skewed, nine or ten to one, with a mean age at diagnosis in the forties. Primary sclerosing cholangitis inverts the picture, mostly men, mean diagnosis around forty, and bound to inflammatory bowel disease, since the majority of sclerosing cholangitis patients have concurrent IBD, predominantly ulcerative colitis, though the reciprocal rate is small, only a small percentage of IBD patients developing it. The antibody panel closes the sort: autoimmune hepatitis is the ANA and anti-smooth-muscle disease, with anti-LKM-one and anti-LC-one marking the rarer type two mostly in children; primary biliary cholangitis is the anti-mitochondrial antibody disease, positive in about ninety-five percent of cases with the subtype against pyruvate dehydrogenase the most specific; and primary sclerosing cholangitis has no diagnostic antibody, since the atypical p-ANCA is positive in most patients but appears across many autoimmune diseases, so it's an imaging diagnosis. The IgG profile reinforces the same sort, with polyclonal IgG tracking autoimmune hepatitis and built into its score, polyclonal IgM tracking primary biliary cholangitis, and IgG4 elevation pulling toward IgG4-related cholangitis. So three patients sit in front of you and the reflex pattern is the teaching: a middle-aged woman with a hepatocellular pattern, smooth-muscle antibody, and elevated IgG is autoimmune hepatitis until biopsy proves otherwise; a middle-aged woman with cholestatic biochemistry, fatigue and pruritus, and a positive mitochondrial antibody is primary biliary cholangitis, and you start ursodeoxycholic acid; and a man in his forties with ulcerative colitis and a cholestatic pattern needs an MRCP, where the multifocal stricture pattern with intervening normal-caliber duct closes the diagnosis. The temptation is to anchor on biochemistry alone, but mixed patterns are common, so pattern plus demographics plus antibodies sort the three within the first two sentences of the vignette, and that's the backbone of the chapter.

3:33That brings us to autoimmune hepatitis on its own terms. It's the immune attack on hepatocytes by activated T cells responding to self-antigens that escaped tolerance, with the B-cell-derived autoantibodies serving as markers rather than effectors, genetic at base with HLA susceptibility alleles, and triggered by a virus, drug, or environmental exposure in primed hosts. The presentation has two faces: the acute face is transaminases in the hundreds with variable jaundice, sometimes looking like a viral hepatitis, and the chronic face is an incidental transaminase elevation in an asymptomatic patient or cryptogenic cirrhosis discovered late, with a substantial fraction already cirrhotic at diagnosis, which is why it has to be considered early even when the labs look indolent, since the reflex to delay biopsy and watch the numbers is the one that costs the patient an organ. Two serologic types matter: type one is the classical adult disease marked by ANA, smooth-muscle antibody, and the less common soluble-liver-antigen antibody, female-predominant with documented overlap; type two is less common in North America, mostly in children and young adults, marked by anti-LKM-one and anti-LC-one, even more female-skewed, running more aggressively, presenting with acute liver failure more often, and clustering with the autoimmune polyglandular syndrome, so the young woman with type one diabetes who arrives with two weeks of jaundice and encephalopathy is the type two stem.

4:56The diagnostic framework the boards test is the simplified scoring system, four parameters and a small set of points: autoantibodies score points scaled to titer, IgG scores points scaled to how far above normal it sits, histology scores points for being compatible or typical, and absence of viral hepatitis scores points, with a total of six probable and seven or more definite. The score has a known blind spot, because a meaningful fraction of confirmed patients score six or less without histology, and a meaningful fraction are seronegative for the conventional autoantibodies, so the diagnosis in those patients hinges on histology with elevated IgG and exclusion of competing causes, and the score doesn't replace biopsy, it points to it, and the biopsy settles the question. Histology is the diagnostic anchor, with the hallmark being interface hepatitis, where the inflammatory infiltrate breaches the limiting plate at the portal-lobular interface and produces piecemeal necrosis, the infiltrate plasma-cell-rich in most cases with lobular hepatitis common, and two ancillary features adding specificity, emperipolesis, one cell engulfing another with both surviving, and hepatocyte rosettes, clusters of reactive hepatocytes surrounded by inflammatory cells, with eosinophils sometimes present, which shouldn't push toward drug injury alone because idiopathic and drug-induced autoimmune-like hepatitis are histologically indistinguishable.

6:12Treatment is the second half. The structure is steroid-based induction with steroid-sparing immunosuppression for maintenance. Induction in a non-cirrhotic adult is prednisone at thirty to sixty milligrams daily, with azathioprine added at fifty to a hundred fifty milligrams daily once a TPMT level is checked, and the check matters because TPMT-deficient patients metabolize azathioprine to toxic levels and develop severe myelosuppression, so it happens before the drug, not after the cytopenia. The steroid is tapered over six months to a maintenance of five to ten milligrams once biochemical response is established at a few weeks, and azathioprine takes over maintenance. Budesonide at nine milligrams daily is an alternative induction agent for non-cirrhotic patients because of its high first-pass metabolism, which reduces systemic side effects, but the exclusion is critical: it must not be used in cirrhosis, because the portosystemic shunting bypasses the first-pass effect, lowering efficacy and raising systemic exposure, so cirrhotic patients on budesonide get the worst of both worlds. Acute severe autoimmune hepatitis is a different situation: the hospitalized patient with very high transaminases, jaundice, and a rising INR is treated with high-dose prednisone or intravenous methylprednisolone, without azathioprine and without budesonide in the acute phase, with response assessed every one to two weeks, and failure to respond in that window is a signal to refer for transplant, not to escalate the dose, because the patient who doesn't move on the labs in two weeks of high-dose steroids won't be salvaged by week three.

7:38The endpoint of therapy is biochemical and histologic remission, biochemical being normal transaminases and normal IgG, and histologic remission lagging the biochemical by many months, which is exactly why immunosuppression shouldn't stop the moment the transaminases normalize, since the dominant teaching error is premature withdrawal, because most patients relapse within a few years of stopping, and recurrent relapse drives cirrhosis and mortality. So the standard is to maintain biochemical remission for at least two years before considering withdrawal, with many experts repeating a biopsy first because patients with normal histology relapse less. So the patient who's been in normal range for eight months wants to stop, and the answer is to keep going, while the patient with two years of normal labs and IgG can be considered for withdrawal, ideally after a biopsy confirms histologic remission, and even then a meaningful fraction relapse and need to know they may have to come back to the drug. Drug-induced autoimmune hepatitis is the named exception to lifelong immunosuppression, because nitrofurantoin, minocycline, methyldopa, hydralazine, a statin, and the immune checkpoint inhibitors can produce an autoimmune-like phenotype that remits with drug withdrawal and a steroid taper without long-term therapy, with the histology identical to idiopathic disease, which is why the medication history is the discriminator. A few practical pieces complete it: mycophenolate is the preferred second-line agent for azathioprine intolerance or non-response, with calcineurin inhibitors and rituximab for refractory disease, and the most common reason for early treatment failure is not the drug but the dose, since too-rapid prednisone reduction and non-adherence dominate, so the move is to re-induce and re-engage rather than switch agents. Pregnancy is where the rules look most counterintuitive: both prednisone and azathioprine are continued, because the risk of a flare on withdrawal exceeds the fetal risk, and postpartum flare is common, so the right answer for a stable patient who arrives pregnant is to continue both. And transplantation is the answer in fulminant disease unresponsive to steroids and in decompensated cirrhosis, with acute rejection more common in the first post-transplant year and a minority recurring in the allograft.

9:51So the synthesis is short. Pattern plus demographics plus antibodies sort the three diseases in two sentences. Autoimmune hepatitis then has its own logic, the immune attack on hepatocytes in a middle-aged woman with elevated IgG and ANA or smooth-muscle antibody, confirmed by interface hepatitis on biopsy. Treatment is steroids and azathioprine for induction, maintained on the steroid-sparing agent, with type one the adult disease and type two the pediatric, more aggressive disease with anti-LKM-one. The score points to biopsy, the biopsy settles the diagnosis, maintenance runs at least two years, and withdrawal is cautious because relapse is the rule rather than the exception.

10:31Episode two picks up primary biliary cholangitis, the mitochondrial-antibody-positive cholestatic disease in a middle-aged woman, with ursodeoxycholic acid first-line, the changed second-line landscape after obeticholic acid was withdrawn from the US market leaving the two newer agents, and the pruritus management sequence of cholestyramine, rifampin, naltrexone, and the emerging bile-acid-transporter inhibitors.

10:54For 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.