Liver Test Interpretation and Workup: Liver Test Interpretation: Patterns and Workup

0:00Welcome to Board Pearls. This is episode one of two of the Liver Test Interpretation chapter, in the Liver Disease module. This episode is reading abnormal liver tests and working them up. A woman whose mildly high alanine aminotransferase, ALT, was called normal for two years shows what a normal ALT really is, how the R ratio decides the workup, what the size of the elevation means, and what the ratio of aspartate aminotransferase, AST, to ALT adds. A man with acute hepatocellular injury shows the workup in five layers, sent in one round, with his answer on his supplement shelf, and two women show the injuries recognized by pattern and the one that can't wait for a test. A man with cholestasis and ulcerative colitis shows why ultrasound comes first, followed by the antibody, the magnetic resonance cholangiopancreatography, MRCP, and the immunoglobulin G4, IgG4, level, and a woman with itching and a man with a stricture show what the antibody and the IgG4 each pick out.

1:07Here's the first patient. A forty-six-year-old woman with obesity, type two diabetes, and high triglycerides has had an ALT of thirty-eight on three checks over two years. Her lab's upper limit is forty, so each time it was reported as normal. Her AST is thirty. Her alkaline phosphatase is normal, at eighty, with an upper limit of one hundred twenty. Her bilirubin and albumin are normal. She drinks two glasses of wine a week. She's been told her liver is fine.

1:35Her liver isn't fine. She has a persistently abnormal ALT, most likely from metabolic dysfunction-associated steatotic liver disease, MASLD, though that has to be shown, not assumed. An abnormal liver test is a screening signal, not a diagnosis. The workup is organized by which enzyme dominates and by how many multiples of normal it sits at, so normal has to be defined first.

2:00Her ALT was called normal against the wrong limit. The historical ranges were set on populations that included people with fatty liver and undiagnosed hepatitis C, which pushed the upper limit up. The twenty seventeen American College of Gastroenterology, ACG, guideline moved the practical upper limit to under twenty-five for women and under thirty-three for men. Her ALT of thirty-eight is about one and a half times her true upper limit, in a patient with every metabolic risk factor. In biopsy series of unexplained mild aminotransferase elevation, about a third are MASLD and about a quarter pure steatosis, and the rest are chronic active hepatitis, sinusoidal fibrosis, and minimal change, so very few are nothing. Small elevations are worked up.

2:49Before anything is ordered, her pattern is set by the R ratio: the ALT as a multiple of its upper limit, divided by the alkaline phosphatase as a multiple of its upper limit. A ratio over five is hepatocellular, under two is cholestatic, and two to five is mixed. The hepatocellular and cholestatic patterns have almost completely different causes. Her alkaline phosphatase is normal and her ALT is the only raised value, so she's on the hepatocellular list.

3:17The ratio has to use multiples, not raw numbers. An ALT of two hundred eighty and an alkaline phosphatase of three hundred twenty look equal as raw values, but with upper limits of thirty-three and one hundred twenty they're about eight and a half and two point seven times normal, an R of about three, which is mixed.

3:39Each pattern has its own list. Hepatocellular injury means hepatocytes are dying or leaking. The acute list is viral hepatitis, drug-induced liver injury, ischemic hepatitis, autoimmune hepatitis, and Wilson disease in younger patients, and the chronic list is MASLD, alcohol, chronic viral hepatitis, autoimmune hepatitis, hemochromatosis, and alpha-one antitrypsin deficiency. Cholestatic injury involves the biliary epithelium and the canalicular machinery, and the first decision is whether the obstruction is mechanical, which is why ultrasound comes first. The mixed pattern is most often drug injury, Epstein-Barr virus, EBV, or cytomegalovirus, CMV, hepatitis, or primary sclerosing cholangitis with active inflammation, and it's worked up both ways.

4:30The size of her elevation narrows the list further. Chronic viral hepatitis and MASLD typically sit at forty to two hundred. Alcohol-associated hepatitis runs forty to one hundred fifty, with the AST capped under four hundred. Autoimmune hepatitis runs roughly two hundred to one thousand in its chronic phase. Acute viral or drug-induced hepatitis, other than acetaminophen, typically runs five hundred to two thousand. Over one thousand, the causes are acute viral, ischemic, drug-induced, autoimmune, and Wilson disease, told apart by tempo, history, and serology.

5:08Above five thousand there are essentially two answers: acetaminophen, with a rapid rise and fall over days, and ischemic hepatitis, with a markedly raised lactate dehydrogenase, LDH, and a preceding hypotensive episode. Her ALT sits in the range of MASLD and chronic viral hepatitis. With her metabolic profile MASLD leads, but she still needs the full workup and a fibrosis assessment, covered in episode two.

5:36The AST-to-ALT ratio adds one more clue. Her AST is lower than her ALT, the usual pattern in MASLD and in chronic hepatitis before cirrhosis. In otherwise typical chronic hepatitis, an AST greater than the ALT suggests cirrhosis from any cause, because the fibrosing liver makes less ALT, which has the shorter half-life, so the ratio flips. In a drinker, a ratio over two with an AST under four hundred is alcohol-associated hepatitis until proven otherwise, so a forty-nine-year-old heavy drinker with jaundice, an AST of two hundred forty, and an ALT of ninety-five fits. Alcohol doesn't drive the AST into the thousands, so an AST of three thousand in a drinker means looking for something else, such as acetaminophen. And a seventy-year-old resuscitated from septic shock yesterday, with an AST of six thousand and an ALT of four thousand eight hundred, has ischemic hepatitis as the leading diagnosis.

6:38Once the pattern is hepatocellular, the workup follows, and the second patient shows how it's sent. A thirty-four-year-old man has had fatigue, dark urine, and nausea for a week. His ALT is seven hundred eighty, his AST five hundred ten, his alkaline phosphatase normal, and his bilirubin three point two, with a normal international normalized ratio, INR. His R ratio is well over five. He takes no prescription drugs, drinks rarely, and hasn't traveled. He says he takes nothing else, but he lifts weights several times a week.

7:11He has acute hepatocellular injury with jaundice but a normal INR, and the cause has to be found quickly. The workup has five layers, viral, autoimmune, metabolic and inherited, ischemic, and drug, each with a defined first panel. The first-line tests from every layer are sent together in one round, because working down the list one layer at a time lets a treatable cause progress.

7:35The viral layer leads because viral hepatitis is treatable, transmissible, and common. It's hepatitis B surface antigen, hepatitis C antibody with a reflex to ribonucleic acid, RNA, and hepatitis A immunoglobulin M, IgM, for any acute presentation, with hepatitis E IgM in returning travelers and immunocompromised patients. EBV serology and CMV polymerase chain reaction, PCR, are added when the picture suggests mononucleosis, a young patient with fatigue, sore throat, splenomegaly, lymphocytosis, and an ALT two to ten times normal, and at his age they're included. Chapter eighteen covers viral hepatitis.

8:25The autoimmune layer is the antinuclear antibody, ANA, anti-smooth muscle antibody, anti-liver-kidney microsomal antibody type one, anti-LKM-1, and a quantitative immunoglobulin G, IgG. Type one autoimmune hepatitis presents with ANA and smooth muscle antibody, classically in a middle-aged woman with a high IgG, while type two, with anti-LKM-1, is more common in children, and anti-soluble liver antigen, anti-SLA, is a more specific second-line antibody.

9:01The simplified score turns those results into a probability. It gives one point for an ANA or smooth muscle antibody at one to forty, and two at one to eighty or for anti-LKM or anti-SLA; one point for an IgG above normal and two above one point one times normal; one or two for compatible or typical histology; and two for the absence of viral hepatitis. Six is probable, and seven or more is definite. Nineteen to thirty-five percent of autoimmune hepatitis is seronegative, so negative antibodies don't exclude it, and a biopsy is recommended in most suspected cases. Chapter twenty covers it.

9:43The metabolic and inherited layer comes next. Hemochromatosis is screened with a fasting transferrin saturation and ferritin. A saturation over forty-five percent with a high ferritin, outside an inflammatory state, goes to HFE testing for C282Y and H63D. But ferritin alone doesn't warrant HFE testing, because it's an acute phase reactant raised in MASLD, alcohol, and inflammation.

10:11The other two inherited diseases need their own tests. Wilson disease is screened with a ceruloplasmin in anyone under fifty-five with unexplained liver disease, and a low alkaline phosphatase and a low uric acid support it. A ceruloplasmin under twenty milligrams per deciliter is suggestive but not specific, since it also falls in malabsorption and synthetic failure, while one over thirty effectively excludes it, and the Leipzig score confirms the diagnosis at four or more. Alpha-one antitrypsin deficiency is screened by phenotype and genotype, not by level, because the level rises with inflammation and Pi ZZ can show intermediate levels. Chapter twenty-two covers these diseases.

10:54MASLD sits in this layer too, and it explains two common false leads. It's diagnosed by imaging steatosis plus at least one cardiometabolic criterion, obesity, diabetes, dyslipidemia, or hypertension, with significant alcohol and other causes excluded. About thirty percent of patients with MASLD have a high ferritin without needing HFE testing, and about twenty-one percent have an ANA of one to one hundred sixty or more, or a smooth muscle antibody of one to forty or more, without autoimmune hepatitis.

11:30The drug layer is a diagnosis of exclusion, not of last resort, so it's sought actively from the start and confirmed by excluding the rest. The history covers prescription drugs, over-the-counter medications, herbal and dietary supplements, recreational drugs, and acetaminophen. Supplements account for about sixteen percent of drug-induced liver injury in the United States registry, with anabolic steroids and green tea extracts dominant. Asked specifically about anything he takes for training, weight, or energy, he mentions a fat-burner supplement and a testosterone booster started two months ago.

12:00Drug injury is suspected when the timing fits a new agent, the pattern matches a known offender, such as amoxicillin-clavulanate, isoniazid, nitrofurantoin, trimethoprim-sulfamethoxazole, minocycline, statins, or herbals, and other causes are excluded. Hy's law applies: an ALT over three times normal with a bilirubin over two milligrams per deciliter, in drug injury without obstruction, carries about a ten percent mortality even when the drug is stopped promptly. Chapter nineteen covers drug injury.

12:32His results come back together. His panel is negative for hepatitis A, B, C, and E, EBV, and CMV. His ANA, smooth muscle antibody, IgG, transferrin saturation, and ceruloplasmin are normal. His ultrasound shows no biliary dilation. And his fat burner contains green tea extract. He meets Hy's law, so he's monitored closely with his INR. He stops both supplements, and over six weeks his ALT and bilirubin normalize.

13:07When the whole panel is negative and there's no drug, the list widens. It includes hepatitis E in immunocompromised patients with unexplained chronic hepatitis; CMV, EBV, adenovirus, leptospirosis, and Q fever; partial or intermittent biliary obstruction, which can give a hepatocellular pattern; infiltrating malignancy; and hemophagocytic lymphohistiocytosis. A patient who isn't improving without a cause needs a biopsy.

13:38The ischemic layer has no blood test, and the third patient shows how it's recognized by pattern. A sixty-seven-year-old woman had a cardiac arrest two days ago, and now her ALT is six thousand two hundred, with a far higher LDH and a bilirubin of only one point eight. Shock liver follows a hypotensive event, such as a cardiac arrest, septic shock, severe heart failure, or a bleed with instability. The LDH is raised out of proportion: an ALT-to-LDH ratio under one point five favors ischemic or acetaminophen injury, and over one point five favors viral hepatitis. The bilirubin is only mildly raised relative to the aminotransferases, whereas in acute viral hepatitis it tracks with them.

14:21Its course confirms it. The aminotransferases peak within twenty-four to forty-eight hours and fall by half every twenty-four to forty-eight hours as perfusion returns, and liver failure is unusual unless the shock persists. In a multicenter study, ischemia caused forty-three percent of aminotransferase elevations over one thousand, and above five thousand, acetaminophen and ischemia together accounted for most cases.

14:47One cause can't wait for its test. A twenty-nine-year-old woman in her third trimester has fever, confusion, and an ALT of four thousand. Herpes simplex hepatitis is rare but lethal in pregnant and immunocompromised patients, with about eighty percent mortality untreated. It's diagnosed by herpes simplex virus, HSV, PCR or by a biopsy showing intranuclear inclusions, but with high suspicion, empiric acyclovir starts immediately.

15:14A cholestatic pattern sends the workup in a different direction, and the fourth patient shows it. A thirty-eight-year-old man has had ulcerative colitis for ten years. His alkaline phosphatase is three hundred forty, his ALT seventy, and his bilirubin normal, an R ratio under two. He feels well apart from mild itching, and his colitis is in remission on mesalamine.

15:39He has a cholestatic pattern, and his ulcerative colitis already suggests a diagnosis, but imaging still comes first. The cholestatic differential turns on whether the obstruction is mechanical, so the first test is a right upper quadrant ultrasound, for an isolated alkaline phosphatase elevation, a cholestatic ratio, or jaundice with cholestatic labs. Sending antibodies before knowing whether the ducts are dilated can mean chasing an autoimmune diagnosis in a patient with a stone.

16:08A dilated tree means a mechanical cause, and the next step defines its level and cause. MRCP is the noninvasive test of choice, imaging the whole biliary tree without contrast or procedural risk. Computed tomography, CT, is added when a mass outside the duct is suspected, such as pancreatic cancer, hilar cholangiocarcinoma, or nodes compressing the porta hepatis. Endoscopic ultrasound, EUS, with fine-needle aspiration follows when a mass needs tissue, particularly for distal obstruction, where pancreatic head adenocarcinoma leads. Endoscopic retrograde cholangiopancreatography, ERCP, is reserved for planned therapy, stone extraction, stenting, or brushings, because post-ERCP pancreatitis occurs in three to ten percent, and chapters twenty-seven and twenty-nine cover the stone algorithm. His ultrasound shows no ductal dilation and a normal-looking liver, so his disease is inside the liver.

17:09With a normal tree, the first test is the anti-mitochondrial antibody, AMA, with a quantitative IgM, for primary biliary cholangitis. His is negative, and his colitis points to the next step.

17:25AMA-negative cholestasis with inflammatory bowel disease points to primary sclerosing cholangitis, diagnosed by an MRCP showing multifocal intrahepatic and extrahepatic strictures with a beaded appearance. MRCP performs about as well as ERCP at lower cost and risk, and a supportive cholangiogram means no biopsy is needed. Primary sclerosing cholangitis has a seventy to eighty percent association with inflammatory bowel disease, IBD, typically ulcerative colitis with a right-sided, pancolitic predominance and backwash ileitis. Perinuclear antineutrophil cytoplasmic antibody, pANCA, is positive in seventy to eighty percent but isn't specific enough to diagnose it. The alkaline phosphatase is normal in forty to fifty percent and fluctuates, so a normal value doesn't exclude it. His MRCP shows multifocal intrahepatic strictures with beading, so he has primary sclerosing cholangitis, diagnosed by the picture of his ducts rather than by any antibody.

18:24One mimic has to be weighed. IgG4-related sclerosing cholangitis can draw the same picture and responds to steroids, unlike primary sclerosing cholangitis. It's the diagnosis to consider when the patient doesn't fit, typically an older man with pancreatic involvement and strictures near the pancreas, and a high IgG4 alone doesn't settle it, because some patients with primary sclerosing cholangitis have one. He's thirty-eight, with ten years of colitis and intrahepatic strictures, so he fits primary sclerosing cholangitis.

18:58His typical cholangiogram means no biopsy. Primary sclerosing cholangitis still needs one in two settings. The first is small-duct disease, about five percent of cases, with a normal cholangiogram and a biopsy showing concentric periductal, onion-skin fibrosis with ductular proliferation. The second is a suspected overlap with autoimmune hepatitis, an ALT at least five times normal without obstruction. He needs neither. His follow-up, including cholangiocarcinoma surveillance, is in chapter twenty, and his colonoscopy surveillance becomes yearly, as chapter fourteen described.

19:34In primary biliary cholangitis, the antibody makes the diagnosis, and the fifth patient shows it. A fifty-seven-year-old woman has fatigue and itching, an alkaline phosphatase of two hundred eighty, more than one and a half times normal, an AST under five times normal, a normal ultrasound, and a positive AMA. That combination establishes primary biliary cholangitis, with about a ninety-eight percent positive predictive value, without a biopsy.

20:02The antibody has limits at both ends. About five percent of cases are AMA-negative, and the twenty eighteen American Association for the Study of Liver Diseases, AASLD, guideline accepts anti-gp210 and anti-sp100, which are about ninety-eight percent specific but about twenty-five percent sensitive. When they're unavailable or negative, a biopsy shows non-suppurative destructive cholangitis of the small intralobular ducts, with granulomas. And AMA isn't screened in people with normal liver tests, because about one in six thousand two hundred fifty adults without disease has a positive AMA, and only about one in six of those develops primary biliary cholangitis over five years. Chapter twenty covers treatment.

20:49The last patient shows a stricture that looks like cancer. A sixty-six-year-old man has painless jaundice, a distal bile duct stricture, and fullness of the pancreatic head, and everyone assumes pancreatic cancer, but his IgG4 is two hundred forty. IgG4-related sclerosing cholangitis is the biliary manifestation of IgG4-related disease, often with type one autoimmune pancreatitis, salivary gland inflammation, and retroperitoneal fibrosis. A serum IgG4 over one hundred thirty-five milligrams per deciliter is suggestive but not specific. The strictures tend to involve the bile duct inside the pancreas and the proximal ducts, and they respond to corticosteroids, which is both the diagnostic and the therapeutic discriminator. He starts prednisone, and his stricture and jaundice resolve.

21:40When none of these fit, the list widens. It includes drug-induced cholestasis from amoxicillin-clavulanate, anabolic steroids, oral contraceptives, parenteral nutrition, and antibiotics; infiltrative disease, meaning sarcoidosis with granulomatous hepatitis, amyloidosis, lymphoma, and metastases; genetic and congenital causes, such as cystic fibrosis liver disease, the inherited intrahepatic cholestasis syndromes, choledochal cysts, and sickle cell hepatopathy; cholestasis of sepsis; and obstruction missed on the first imaging, from intermittent stones or a periampullary cancer that needs EUS. Rarer causes include diabetic hepatosclerosis, nodular regenerative hyperplasia, graft-versus-host disease, the cholangiopathy of acquired immunodeficiency syndrome, AIDS, idiopathic adulthood ductopenia, and macro-alkaline phosphatase. A persistently high alkaline phosphatase after a negative workup goes to biopsy.

22:34Here are the traps the boards set in these patients. The lab's upper limit isn't the true normal for ALT. The R ratio comes first, and it's calculated from multiples, not raw values. Alcohol doesn't produce an AST in the thousands. An AST above the ALT in chronic hepatitis can signal cirrhosis. And a level over five thousand means acetaminophen or ischemia, not a broad workup.

22:58The hepatocellular traps follow. The workup isn't sent one layer at a time. A high ferritin alone doesn't call for HFE testing. A low-titer ANA in fatty liver isn't autoimmune hepatitis, and negative serology doesn't rule autoimmune hepatitis out. Alpha-one antitrypsin isn't checked by level. "No medications" isn't the end of the history until supplements are asked about. Ischemia after a hypotensive event, with a high LDH, is easy to miss. And acyclovir doesn't wait for an HSV result.

23:30The cholestatic traps close the list. Cholestatic serology doesn't come before the ultrasound. ERCP isn't a diagnostic test. A normal alkaline phosphatase doesn't rule out primary sclerosing cholangitis, and pANCA doesn't diagnose it. Typical primary sclerosing cholangitis on MRCP doesn't need a biopsy, and neither does a patient with a positive AMA and a cholestatic pattern. AMA isn't screened with normal liver tests. And a distal stricture with pancreatic fullness isn't called cancer without checking the IgG4.

24:03So, start to finish. An abnormal liver test is a signal, and an ALT over twenty-five in women or thirty-three in men is abnormal. Calculate R from multiples of the upper limit: over five is hepatocellular, under two cholestatic, and two to five mixed. Read the size: forty to two hundred for MASLD and chronic viral hepatitis, over one thousand for viral, ischemic, drug, autoimmune, or Wilson disease, and over five thousand for acetaminophen or ischemia. An AST-to-ALT ratio over two with an AST under four hundred in a drinker is alcohol-associated hepatitis, and an AST above the ALT in chronic hepatitis suggests cirrhosis.

24:45Send the hepatocellular workup in one round across the viral, autoimmune, metabolic and inherited, ischemic, and drug layers, with a full supplement history, and remember Hy's law and empiric acyclovir for suspected HSV.

25:00Start cholestasis with an ultrasound. A dilated tree goes to MRCP, with ERCP only for therapy. A normal tree goes to AMA, and a positive AMA with an alkaline phosphatase over one and a half times normal diagnoses primary biliary cholangitis without a biopsy. AMA-negative cholestasis with IBD goes to MRCP for primary sclerosing cholangitis. And a sclerosing picture with pancreatic involvement or a high IgG4 is IgG4-related cholangitis until steroids prove otherwise.

25:35Episode two is the single abnormal value, an isolated bilirubin or an isolated alkaline phosphatase, and then staging the liver: noninvasive fibrosis tests, liver biopsy, and the hepatic venous pressure gradient.

25:50For 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 two of chapter seventeen, and I'll see you in the next one.