Drug-Induced Liver Injury and Acute Liver Failure: Drug Injury and Liver Failure: Acute Liver Failure

0:00Welcome to Board Pearls. This is episode two of two of the Drug-Induced Liver Injury and Acute Liver Failure chapter, in the Liver Disease module. This episode is acute liver failure, from the definition to the transplant decision. A previously healthy woman who becomes confused shows what defines it, which chronic diseases still count, why it swells the brain when cirrhosis doesn't, what causes it, which causes recover on their own, and how each complication is managed. Three more patients show the causes where treatment starts before the diagnosis is confirmed: a young man with fulminant Wilson disease, a pregnant woman with herpes simplex hepatitis, and a man whose liver failure isn't from acetaminophen but still responds to N-acetylcysteine. Then a woman after an acetaminophen overdose shows when to transfer, how the King's College criteria work, what Status one-A listing means, and what rules a patient out, and a man injured by a supplement shows the other version of the criteria.

0:56Here's the first patient. A thirty-three-year-old woman, previously healthy, has had nausea and jaundice for two weeks. Today she's confused and drowsy, with asterixis. Her alanine aminotransferase, ALT, is two thousand one hundred, her bilirubin nine, her international normalized ratio, INR, two point eight, and her glucose fifty-eight. She has no history of liver disease, and her ultrasound shows a normal-sized liver without cirrhosis. She takes no medications or supplements, and her acetaminophen level is undetectable.

1:31Jaundice, coagulopathy, and new confusion in a previously healthy woman is acute liver failure, and she meets the definition exactly. Acute liver failure is the rapid loss of the liver's synthetic and detoxifying function in a previously healthy liver, and it has four criteria. The first is an INR of at least one and a half, which is the international consensus, though some references use two. The second is any grade of hepatic encephalopathy. The third is a total illness of less than twenty-six weeks. The fourth is no pre-existing cirrhosis. Her INR is two point eight, her asterixis and confusion are encephalopathy, she's been ill for two weeks, and her liver shows no cirrhosis, so she meets all four.

2:15The last two criteria draw the edges. The twenty-six-week limit separates acute liver failure from acute-on-chronic liver failure, and the absence of cirrhosis separates it from decompensation of established disease. An illness over twenty-six weeks, or a nodular liver, would take her out of the diagnosis.

2:34Underlying chronic disease doesn't always exclude it. Three exceptions qualify as acute liver failure despite underlying disease. Autoimmune hepatitis presenting fulminantly without a prior diagnosis can show established chronic changes on histology. Wilson disease presenting fulminantly always has years of copper deposition. And hepatitis B reactivation during immunosuppression or chemotherapy may sit on prior chronic infection but presents as acute decompensation. All three behave like acute liver failure and need the same urgent management. Chapter eighteen covers reactivation.

3:12The distinction from acute-on-chronic failure matters because of her brain. Acute liver failure is the condition in which cerebral edema develops, with raised intracranial pressure and herniation. Acute-on-chronic failure, despite often having the higher Model for End-Stage Liver Disease with sodium, MELD-Na, score and the more decompensated state, doesn't produce significant cerebral edema, because the brain has adapted over time to chronic ammonia, so astrocyte swelling doesn't progress to herniation the same way.

3:44A man with cirrhosis shows the contrast. A sixty-two-year-old man with alcohol-related cirrhosis has a new INR of three, jaundice, and confusion after an infection. He has acute-on-chronic failure, not acute liver failure, and his encephalopathy is managed as cirrhotic encephalopathy, without the cerebral edema measures that follow.

4:07Her timeline places her in the course. Acute liver failure runs through four overlapping phases: a prodrome of malaise and nausea; biochemical evolution, with rising aminotransferases and jaundice; synthetic failure, with a rising INR and falling fibrinogen; and encephalopathy, the progressive cognitive impairment that defines the diagnosis. Her two weeks of nausea and jaundice were the first two phases, her INR shows the third, and her confusion is the fourth.

4:35Her encephalopathy grade predicts what her brain will do. Grade one is mild confusion, a reversed sleep-wake cycle, and inattention. Grade two is lethargy, asterixis, and disorientation to time. Grade three is somnolence with confusion, marked asterixis, and disorientation to place. Grade four is coma without response to pain. The risk of cerebral edema rises sharply at grades three and four. She's lethargic, with asterixis, and doesn't know the date, but she knows she's in a hospital, so she's grade two.

5:11With the diagnosis made, the cause comes next, because it determines management. In the United States Acute Liver Failure Study Group registry, acetaminophen accounts for roughly forty-five to forty-seven percent of cases. Indeterminate cases, presumed but unproven viral or drug injury, make up about fourteen percent; drug injury other than acetaminophen, about eleven percent; viral hepatitis, about seven percent, mostly hepatitis B, with a small share of hepatitis A and an even smaller share of hepatitis E outside pregnancy; autoimmune hepatitis, about five percent; ischemic injury, or shock liver, about four percent; and Wilson disease, two to three percent. Pregnancy-associated causes, acute fatty liver of pregnancy, severe preeclampsia, and the syndrome of hemolysis, elevated liver enzymes, and low platelets, HELLP, and Budd-Chiari syndrome each account for a small share.

6:01Where she's been can matter too. In developing countries, viral hepatitis B, A, and E becomes the dominant cause, and genotype one hepatitis E carries a high mortality in pregnant women in South Asia, so a returning traveler, especially a pregnant one, is asked where she's been. Her acetaminophen level is undetectable, she takes nothing, and she's had no shock, so her viral panel, autoimmune markers, copper studies, and pregnancy test all go out at once.

6:30The cause also predicts whether she'll recover on her own. Acetaminophen has the best spontaneous recovery, with about sixty-five to seventy percent surviving without transplant, because the injury stops once the drug is metabolized and N-acetylcysteine is given. Indeterminate and idiosyncratic drug cases recover least often, because the immune injury continues after the drug is withdrawn. Her workup comes back negative, so she's indeterminate, in the group least likely to recover. She's transferred to the intensive care unit, ICU, of a transplant center, and with grade two encephalopathy she gets N-acetylcysteine even without acetaminophen, for reasons covered later in this episode.

7:10In the ICU, her complications outside the brain are managed first. Hypoglycemia comes from the loss of hepatic gluconeogenesis and is treated with a continuous dextrose infusion, so with a glucose of fifty-eight she starts one now. Coagulopathy, with an INR routinely above two or three, isn't corrected with fresh frozen plasma unless she bleeds or needs an invasive procedure, because the INR is the prognostic marker and correcting it hides the trend. Vitamin K is given to exclude a reversible deficiency, and her INR of two point eight is followed, not corrected.

7:46The other organs follow. Kidney failure complicates about thirty to fifty percent of cases, often from acute tubular necrosis or hepatorenal physiology, and it's managed with continuous renal replacement therapy when needed. Sepsis is common, from bacterial translocation from the gut and from line infections, and broad-spectrum antibiotics are often given empirically, because the systemic inflammatory response of acute liver failure mimics infection. The circulation is hyperdynamic, like cirrhotic physiology, with a high cardiac output, a low systemic vascular resistance, and a frequent need for vasopressors.

8:23Her biggest threat is cerebral edema, the complication that kills. Ammonia crosses the blood-brain barrier, and astrocytes convert it to glutamine through glutamine synthetase. Glutamine is osmotically active, so the astrocytes swell, the brain swells, intracranial pressure rises, and herniation follows. An arterial ammonia over one hundred fifty to two hundred micromoles per liter is the threshold above which the risk rises sharply, and mortality climbs with the ammonia.

8:51Two days later, she's somnolent and doesn't know where she is, which is grade three. She's intubated to protect her airway, and her arterial ammonia is one hundred ninety, over the threshold. Her head is raised to thirty degrees. She's sedated to reduce metabolic demand and ventilated to a normal or mildly low carbon dioxide, a PaCO2 of thirty to thirty-five. Hypertonic saline holds her sodium at one hundred forty-five to one hundred fifty-five, and mannitol, half to one gram per kilogram, is kept for breakthrough intracranial hypertension. Because her ammonia stays very high, renal replacement therapy starts to remove it.

9:31Some familiar tools help less here. Lactulose helps less than in chronic encephalopathy, because the timeline is too short for clearing the gut to matter. Therapeutic hypothermia, to thirty-two to thirty-three degrees, is sometimes used in refractory intracranial hypertension as a bridge to transplant. Intracranial pressure monitors are no longer routine, because of the bleeding risk in a coagulopathic patient, and experienced centers use transcranial Doppler or optic nerve sheath ultrasound instead.

10:01Her workup checked for three causes that can't wait for their confirmatory tests, and the next patients show each one. A nineteen-year-old man has had jaundice for a week and is now confused. His hemoglobin has fallen to eight, his Coombs test is negative, and his bilirubin is twenty-four, much of it indirect. His alkaline phosphatase is forty, his aspartate aminotransferase, AST, three hundred twenty, his ALT one hundred ten, his uric acid low, and his INR three point one. He's never had liver disease, takes nothing, and his acetaminophen level is undetectable.

10:37A young patient with liver failure and Coombs-negative hemolysis has fulminant Wilson disease until proven otherwise. Fulminant Wilson disease is the presentation of a previously undiagnosed patient, almost always young, most between five and thirty-five, and it's universally fatal without transplant. As liver cells die, they release their stored copper all at once, and the copper damages red cells oxidatively and causes massive hemolysis, so the jaundice comes from hemolysis and synthetic failure together.

11:09Four findings give it away. The first is a Coombs-negative hemolytic anemia with a high indirect bilirubin, Coombs-negative because copper, not antibody, damages the cells. The second is an alkaline phosphatase-to-bilirubin ratio under four, because the alkaline phosphatase is low for a bilirubin driven up by hemolysis, and his is forty over twenty-four, about one point seven. The third is an AST-to-ALT ratio over two point two, and his is three hundred twenty over one hundred ten, about two point nine. The fourth is a low uric acid, because copper injures the proximal renal tubules, causing a Fanconi-like proximal tubular acidosis, so uric acid is lost in the urine. He has all four. His aminotransferases are only in the hundreds, nothing like the thousands of acetaminophen or ischemia, so a modest ALT doesn't exclude acute liver failure.

12:04The usual Wilson tests can mislead in this setting. Kayser-Fleischer rings are present in more than eighty percent of patients with neurologic Wilson disease but in only about half of those with liver disease alone, so their absence doesn't exclude it. The slit lamp exam is still done, because rings support the diagnosis, and his shows faint rings. Ceruloplasmin is unreliable acutely, because it's an acute phase reactant that illness pushes up, and it's made by the liver, so synthetic failure of any cause pushes it down. A normal value doesn't exclude Wilson disease, and a low value doesn't prove it.

12:37The reliable acute markers measure copper itself. Serum free copper, the copper not bound to ceruloplasmin, is very high, and twenty-four-hour urine copper is raised. ATP7B genetic testing confirms the diagnosis but takes time, and a liver copper over two hundred fifty micrograms per gram of dry weight is diagnostic, but a biopsy is rarely safe in a coagulopathic patient. His serum free copper and urine copper are both very high. A normal ceruloplasmin and no rings wouldn't have cleared him, because his hemolysis, his two ratios, and his low uric acid would still have called for copper studies and a call to the transplant center.

13:17His treatment is urgent transplant. Penicillamine, the first-line chelator for chronic Wilson disease, is avoided in the fulminant form, because it can worsen the acute phase by mobilizing copper too rapidly. Plasmapheresis and albumin dialysis, with systems such as the molecular adsorbent recirculating system, MARS, or Prometheus, remove free copper and bridge to transplant. He's listed for urgent transplant and has plasmapheresis while he waits. Chapter twenty-two covers the chronic management of Wilson disease.

13:46The second cause is herpes simplex hepatitis. A thirty-year-old woman at thirty-one weeks of pregnancy has a fever of thirty-nine point two, an ALT of six thousand, a bilirubin of only one point eight, and new confusion. Her skin is clear. Pregnancy, fever, and liver failure without jaundice point to herpes simplex hepatitis, from herpes simplex virus, HSV, one or two.

14:11It's rare, and it strikes two groups. The first is pregnant women, especially in the second and third trimesters, who account for roughly a quarter of reported disseminated HSV. The second is immunocompromised patients, including transplant recipients, people with human immunodeficiency virus, HIV, infection, and those on immunosuppression for autoimmune disease. Primary or reactivated infection escapes immune control and kills liver cells directly, and a biopsy shows patchy zonal necrosis, with Cowdry type A inclusions in the surviving cells beside the dead ones.

14:49The pattern is the clue. The aminotransferases are very high, often twenty-five to forty times normal, but the bilirubin is normal or low, often under three, an anicteric liver failure. Fever is almost universal. The vesicular rash is absent in up to eighty percent, so clear skin doesn't exclude it. A sore throat or upper respiratory symptoms can be present, and encephalitis can coexist. She has pregnancy, an ALT of six thousand with a bilirubin under two, fever, and confusion with no other cause.

15:19Her treatment can't wait for the test. The diagnosis is made by HSV polymerase chain reaction, PCR, on serum, or on liver tissue when a biopsy is feasible, but it's missed in most cases before death, and delay drives mortality over eighty percent without acyclovir. So intravenous acyclovir, five to ten milligrams per kilogram every eight hours, starts empirically in any pregnant or immunocompromised patient with indeterminate acute liver failure. She starts it in the emergency department while the PCR is sent. The PCR comes back positive, and she improves. Had she worsened, herpes simplex hepatitis is a cause that can be listed for transplant.

15:57The third measure is N-acetylcysteine when acetaminophen isn't the cause. A fifty-two-year-old man has acute liver failure from a supplement, an undetectable acetaminophen level, and grade two encephalopathy. In a randomized trial in acute liver failure not caused by acetaminophen, intravenous N-acetylcysteine improved transplant-free survival in patients with grade one or two encephalopathy at presentation, while patients arriving at grade three or four gained no overall survival benefit. The benefit probably comes from anti-inflammatory and microcirculatory effects, since the toxic acetaminophen metabolite, NAPQI, isn't involved.

16:38So any patient with acute liver failure of indeterminate or drug cause and grade one or two encephalopathy gets empiric intravenous N-acetylcysteine, dosed as for acetaminophen toxicity: one hundred fifty milligrams per kilogram over sixty minutes, then fifty over four hours, then one hundred over sixteen hours, continued until improvement or transplant. He starts it on arrival, as the first woman did. Corticosteroids are a different matter. They don't improve mortality in drug-induced acute liver failure and may raise the risk of infection, so they're reserved for autoimmune hepatitis presenting as liver failure and for high-grade checkpoint inhibitor hepatitis.

17:18The last decision is who needs a transplant, and the next patient shows how early that starts. A twenty-six-year-old woman was brought to a community hospital thirty hours after a large acetaminophen overdose and got N-acetylcysteine on arrival. By the next morning she's confused, her INR is three point two, and her creatinine has risen from zero point eight to one point nine. After two liters of fluid, her arterial pH is seven point two six. The team plans to watch her another day and transfer her if she gets worse.

17:48She has acetaminophen-induced acute liver failure, and she needs transfer now. The King's College criteria identify patients who already have severe organ failure, so waiting for them means waiting too long, and the goal is to have the patient at a transplant center while there's still time to optimize and list her. American Association for the Study of Liver Diseases, AASLD, guidance gives the triggers for transfer. For liver failure not caused by acetaminophen, any patient with grade one or two encephalopathy, an INR over one and a half, or a rising bilirubin is transferred. For acetaminophen, transfer comes at the first sign of encephalopathy, persistent acidosis after volume resuscitation, an INR over two that doesn't improve with vitamin K, or any rise in creatinine, because these patients can deteriorate fast. She's confused, her acidosis persists after fluids, her INR is over two, and her creatinine has risen, so she meets all four triggers. Watching her another day is the mistake, and she's transferred that morning.

18:54At the transplant center, whether she needs a transplant is judged by the King's College criteria. They're prognostic tools, developed at King's College Hospital in London, that identify patients unlikely to recover without transplant, by capturing the threshold of synthetic, acid-base, kidney, and brain failure beyond which spontaneous recovery is rare. They come in acetaminophen and non-acetaminophen versions, because the prognostic factors differ.

19:20The acetaminophen version has two paths. The first is an arterial pH under seven point three zero after adequate volume resuscitation, which captures the profound lactic acidosis of the failing liver. The pH is measured after fluid, because a dehydrated patient can be acidotic for reasons that fluid fixes. The second path requires all three of grade three or four encephalopathy, an INR over six point five, which is a prothrombin time over one hundred seconds, and a creatinine over three point four milligrams per deciliter, or three hundred micromoles per liter, and one or two of the three isn't enough. Some references add an arterial lactate over three and a half millimoles per liter on admission, or over three after fluid resuscitation.

20:07Her pH of seven point two six after two liters meets the first path alone. With a pH of seven point three four, she'd meet none of the second path's three features, so she wouldn't meet the criteria, but she'd still meet every transfer trigger. Transfer comes early, and listing comes when the criteria are met.

20:27Meeting the criteria changes her place in line. In the United States, it means immediate Status one-A listing, the highest priority in the national allocation system, above any calculated MELD-Na in the regional and national queues. Status one-A in acute liver failure requires admission to the ICU, an age over eighteen, with separate criteria for children, a life expectancy without transplant of less than seven days, and at least one of ventilator dependence, renal replacement therapy, or an INR over two. The listing is time-limited, and most centers re-list every seven days. She's in the ICU, she's an adult, and her INR is over two, so she's listed Status one-A that afternoon.

21:11Before she gets a liver, the contraindications are checked. Active uncontrolled sepsis is a contraindication until it's controlled, with positive blood cultures and pressor-dependent shock the disqualifiers, but a controlled infection, such as cholangitis or a urinary source, doesn't necessarily preclude listing. Severe lung disease with poor functional reserve is a contraindication, and so is cancer outside the liver, other than completely treated non-melanoma skin cancer.

21:38The brain matters here too. Irreversible brain injury, from prolonged intracranial hypertension or a sustained loss of brainstem reflexes, rules out transplant, which is why controlling cerebral edema with hypertonic saline, mannitol, sedation, and normocapnia matters for eligibility as well as survival. Persistent non-adherence, a lack of social support, and active substance use are relative contraindications left to the transplant center's judgment, so her overdose isn't an absolute bar. She has no absolute contraindications, receives a liver two days later, and recovers.

22:16Other causes use the other version, and a final patient shows it. A fifty-seven-year-old man developed jaundice from a weight-loss supplement and was jaundiced for twelve days before he became confused. His INR is four and his bilirubin twenty. This version also has two paths. The first is an INR over six point five, a prothrombin time over one hundred seconds, on its own. The second is any three of five features: an age under ten or over forty, an indeterminate or drug-induced cause, jaundice for more than seven days before encephalopathy, an INR over three and a half, which is a prothrombin time over fifty seconds, and a bilirubin over seventeen and a half milligrams per deciliter, or three hundred micromoles per liter. A subacute course and drug-induced failure score heavily because they recover worst, as the immune injury continues after the drug stops.

23:09His INR of four misses the first path, but he's over forty, his cause is a drug, he had twelve days of jaundice before encephalopathy, his INR is over three and a half, and his bilirubin is over seventeen and a half, so he has five of five. He's transferred and listed Status one-A, and his N-acetylcysteine continues. His INR alone would look modest beside an acetaminophen case, but his slow course and his cause put him over the line, because twelve days of jaundice before confusion means he isn't improving.

23:41The first woman scores two of five. Her age of thirty-three doesn't count, her indeterminate cause and her two weeks of jaundice before confusion do, and her INR of two point eight and bilirubin of nine fall short. She didn't meet the criteria at presentation, but she was already at a transplant center, where her rising encephalopathy grade was watched.

24:02The criteria are good but not perfect, and other scores exist. MELD is used as a continuous measure of severity, not as the transplant trigger in acute liver failure. The French Clichy criteria use the factor five level with encephalopathy, under thirty percent in patients over thirty or under twenty percent in younger patients, and were developed for hepatitis B liver failure. The United States registry prognostic index combines encephalopathy grade, INR, bilirubin, phosphate, and cause to estimate transplant-free survival. The King's College criteria are the standard, and the systems generally agree on who's at highest risk, but the King's College sensitivity for predicting the need for transplant is about seventy to eighty percent and its specificity eighty-five to ninety-five percent. So the criteria serve as a transplant threshold, not a strict rule. The transplant team decides borderline cases, and a clearly deteriorating patient can be listed when the criteria will soon be met.

25:00Here are the traps the boards set in these patients. A decompensated patient with cirrhosis doesn't have acute liver failure. Fulminant Wilson disease, autoimmune hepatitis, and hepatitis B reactivation count despite chronic disease. Cerebral edema care isn't applied to acute-on-chronic failure. Fresh frozen plasma doesn't correct the INR without bleeding. Hypoglycemia is easy to miss. Lactulose isn't the treatment for cerebral edema. And hepatitis E matters in a pregnant traveler.

25:31The traps for the specific causes follow. A normal ceruloplasmin or absent Kayser-Fleischer rings don't rule out Wilson disease, and Coombs-negative hemolysis is the clue. Aminotransferases only in the hundreds don't exclude liver failure. Fulminant Wilson disease needs transplant listing, not penicillamine. A missing rash or a low bilirubin doesn't rule out herpes simplex hepatitis, and acyclovir doesn't wait for the PCR. N-acetylcysteine isn't withheld because acetaminophen isn't involved, and it helps at grade one or two, not three or four. And steroids don't help drug-induced liver failure.

26:11The transplant traps close the list. Transfer doesn't wait for the King's College criteria. The pH is measured after fluid resuscitation. In acetaminophen failure, one or two of the three combined criteria aren't enough. The acetaminophen version isn't applied to a drug-induced case, or the reverse. A subacute course and a drug cause count against recovery. MELD isn't the trigger. A controlled infection isn't a contraindication. And uncontrolled sepsis or irreversible brain injury rules out transplant.

26:45So, start to finish. Acute liver failure is an INR of one and a half or more with any encephalopathy, less than twenty-six weeks of illness, and no cirrhosis, and fulminant Wilson disease, autoimmune hepatitis, and hepatitis B reactivation qualify despite underlying disease. Only acute liver failure causes cerebral edema, with the risk rising at grade three or four encephalopathy and an arterial ammonia over one hundred fifty to two hundred. Acetaminophen causes nearly half of United States cases and recovers best, while indeterminate and drug cases recover worst.

27:20Give dextrose, avoid plasma unless there's bleeding, and treat raised intracranial pressure with head elevation, sedation, a PaCO2 of thirty to thirty-five, hypertonic saline, mannitol, and ammonia removal by renal replacement therapy.

27:35Three treatments start before confirmation. A young patient with Coombs-negative hemolysis, an alkaline phosphatase-to-bilirubin ratio under four, an AST-to-ALT ratio over two point two, and a low uric acid has fulminant Wilson disease and needs urgent transplant listing, not penicillamine, with plasmapheresis as a bridge. A pregnant or immunocompromised patient with very high aminotransferases, a low bilirubin, and fever gets intravenous acyclovir before the PCR returns. And liver failure not caused by acetaminophen, with grade one or two encephalopathy, gets intravenous N-acetylcysteine, without steroids.

28:17Transfer early, before the King's College criteria are met. Acetaminophen failure meets them with a pH under seven point three zero after resuscitation, or with grade three or four encephalopathy plus an INR over six point five plus a creatinine over three point four. Other failure meets them with an INR over six point five, or with any three of an age under ten or over forty, an indeterminate or drug cause, jaundice more than seven days before encephalopathy, an INR over three and a half, and a bilirubin over seventeen and a half. Meeting them means Status one-A listing, and uncontrolled sepsis and irreversible brain injury rule out transplant.

28:56That completes the Drug-Induced Liver Injury and Acute Liver Failure chapter. The next chapter is autoimmune and cholestatic liver disease.

29:05For 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 two of two of chapter nineteen, and I'll see you in the next one.