LIVER CONSULT
| DRUG-INDUCED HEPATOTOXICITY —Willis C. Maddrey, MD, Professor of Internal Medicine, University of Texas
Southwestern Medical Center, Dallas
|
| Diagnosis of drug-induced liver injury: hardly any liver disease that cannot be mimicked by drug injury; diagnosis
of exclusion; underlying liver injury itself can divert attention from role of drug, eg, patient with HIV AIDS
coinfected with hepatitis C virus (HCV) and on multiple drugs; deceleration of injury after withdrawal of drug supports
diagnosis of drug-induced liver injury; detection of drug-induced injury particularly difficult when drug for
treatment of liver disease or administered to patient with liver disease; most experts do not believe that people with
underlying liver disease more likely to develop serious liver injury from drugs than those in general population, but
more likely to express liver injury because they do not have adequate defense system to handle metabolic products
of drugs
|
| Risk factors: age and sex—crucial; most serious liver injuries caused by drugs occur in people >40 yr of age and
in women; obesity—obese people not more likely to sustain liver injury from drug, but more likely (in certain circumstances)
to manifest it; drug’s intermediate passes through mitochondria already under assault because of obesity;
alcohol—causes fatty liver and other liver damage; metabolized by cytochrome P (CYP)450 2E1; diseases
that affect likelihood of hepatotoxicity—patients with chronic hepatitis B or C taking isoniazid have increased risk
(drug-disease interaction); sulfonamides in patients with HIV; tamoxifen in patients with steatohepatitis
|
| Mechanisms: overproduction and inadequate disposition of some intermediate metabolites; production of intermediate
dependent on variety of enzymes and on “staircasing” of enzymes, all of which have polymorphisms; focus
now on disposition of metabolites (also dependent on enzymes); reaction to liver damage swift and occurs in innate
immune system; overly exuberant reaction may cause damage from factors that try to minimize reaction
|
 | Adaptation: statins—cause elevated aminotransferases in ≤10% of patients; slightly dose-related, but probably mechanism-based;
statin blocks 3-hydroxy-3-methylglutaryl coenzyme A (HMG CoA), leading to backup of precursor, and
causing liver to leak enzyme; as drug continued, liver adapts and aminotransferases return to normal levels; so statins
seldom cause liver disease but frequently cause elevated aminotransferases; isoniazid—10% to 20% of people who
take isoniazid have elevated aminotransferases in first 2 to 3 wk, but only 1% develop liver injury
|
| Clinical course: if alanine aminotransferase elevated up to 3 times upper limit of normal (ULN), but without associated
jaundice or symptoms, recheck few days later; most slight elevations of aminotransferases caused by over-the-
counter drugs and obesity (hepatitis C causes 1%-3%); elevation of aminotransferases does not necessarily progress to
liver disease; drugs tend to have own “signature,” but may be “forged”; eg, in obese patient taking multiple medications,
may find mixed reaction, ie, drug that usually has hepatocellular effects, may show cholestatic features; jaundice
ominous sign (indicates serum bilirubin ≥3 mg/dL); time of onset—no case of dilantin-induced liver injury
occurring in patient taking it for >6 wk; most of nitrofurantoin-induced liver injury occurs after taking drug 6 mo; ticrynafen
(Selacrine)—uricosuric diuretic that caused several deaths; Zimmerman (1999) observed that all patients
who died had jaundice; elevated aminotransferase with clinical jaundice due to drug associated with ≈10% risk for
death; true for halothane, methyldopa, diclofenac, and troglitazone
|
| Acetaminophen: key mechanism of toxicity metabolism of acetaminophen to quinone form (through CYP450
2E1); regular heavy intake of alcohol raises CYP450 2E1 level; also produces toxic metabolite and blunts ability to
produce glutathione (hepatoprotectant); leads to cell damage; study from University of Arkansas—identified in serum
adducts of N-acetylquinone cysteine (appears only when acetaminophen level too high); test now available for
diagnosis of acetaminophen poisoning in emergency department (ED; see James et al in Suggested Reading); innate
immune system—first line of defense; Kupffer cells, macrophages, and natural killer T (NKT) cells; produce
cytokines, chemokines, reactive O2 species, and nitric oxide; about one-third of leukocytes in liver NK and NKT
cells; major source of interferon-γ (leads to activation of macrophages); macrophages and Kupffer cells produce tumor
necrosis factor (TNF)-α inflammatory cells recruited through interleukin (IL)-8; study by Kaplowitz—mice
depleted of NK and NKT cells and given acetaminophen dose toxic to mice; reported increased survival, decreased
necrosis on biopsy, decreased messenger RNA for interferon-γ, and decreased neutorophils and chemokines
|
| Isoniazid: normally metabolized to hydrazine; 2 enzymes involved (N-acetyl transferase and CYP450 2E1); hepatotoxins
related to production of acetylisoniazid and level of CYP450 2E1; CYP450 2E1 main determinant of development
of isoniazid hepatitis; study—measured CYP450 2E1 in group of people, 80% to 90% of whom had C1C1
(wild or common type); others had variant (C1C2 or C2C2); wild type had 20% chance of having decreased
CYP450 2E1 (vs 9% with variant); suggests that if CYP450 2E1 can be inhibited, can reduce likelihood of isoniazid
liver injury; increased risk for isoniazid liver injury in patients with hepatitis B or C; risk disappears if hepatitis B
virus (HBV) suppressed to low levels or HCV cleared
|
| Troglitazone (Rezulin): stops progression of diabetes; 2 yr after launch, 100 of 2 million people on drug developed serious
liver disease; diabetics have elevated aminotransferases (related to nonalcoholic steatohepatitis); in clinical trials,
almost 2% of patients had aminotransferases increased 3 times ULN, although most adapted; 2 patients (in ≈2500) had
jaundice; mechanism unknown, although drug did not cause accumulation of quinone; led to question of whether drug
taken today can cause liver disease to manifest 5 to 10 yr later; at present, prevailing thought that most of harm caused by
drugs becomes apparent in reasonable time after starting drug
|
| Phenytoin: hepatotoxicity rarely seen in those ≤14 yr of age; presents within first 4 wk of beginning drug and can
cause Stevens-Johnson syndrome; metabolized by CYP450 into arene oxide (important in carcinogenesis, and in
large amounts can cause injury to liver); detoxified by epoxide hydrolase; later discovered that patients with this
syndrome had genetic defect (dominant) in epoxide hydrolase; patient with phenytoin reactivity also likely to have
carbamazepine reactivity (same pathway); minor genetic changes may cause drug-induced liver disease
|
| Amiodarone: now being used for atrial fibrillation; previously used only for ventricular fibrillation and arrhythmias;
composed of 60% to 70% iodine by weight (causing thyroid issues); accumulates in lysosomes (as in Niemann-Pick
disease); can get into mitochondria and uncouple oxidation and phosphorylation, leading to steatohepatitis and injury;
patient with liver injury from amiodarone who has taken drug for 2 yr, then stopped, will have slow release of amiodarone
from lysosomes for up to 150 days (autodose), and injury will continue; interaction between alcohol-induced
liver disease and drug-induced liver disease most evident in acetaminophen; interaction between drug-induced liver
disease and nonalcoholic steatohepatitis evident with tamoxifen; all produce oxidative stress; question—are people
with fatty liver more likely to develop clinically apparent drug-induced hepatotoxicity? speaker believes moderate
predisposition present, and safest drugs diazepam (Valium) and statins
|
| Minocycline: tetracycline derivative used for acne; can mimic autoimmune hepatitis; oxyphenisatin first drug recognized
to cause autoimmune hepatitis
|
| Bosentan: lifesaving for patients with primary pulmonary hypertension; also leads to liver disease; can cause elevated
aminotransferases >3 times ULN in 16%; only few hundred doctors in world licensed to prescribe it; mechanism
is blockage of bile salt excretory pump (unknown whether same mechanism leads to liver injury); example of
drug released because its benefits outweigh risks
|
| Amoxicillin and potassium clavulanate (Augmentin): second only to acetaminophen in causing liver injury;
causes mild liver disease (rarely seen while patient taking drug; generally occurs 1 to 2 wk after stopping drug); patient
develops pruritus and elevated aminotransferases
|
| Herbs: vitamin A—dose-dependent hepatotoxin; should not take >5000 IUs per day; causes chronic liver disease
with cirrhosis and ascites; poor prognosis; kava—200 times more kava in pill than in cup of kava tea; caused
deaths
|
| THE FAILING LIVER —David R. Nelson, MD, Associate Professor and Director, Hepatology and Liver Transplantation,
University of Florida College of Medicine, Gainesville
|
| Case 1: black man 58 yr of age has chronic hepatitis C for 3 to 4 decades with significant comorbidities; genotype 1b
with high viral load; has 1 in 5 chance of responding to conventional therapy; has well preserved synthetic function; biopsy
shows advanced disease; ultrasonography (US) shows no cancer; has mild hepatosplenomegaly; platelet count
135,000/mL; endoscopy negative for varices, but portal gastropathy present; goal to treat aggressively; understanding
natural history of disease critical in making risk-benefit decisions
|
| HCV infection: 20% to 30% of patients clear virus spontaneously; of those who develop chronic infection, 1 in 5 develop
cirrhosis and 1 in 5 have bad outcome, eg, liver failure, cancer, need for transplantation; study—started in 1986 to
look at natural history of cirrhosis; hepatitis C unknown at that time; after hepatitis C discovered, HCV-positive cohorts
separated; (11% of 214 patients) and treated with interferon; none cured; 35% of people died within 17 yr (annual death
rate 4%); 32% developed liver cancer (annual incidence 3.9%); cancer most common cause of death in 50% of patients
and first complication to develop in 1 in 5 patients; ascites next most common complication, followed by jaundice and
gastrointestinal (GI) bleeding; encephalopathy rare; follow-up to study—to determine mortality rate once patient decompensates;
if patient develops cancer, mortality rate 31% within 1 yr of diagnosis; preventive measures—now recommended
that patients with cirrhosis be on screening regimen (6 or 12 mo); screen for varices; monitor patients using
model for end-stage liver disease (MELD) scoring system; speaker suggests monitoring all cirrhotic patients for 4 mo
|
| Case 2: woman 54 yr of age presents to ED with abdominal pain after eating; has had hepatitis C for 2 decades; nonresponder;
no varices, ascites, or encephalopathy; US shows 3 large gallstones and nondilated ducts; laboratory
tests show reasonable synthetic function; elective cholecystectomy planned for following day
|
| MELD system: created to predict postoperative mortality after transjugular intrahepatic portosystemic shunt
(TIPS); composed of creatinine, bilirubin, prothrombin time (PT), and international normalized ratio; good predictor
of mortality; score from 6 to 40; as score gets higher, mortality increases; current indication for organ allocation
in United States; with score of 14 to 15, 1-yr survival drops to <90%; most transplant centers have 1-yr survival
rate of 90%; when to refer for transplantation—in past, MELD score >10; at present, score of 14 to 15; elliptical
relationship between MELD score and postoperative mortality
|
| Child-Pugh classification: objective and subjective; composed of albumin, bilirubin, PT, ascites, and encephalopathy;
3-point scoring system; class A, B, or C based on score; also good at predicting postoperative mortality; all
patients with cirrhosis have increased risk; Child’s A has 10% mortality risk, B has 30%, and C has 75% to 80%; in
case above, woman had Child-Pugh score of B7 (30% risk for death from abdominal surgery) and MELD score of
16 (16% risk for 30-day mortality)
|
| Case 3: woman 64 yr of age with fatty liver, known cirrhosis, portal hypertension, and refractory ascites presents to
ED with melena; not orthostatic and admitted to floor for GI consultation; MELD score 11; anemia; started on
octreotide and proton pump inhibitor; endoscopy next morning showed large varices
|
| Spontaneous bacterial peritonitis (SBP): prevalence—10% to 30% of all hospitalized patients with ascites have
SBP on admission; in speaker’s hospital, all patients admitted to hospital with ascites get abdominal tap; patient
with GI bleeding and ascites (by definition, bacteremic [can seed ascites]); recommendation to give antibiotics to
patients with GI bleeding and ascites (usually 3- to 7-day course); mortality high; first perform diagnostic paracentesis;
in setting of SBP, avoid large-volume paracentesis; urine dipstick—used to look for polymorphonuclear
leukocytes in ascitic fluid and get rapid sense of whether patient has SBP; standard usually absolute
neutrophil count (ANC) >250/mm3 or absolute white blood cell (WBC) count of 500/mm3 to 750/mm3; at bedside,
can remove 10 mL of fluid and use dipstick; high score has positive predictive value of 98% (start patient on antibiotics);
score of 0 or 1 has high negative predictive value, so no need to do anything further, except large-volume
paracentesis to drain fluid; score of 2 equivocal; rapid and cheap; can use in outpatient setting
|
| Refractory ascites: infrequent; occurs in <5% to 10% of patients; more frequent occurrence means patient noncom-
pliant with diet or diuretics; precondition for hepatorenal syndrome; patients also at risk for SBP (should be on prophylactic
antibiotics); poor long-term survival; need transplantation, regardless of MELD; treatment options—
large-volume paracentesis, TIPS, or liver transplantation
|
| Case 3 continued: diagnostic paracentesis performed previously, and reagent strip positive; decided not to perform
large-volume paracentesis; patient started on antibiotic; only use for intravenous albumin prospectively in these
patients is presence of SBP; given on day 1 and day 3, can improve renal failure and survival; underwent transplant
evaluation; TIPS considered because of umbilical hernia, but patient’s son refused because of risk for encephalopathy
|
| TIPS: study—75 patients sent for TIPS; no treatment arm; given lactulose and rifaximin; results showed prophylactic
lactulose and rifaximin had no ability to prevent encephalopathy; best predictor of encephalopathy previous
history; the better the reduction in portal gradients or the bigger the shunt, more likely development of
encephalopathy
|
Suggested Reading
Bansal S et al: Acute liver failure. Indian J Pediatr 73:931, 2006; Bernal W et al: Acute liver failure. Curr Opin
Anaesthesiol 13:113, 2000; Cucchetti A et al: Recovery from liver failure after hepatectomy for hepatocellular carcinoma
in cirrhosis: meaning of the model for end-stage liver disease. J Am Coll Surg 203:670, 2006; Garcia-Gil FA
et al: Liver transplant, in emergency 0 (UNOS Status 1). Transplant Proc 38:2465, 2006; James LP et al: Detection
of Acetaminophen Protein Adducts in Children With Acute Liver Failure of Indeterminate Cause. Pediatrics
118:681, 2006; Kaplowitz N: Idiosyncratic drug hepatotoxicity. Nat Rev Drug Discov 4:489; 2005; Kuffner EK
et al: Retrospective analysis of transient elevations in alanine aminotransferase during long-term treatment with acetaminophen
in osteoarthritis clinical trials. Curr Med Res Opin 22:2137, 2006; Lewis JH: Drug-induced liver disease.
Curr Opin Gastroenterol 18:307, 2002; Lucena MI et al: Determinants of the clinical expression of
amoxicillin-clavulanate hepatotoxicity: a prospective series from Spain. Hepatology 44:850, 2006; Luedde T et al:
Intracellular survival pathways in the liver. Liver Int 26:1163, 2006; Mayoral W et al: Drug-induced liver disease.
Curr Opin Gastroenterol 15:208, 1999; Nathwani RA et al: Drug hepatotoxicity. Clin Liver Dis 10:207, 2006;
Onaca N et al: MELD-based allocation scheme for liver transplantation in Israel. Isr Med Assoc J 8:651, 2006;
Sotiropoulos GC et al: Split liver transplantation for hepatocellular carcinoma. Hepatogastroenterology 53:764,
2006; Steel JL et al: Health-related quality of life: Hepatocellular carcinoma, chronic liver disease, and the general
population. Qual Life Res Nov 21, 2006; Williams DP: Toxicophores: investigations in drug safety. Toxicology
226:1, 2006; Zimmerman HJ et al: Ticrynafen-associated hepatic injury: analysis of 340 cases. Hepatology 4:315,
1984.
Educational Objectives
| The goal of this program is to improve management of drug-induced hepatotoxicity and the failing liver. After hearing
and assimilating this program, the clinician will be better able to:
|
| 1. Cite the risk factors for drug-induced hepatotoxicity.
|
| 2. Review the mechanisms of hepatotoxicity due to acetaminophen, isoniazid, troglitazone, phenytoin, amiodarone,
minocycline, and amoxicillin and potassium clavulanate.
|
| 3. Discuss liver failure resulting from hepatitis C infection.
|
| 4. Interpret the model for end-stage liver disease (MELD) system and Child-Pugh classification scores to monitor
and predict mortality from liver disease.
|
| 5. Recognize spontaneous bacterial peritonitis and refractory ascites.
|
Faculty Disclosure
In adherence to ACCME Standards for Commercial Support, Audio-Digest requires all faculty members to disclose relevant
financial relationships within the past 12 months that might create any personal conflicts of interest. Any identified conflicts
were resolved to ensure that this educational activity promotes quality in health care and not a proprietary business or commercial
interest. For this program, the faculty reported nothing to disclose.
Acknowledgements
Dr. Maddrey was recorded at New Treatments in Chronic Liver Disease, held April 1-2, 2006, in La Jolla, CA, and
sponsored by the Scripps Clinic, La Jolla. Dr. Nelson was recorded at the 7th Annual Update in Gastoenterology,
held September 15-17, 2006, in Napa, CA, and sponsored by the Cedars-Sinai Medical Center, Los Angeles, CA. The
Audio-Digest Foundation thanks the speakers and the sponsors for their cooperation in the production of this program.
|
