Measurement of Serum Acetaminophen Protein Adducts in Patients With Acute Liver Failure

Transcription

1 GASTROENTEROLOGY 2006;130: Measurement of Serum Acetaminophen Protein Adducts in Patients With Acute Liver Failure TIMOTHY J. DAVERN II,* LAURA P. JAMES, JACK A. HINSON, JULIE POLSON, ANNE M. LARSON, ROBERT J. FONTANA, # EZMINA LALANI,** SANTIAGO MUNOZ, A. OBAID SHAKIL, WILLIAM M. LEE, and The Acute Liver Failure Study Group *University of California at San Francisco, San Francisco, California; University of Arkansas for Medical Sciences and Arkansas Children s Hospital Research Institute, Little Rock, Arkansas; University of Arkansas for Medical Sciences, Little Rock, Arkansas; University of Washington, Seattle, Washington; University of Texas Southwestern Medical Center, Dallas, Texas; # University of Michigan, Ann Arbor, Michigan; **University of Texas Southwestern Medical Center, Dallas, Texas; Albert Einstein Medical Center, Philadelphia, Pennsylvania; University of Pittsburgh, Pittsburgh, Pennsylvania; and University of Texas Southwestern Medical Center, Dallas, Texas Background & Aims: Acetaminophen toxicity is the most common cause of acute liver failure (ALF) in the United States and Great Britain, but may be underrecognized in certain settings. Acetaminophen protein adducts are specific biomarkers of drug-related toxicity in animal models and can be measured in tissue or blood samples. Measurement of serum adducts might improve diagnostic accuracy in acute liver failure (ALF) patients. Methods: We measured serum acetaminophen protein adducts using high-pressure liquid chromatography with electrochemical detection in coded sera of 66 patients with ALF collected prospectively at 24 US tertiary referral centers. Samples were included from 20 patients with well-characterized acetaminophen-related acute liver failure, 10 patients with ALF owing to other welldefined causes, 36 patients with ALF of indeterminate etiology, and 15 additional patients without ALF but with known acetaminophen overdose and minimal or no biochemical liver injury. Results: Acetaminophen protein adducts were detected in serum in 100% of known acetaminophen ALF patients and in none of the ALF patients with other defined causes, yielding a sensitivity and specificity of 100%. In daily serial samples, serum adducts decreased in parallel with aminotransferase levels. Seven of 36 (19%) indeterminate cases demonstrated adducts in serum suggesting that acetaminophen toxicity caused or contributed to ALF in these patients. Low adduct levels were present in 2 of 15 patients with acetaminophen overdose without significant liver injury. Conclusions: Measurement of serum acetaminophen protein adducts reliably identified acetaminophen toxicity, and may be a useful diagnostic test for cases lacking historical data or other clinical information. Acetaminophen (N-acetyl-P-aminophenol; paracetamol) is the most widely used analgesic-antipyretic in the United States and Europe, with more than 1 billion tablets sold annually in the United States alone. 1 Present in more than 600 separate products, acetaminophen is generally safe when taken in modest doses ( 4 g/d, the maximum dose recommended on the package label), but is also a potentially lethal dose-dependent toxin. More than 56,000 emergency visits and nearly 500 deaths in the United States each year result from acetaminophen toxicity, owing to either intentional or unintentional overdoses. 1 Indeed, acetaminophen toxicity is currently the most common cause of acute liver failure in the United States and the United Kingdom. 2,3 Intentional acetaminophen overdose is usually easy to diagnose when a history of a 1-time large-dose ingestion is given or elevated acetaminophen blood levels are found. By contrast, in unintentional cases the diagnosis can be elusive for several reasons: patients unaware of the risks associated with acetaminophen may not recall taking the medication; markedly elevated aminotransferase levels are characteristic of acetaminophen toxicity but are not diagnostic; plasma acetaminophen concentrations are often undetectable by the time liver injury has occurred; and unintentional cases present late when encephalopathy may already be present. 4 As a result, many cases of acetaminophen toxicity are unrecognized, leading to a delay in administration of the potentially life-saving antidote, N-acetylcysteine (NAC). 5 Mechanisms of acetaminophen toxicity have been extensively studied (Figure 1). 6 9 Excessive formation of a highly reactive intermediate, N-acetyl-p-benzoquinoneimine (NAPQI), occurs when large doses are taken. In Abbreviations used in this paper: ALF, acute liver failure; CV, coefficient of variation; NAC, N-acetylcysteine; NAPQI, N-acetylparabenzoquinoneimine by the American Gastroenterological Association Institute /06/$32.00 doi: /j.gastro

2 688 DAVERN ET AL GASTROENTEROLOGY Vol. 130, No. 3 Figure 1. Schematic diagram of acetaminophen metabolic pathways. Metabolism of the parent compound is principally via glucuronidation and sulfation, with only a small fraction normally being metabolized via cytochrome 450 (in particular, CYP2E1) to NAPQI, the reactive intermediate form. In the presence of adequate glutathione stores, NAPQI is converted into a harmless, water-soluble conjugate; however, if the capacity to detoxify is exceeded, then NAPQI can covalently bind to cell proteins via cysteine residues. The acetaminophen protein adducts so formed may enter the serum with cell death and release of intracellular proteins. the absence of glutathione, covalent binding of NAPQI to cysteine groups on hepatocyte proteins occurs forming acetaminophen protein adducts that may be released into serum Studies using radiolabeled acetaminophen and anti-acetaminophen antibodies have shown that this covalent binding represents the initial and irreversible step in the development of hepatic toxicity. 12 An accurate and reliable assay that could identify patients with occult acetaminophen-related liver injury would be of significant value to clinicians. Utilizing a highly specific assay employing high-performance liquid chromatography with electrochemical detection (HPLC- EC), 13 we measured acetaminophen protein adducts in serum in patients with acute liver failure (ALF) of differing etiologies to determine the validity of such an assay for detecting acetaminophen hepatotoxicity. We also compared the presence and quantity of the adducts detected in serum samples from patients with known acetaminophen-related ALF, patients with other known causes of ALF, patients with acetaminophen overdoses without significant injury, and those with acute liver failure of indeterminate etiology. Materials and Methods Study Participants Serum and data were prospectively collected from January 1998 to the present at 24 sites participating in the U.S. ALF Study Group on more than 880 adult patients who met criteria for ALF, defined as the presence of coagulopathy (prothrombin time 15 seconds or international normalized ratio 1.5) and hepatic encephalopathy within 26 weeks of first symptoms of illness and without evidence of previous liver

3 March 2006 SERUM ACETAMINOPHEN PROTEIN ADDUCTS 689 Table 1. Demographic and Clinical Characteristics of Patients With Acetaminophen or Non-Acetaminophen Related ALF Acetaminophen cases (n 20) Non-acetaminophen cases (n 10) Characteristic Range Median Range Median Age (y) Sex (% female) Coma grade I IV III I III II APAP level (mg/l) a 10 ALT (IU/L) , AST (IU/L) 59 16, International normalized ratio (INR) Bilirubin (mg/dl) b Creatinine (mg/dl) c Adduct concentration (nmol acetaminophen CYS/mg protein) Transplanted (%) Outcome (% alive) a The acetaminophen levels, which were obtained on 4 of the 10 patients, were 0, 10, 10, and 10 mg/dl. b To convert to mol/l multiply by c To convert to mol/l multiply by disease. 14 Informed consent is obtained from the patients legal next of kin before enrollment, according to the guidelines of local institutional review boards. The etiology of ALF is initially determined at each study center by the local site investigator using generally accepted diagnostic criteria based on history, laboratory values, imaging studies, and, in some cases, histologic examination of the liver. 2 A careful history of acetaminophen ingestion is recorded on the case report forms (CRFs) for each patient, including total dose ingested, type of acetaminophen product taken, and duration of use. At the beginning of the study, we established standard criteria for acetaminophen ingestion to be used by the sites: (1) a history of ingestion of doses exceeding the package labeling ( 4 g/24 hours) within 7 days of presentation; in most cases, the quantities greatly exceeded this value; (2) detection of acetaminophen in plasma of a patient meeting other qualifications for ALF; (3) ALT levels 3500 IU/L, because these levels are commonly observed in acetaminophen toxicity and infrequently or rarely with other causes of ALF. 15 Each case in the present study was confirmed to have had rapid onset disease typical of acetaminophen-induced injury. Data provided on the CRFs were then queried to determine whether additional data could be obtained. Each case of presumed acetaminophen ingestion was then subjected to further evaluation at the central site by a systematic review of the CRF. Site visits by the overall principal investigator of the study (W.M.L.) provided a further level of scrutiny of cases by comparing the original medical record with the CRF to be certain of the diagnosis. In addition to the 20 cases with definite acetaminophen toxicity, 10 patients with ALF were chosen at random representing other well-defined etiologies including 4 with idiosyncratic drug-induced liver injury, 2 with autoimmune hepatitis, 2 acute hepatitis B, 1 acute hepatitis A, and 1 with Wilson disease. In this group, there was no history of concomitant acetaminophen ingestion. ALF was considered to be of indeterminate etiology when careful historical review and extensive clinical, radiographic, and laboratory evaluation (including toxicology screens, serologic markers of viral hepatitis A, B, and C, and antinuclear and anti smooth-muscle antibodies), and pathologic analysis (where available) failed to indicate a defined cause. Outcome data and final diagnosis were recorded on a subsequent CRF 3 weeks after admission, at the time of liver transplantation, or death. Serum Analysis Serum samples collected at admission to study and for 6 additional days were aliquoted at room temperature within 2 hours of collection and stored at 80 C prior to analysis. For the 20 patients with established acetaminophen-related ALF, 17 (85%) had a history of excessive acetaminophen ingestion and 15 had ingested 10 g (median ingestion for the group as a whole: 24 g). In 3 of 4 where doses were unknown, the ALT values were 8000, 14,570, and 18,079 IU/L. Fifteen (75%) had detectable acetaminophen levels and 17 (85%) had serum ALT 3500 IU/L. For this known acetaminophen group, the median total dose ingested was 21 g, and the median ALT value was 5600 IU/L. In addition to the 10 patients with ALF where another diagnosis had been established, 36 consecutive patients with ALF of indeterminate etiology were selected for study. All but 2 of the 66 patients studied had samples available from the day of admission to the ALF study. In 6 patients found to have detectable adducts at admission to study, we studied serial samples to determine the kinetics of serum adduct elimination. Finally, we analyzed sera from 15 patients hospitalized with intentional acetaminophen overdose but minimal or no biochemical evidence of liver injury (and thus no ALF) to determine whether large ingestions with low levels of liver injury were associated with development of serum acetaminophen protein adducts. All patients in this group (median SD age, years; 14 women) were

4 690 DAVERN ET AL GASTROENTEROLOGY Vol. 130, No. 3 admitted to the hospital following intentional acetaminophen overdose for treatment with oral NAC and had daily monitoring of ALT values. The median acetaminophen level at presentation was 108 mg/dl (range, mg/dl). Eight of the 15 were judged to be at risk of toxicity by the Rumack nomogram (equivalent to a serum concentration of APAP of 150 g/ml at 4 hours); 5 presented 24 hours after the overdose; and 2 patients did not meet criteria for treatment, but were nonetheless admitted for treatment based on historical data (dose of acetaminophen ingested). In this patient subset, the median time to the first dose of NAC was 9 hours (range hours) and the median peak value for ALT in these patients was 26 IU/L (range, IU/L); all patients in this group recovered uneventfully. Measurement of Acetaminophen Protein Adducts Serum samples were analyzed for acetaminophen protein adducts (acetaminophen-cysteine) by a high-performance liquid chromatography with electrochemical detection (HPLC-EC) method as previously reported. 13 Briefly, samples were dialyzed, treated with protease, and then precipitated with trichloracetic acid. Following centrifugation, the resulting supernatant was injected onto the HPLC machine using a Model 582 solvent delivery system with 2 Model 5600A CoulArray detectors (ESA, Chelmsford, MA). The range of linearity for this method was mol/l acetaminophen-cysteine in serum. The coefficients of variation (CVs) for the assay were consistently 10% at concentrations of 1.0, 6.0, and 30.0 mol/l acetaminophen-cysteine adducts. Based on the CVs for the standard curve for the assay, the lower limit of quantitation for the assay was defined as 1 mol/l acetaminophen-cysteine and thus, a sample containing 1.0 mol/l acetaminophen-cysteine was defined as positive, and those 1.0 mol/l were considered negative. To standardize each sample for its protein content, concentrations of acetaminophen-cysteine determined by HPLC-EC were then calculated as a fraction of the protein content of the sample (Bradford method), 16 the final concentrations being expressed as nmol acetaminophen-cysteine/mg protein. The summary data for patient subgroups are presented as median and range values. Results Adducts in Patients With Acetaminophen and Non-Acetaminophen ALF Acetaminophen protein adducts were present in all 20 serum samples from patients with acetaminophenrelated toxicity (median, nmol acetaminophencysteine/mg protein; range, ). In contrast, no significant adduct levels were detected in the samples of patients with ALF of other known etiologies (median, nmol acetaminophen-cysteine/mg protein; range, ). Based on these data, the sensitivity and specificity of the adduct assay were 100%. In sera from 15 patients with acute acetaminophen overdose without significant liver injury, the concentration of acetaminophen-protein adducts was detectable but only at very low levels (median, nmol acetaminophen-cysteine/mg protein; range, ). Adducts in Serial Samples From Patients With ALF In patients for whom serial measurements of adduct levels were made, 2 (Figure 2A and B) had measurable acetaminophen concentrations prior to the study, whereas another patient (Figure 2C) had a history of excessive acetaminophen ingestion 3 days prior to presenting with liver failure, but no detectable acetaminophen in serum on admission. In this patient the acetaminophen-cysteine value from the admission blood sample was 2.78 nmol acetaminophen-cysteine/mg protein. The highest levels of adducts in these patients were associated with the highest serum aminotransferase levels observed, although peak levels might have occurred prior to admission to study. The decline of acetaminophencysteine paralleled the resolution of biochemical abnormalities. Assuming that the disappearance of adducts follows first-order kinetics, the half-life of adducts in sera ranged from 17.1 to 30.7 hours in these patients. Adducts in Patients With Indeterminate ALF Analysis of acetaminophen protein adducts in sera disclosed that 7 of 36 (19.4%) indeterminate patients had detectable acetaminophen-cysteine adducts (median value 1.06 nmol acetaminophen-cysteine/mg protein; range ). The values and range measured in this group did not differ significantly from those observed in well-characterized acetaminophen cases (Figure 3). In addition, the rate of decline in serum adduct concentration in indeterminate patients (eg, Figure 2D) was similar to that observed in the patients with ALF from acetaminophen and appeared to parallel the decline in serum aminotransferases. In retrospect, despite lacking a history of excessive acetaminophen ingestion, the adduct-positive indeterminate cases studied (Table 2), displayed clinical features that were similar to those with known acetaminophen toxicity: they were younger, more likely to be female, and had higher aminotransferase and serum creatinine levels, but lower serum bilirubin values than the adductnegative indeterminate ALF group. Although the percentage of patients reporting any exposure to acetaminophen (albeit in the therapeutic range) was higher in the adduct-positive than the adduct-negative group, the per-

5 March 2006 SERUM ACETAMINOPHEN PROTEIN ADDUCTS 691 Figure 2. Acetaminophen CYS adduct levels and ALT levels measured serially from day 1 to 7 in 4 patients with either known acetaminophen toxicity (A C) or indeterminate ALF (D). In each case, the decline in serum adduct concentration approximates that of ALT. On presentation, patients A and B had histories of excessive acetaminophen dosing and detectable serum acetaminophen levels, whereas patient C had a history of excessive dosing but an undetectable acetaminophen level, and patient D had neither history nor an elevated serum acetaminophen level. centage with detectable serum acetaminophen levels was identical in both groups (17%). Empiric treatment with NAC was uncommon in both adduct-positive (14%) and adduct-negative (21%) cases, suggesting that the physicians caring for these patients considered acetaminophen toxicity unlikely. In contrast, 92% of suspected acetaminophen cases were treated with this well-established antidote for acetaminophen toxicity. The overall spontaneous (without transplant) survival rate was substantially lower in the indeterminate group (13%) than in the acetaminophen control group (80%), but was higher for the adduct-positive (29%) compared with the adductnegative (10%) group. Discussion In this study, we evaluated the role of an assay to measure acetaminophen adducts in patients with known acetaminophen toxicity, other causes of acute liver failure, a large number with indeterminate ALF, and a group of patients who sustained an overdose but demonstrated little evidence of liver damage. The HPLC-EC assay identified all 30 test cases of ALF with known diagnoses as acetaminophen related (or not). The temporal profile of serum adducts paralleled that of aminotransferase levels, remaining detectable for up to 7 days after admission to study. More importantly, acetaminophen was also implicated in cases not previously recognized as affected, by the detection of adducts in similar quantities to the known acetaminophen cases in nearly 20% of the indeterminate group. The diagnosis of ALF due to acetaminophen is clear cut when a history of ingestion of an excessive amount of acetaminophen coupled with an elevated serum acetaminophen level is obtained. However, those overdose patients with the most severe injury and the poorest prognoses (typically unintentional cases) present late for medical care when acetaminophen levels are often undetectable and a reliable history may not be obtained owing to altered mentation. 2,4 Moreover, false-positive acetaminophen levels may be observed in the presence of

6 692 DAVERN ET AL GASTROENTEROLOGY Vol. 130, No. 3 Figure 3. Serum levels of acetaminophen CYS adducts in patient groups. (A) Patients with ALF secondary to known acetaminophen overdose. (B) Patients with ALF owing to nonacetaminophen causes. (C) Patients with acetaminophen overdose but no ALF. (D) Patients with ALF of indeterminate etiology and detectable serum adducts. (E) Patients ALF of indeterminate etiology and negative adducts. The boxes represent the 25th 75th IQR and the horizontal line represents the median. The extremes of the population are represented by the endmarks. serum bilirubin levels 10 mg/dl, further impeding an accurate diagnosis. 17,18 NAC is a highly effective antidote for acetaminophen poisoning when given early following ingestion, making an accurate diagnosis of signal importance. 19 It remains uncertain how much benefit may accrue if NAC is given after onset of injury, but it may be effective as late as 72 hours following ingestion. 20 Although NAC may be given empirically when acetaminophen toxicity is a possibility, a rapid and accurate biomarker for acetaminophen injury would allow its use only when appropriate. Extensive time course and dose response studies in animal models have characterized acetaminophen protein adduct formation in acetaminophen toxicity. 6,7 Adduct formation appears in the liver following depletion of glutathione and the appearance of unconjugated NAPQI, the reactive metabolite of acetaminophen (Figure 1). Serum adducts are presumed to be of hepatic origin based on the time course for their appearance relative to their formation in the liver. 21 Thus, acetaminophen protein adducts in serum represent liver injury specific to acetaminophen The presence of adducts in the indeterminate cases does not clarify whether acetaminophen toxicity was the only cause or whether another etiology plus acetaminophen together caused the liver injury; however, the levels of adducts were similar to those found in our known ingestions, and the clinical similarities of the cases (low bilirubin, high ALT levels) indicate that acetaminophen may well have been the sole agent, unrecognized because of encephalopathy or specific denial of the ingestion. Our study has limitations. Foremost, we focused mainly on patients with ALF and have not yet studied sera from patients with therapeutic exposure to acetaminophen or with acute acetaminophen overdose and moderate to severe liver injury without encephalopathy. In patients with acute overdose and early NAC administration, injury was prevented and adducts were found in low concentrations in only 2 of the 15 patients. The Table 2. Demographic and Clinical Characteristics of Patients With Indeterminate ALF Indeterminate Group With Adducts (N 7) Indeterminate Group Without Adducts (N 29) Patient Range Median Range Median Age (y) ALT (IU/L) AST (IU/L) 14,580 11,029 18, , International normalized ratio (INR) Bilirubin (mg/dl) a Creatinine (mg/ dl) b Acetaminophen ND level (mg/l) Adduct concentration (nmol acetaminophen CYS/mg protein) ND, not determined. a To convert to mol/l multiply by b To convert to mol/l multiply by 88.4.

7 March 2006 SERUM ACETAMINOPHEN PROTEIN ADDUCTS 693 finding of minimal amounts of adduct formation in patients without overt liver injury is similar to the results observed in experimental animal models; early treatment with NAC in mice can abort toxicity but small amounts of adducts are still observed in the liver in this setting. 24 The HPLC-EC assay used in this study is labor intensive and not yet available in hospital clinical laboratories. However, refinement of the assay has reduced its turnaround time to 2 3 hours. By confirming the diagnosis of acetaminophen toxicity, the practicing physician may be able to consider and use NAC earlier and may prevent future episodes of intentional or unintentional toxicity, for example, by confronting the patient and arranging psychiatric evaluation if appropriate. Recent data suggest that acetaminophen overdoses are often repeated, making an accurate diagnosis of acetaminophen toxicity essential. 25,26 Identification of a surreptitious acetaminophen overdose may also have implications with regard to liver transplant candidacy in certain situations, and also has important forensic and medicolegal implications. Confirming the diagnosis of acetaminophen hepatotoxicity may preclude further etiologic evaluation, thereby decreasing health care costs; refuting the diagnosis may lead to additional testing. Measurement of adducts might more accurately define the epidemiology of acetaminophen toxicity, leading to public health measures to limit cases. Finally, treating all ALF patients with NAC when a specific diagnosis cannot be determined may be appropriate in light of our finding that unrecognized cases of acetaminophen toxicity regularly occur. In conclusion, we demonstrated that measurement of serum acetaminophen protein adducts represents a sensitive and specific method to confirm or exclude acetaminophen as the cause of severe liver injury in patients with ALF. As such, the assay should help to identify acetaminophen-induced liver injury in situations where historical or laboratory data are lacking. Because occult acetaminophen toxicity appears responsible for a significant proportion of cases of indeterminate ALF, a rapid test for adducts is a worthwhile goal. Until such a test is readily available, use of NAC for patients with ALF where the diagnosis is uncertain, particularly where characteristic low bilirubin and high aminotransferase levels are observed, seems indicated. References 1. Nourjah P, Willey M. Epidemiology of acetaminophen-related overdose. Department of Health and Human Services, Center for Drug Evaluation and Research, Food and Drug Administration, Available: 2. Ostapowicz G, Fontana RJ, Schiodt FV, Larson A, Davern TJ, Han SH, McCashland TM, Shakil AO, Hay JE, Hynan L, Crippin JS, Blei AT, Samueal G, Reisch H, Lee WM. Results of a prospective study of acute liver failure at 17 tertiary care centers in the United States. Ann Intern Med 2002;137: Bernal W. Changing patterns of causation and the use of transplantation in the United Kingdom. Semin Liver Dis 2003;23: Schiodt FV, Rochling FA, Casey DL, Lee WM. Acetaminophen toxicity in an urban county hospital. N Engl J Med 1997;337: Lee WM. Drug-induced hepatotoxicity. N Engl J Med 2003;349: Jollow DJ, Mitchell JR, Potter WZ, Davis DC, Gillette JR, Brodie BB. Acetaminophen-induced hepatic necrosis. II. Role of covalent binding in vivo. J Pharmacol Exp Ther 1973;187: Mitchell JR, Thorgeirsson SS, Potter WZ, Jollow DJ, Keiser H. Acetaminophen-induced hepatic injury: protective role of glutathione in man and rationale for therapy. Clin Pharmacol Ther 1974; 16: Bessems JG, Vermeulen NP. Paracetamol (acetaminophen)-induced toxicity: molecular and biochemical mechanisms, analogues and protective approaches. Crit Rev Toxicol 2001;31: James LP, Mayeux PR, Hinson JA. Acetaminophen-induced hepatotoxicity. Drug Metab Dispos 2003;31: Roberts DW, Pumford NR, Potter DW, Benson RW, Hinson JA. A sensitive immunochemical assay for acetaminophen-protein adducts. J Pharmacol Exp Ther 1987;241: Pumford NR, Hinson JA, Potter DW, Rowland KL, Benson RW, Roberts DW. Immunochemical quantitation of 3-(cystein-S-yl) acetaminophen adducts in serum and liver proteins of acetaminophen-treated mice. J Pharmacol Exp Ther 1989;248: Pumford NR, Hinson JA, Benson RW, Roberts DW. Immunoblot analysis of protein containing 3-(cystein-S-yl) acetaminophen adducts in serum and subcellular liver fractions from acetaminophen-treated mice. Toxicol Appl Pharmacol 1990;104: Muldrew KL, James LP, Coop L, McCullough SS, Hendrickson HP, Hinson HA, Mayeux PR. Determination of acetaminophen-protein adducts in mouse liver and serum and human serum after hepatotoxic doses of acetaminophen using high- performance liquid chromatography with electrochemical detection. Drug Metab Dispos 2002;30: Trey C, Davidson CS. The management of fulminant hepatic failure. Prog Liver Dis 1970;3: Zimmerman HJ, Maddrey WC. Acetaminophen (paracetamol) hepatotoxicity with regular intake of alcohol: analysis of instances of therapeutic misadventure. Hepatology 1995;22: Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976;72: Bertholf RL, Johannsen LM, Bazooband A, Mansouri V. False positive acetaminophen results in a hyperbilirubinemic patient. Clin Chem 2003;49: Polson J, Orsulak P, Wians F, Murray N, Balko J, Fuller D, Rossaro L, McGuire B, Lee WM. Elevated bilirubin may cause false positive acetaminophen levels in hepatitis patients. Hepatology 2004;40:496A. 19. Smilkstein MJ, Knapp GL, Kulig KW, Rumack BH. Efficacy of oral N-acetylcysteine in the treatment of acetaminophen overdose. Analysis of the national multicenter study (1976 to 1985). N Engl J Med 1988;319: Jones AL. Mechanism of action and value of N-acetylcysteine in the treatment of early and late acetaminophen poisoning: a critical review. J Toxicol Clin Toxicol 1998;36:

8 694 DAVERN ET AL GASTROENTEROLOGY Vol. 130, No Roberts DW, Bucci TJ, Benson RW, Warbritton AR, McRae TA, Pumford NR, Hinson JA. Immunohistochemical localization and quantification of the 3-(cystein-S-yl)-acetaminophen protein adduct in acetaminophen hepatotoxicity. Am J Pathol 1991;138: Webster PA, Roberts DW, Benson RW, Kearns GL. Acetaminophen toxicity in children: diagnostic confirmation using a specific antigenic biomarker. J Clin Pharmacol 1996;36: James LP, Farrar HC, Sullivan JE, Givens TE, Kearns GL, Wasserman GS, Walson PD, Hinson JA, Pumford NR. Measurement of acetaminophen-protein adducts in children and adolescents with acetaminophen overdoses. J Clin Pharmacol 2001;41: James LP, McCullough SS, Lamps LW, Hinson JA. Effect of N-acetylcysteine on acetaminophen toxicity in mice: relationship to reactive nitrogen and cytokine formation. Toxicol Sci 2003;75: Rumack BH, Peterson RC, Koch GG, Amara IA. Acetaminophen overdose. 662 cases with evaluation of oral acetylcysteine treatment. Arch Intern Med 1981;23: Hawton K, Harris L, Hall S, Simkin S, Bale E, Bond A. Deliberate self-harm in Oxford : a time of change in patient characteristics. Psychol Med 2003;33: Received June 30, Accepted November 9, Address reprint requests to: Timothy J. Davern II, MD, Division of Gastroenterology and Liver Transplant Program, University of California, San Francisco, 513 Parnassus Avenue, Room S-357, San Francisco, California timothy.davern@csf.edu; fax: Supported by NIDDK: DK R and DK (to LJ), the Jean Roberts and the Rollin and Mary Ella King Funds at the Southwestern Medical Foundation, Dallas, and the Stephen B. Tips Fund of Northwestern Medical Foundation. We are grateful to all the coordinators, patients, and families who make this study possible. Thanks to Sadie McFarlane for expert artwork in Figure 1. The U.S. Acute Liver Failure Study Group who participated in this study includes William M. Lee (Principal Investigator), University of Texas Southwestern Medical Center, Dallas, Texas; Anne Larson, University of Washington, Seattle, Washington; Jeffery S. Crippin, Washington University, St. Louis, Missouri; Timothy J. Davern and Nathan Bass, University of California at San Francisco, San Francisco, California; Sukru Emre, Mt. Sinai Medical Center, New York, New York; Timothy M. McCashland, University of Nebraska, Omaha, Nebraska; J. Eileen Hay, Mayo Clinic, Rochester, Minnesota; Natalie Murray, Baylor University Medical Center, Dallas, Texas; A. Obaid Shakil, University of Pittsburgh, Pittsburgh, Pennsylvania; Andres T. Blei, Northwestern University, Chicago, Illinois; Atif Zaman, Oregon Health Sciences University, Portland, Oregon; Steven H.B. Han, University of California, Los Angeles, Los Angeles, California; Robert J. Fontana, University of Michigan, Ann Arbor, Michigan; Brendan McGuire, University of Alabama at Birmingham, Birmingham, Alabama; Raymond Chung, Massachusetts General Hospital, Boston, Massachusetts; Steven Lobritto, Robert Brown and Michael Schilsky, Columbia-Presbyterian Medical Center, New York, New York; M. Edwyn Harrison, Mayo Clinic Scottsdale, Phoenix, Arizona; Adrian Reuben, Medical University of South Carolina, Charleston, South Carolina; Santiago Munoz, Albert Einstein Medical Center, Philadelphia, Pennsylvania; Rajender Reddy, University of Pennsylvania, Philadelphia, Pennsylvania; R. Todd Stravitz, Virginia Commonwealth University, Richmond, Virginia; Lorenzo Rossaro, University of California, Davis Medical Center, Sacramento, California; and Raj Santayanarana, Mayo Clinic, Jacksonville, Florida.