Acute liver failure due to iron overdose in an adult.

Abstract: The vast majority of acute iron toxicity cases occur in children less than 5 years of age. Moreover, clinical hepatic injury is uncommon with most symptoms stemming from the intestinal tract (eg, nausea, vomiting, diarrhea). Therefore, physicians, particularly those who do not routinely treat pediatric patients, are often unfamiliar with hepatotoxicity related to iron overdose. Nevertheless, hepatotoxicity caused by acute iron poisoning is associated with a high mortality rate. We report a case of severe hepatic injury in an adult who overdosed on iron tablets with suicidal intent. Tests for other hepatotoxins (eg, acetaminophen), hepatatrophic viruses, and other causes of acute liver injury were negative. Although peak serum iron level (340 [micro]g/dL) was significantly lower than that reported to cause hepatotoxicity (>1,700 [micro]g/dL), rapid and significant elevations in aminotransferases (>4,000 U/L), total bilirubin (5 mg/dL), and prothrombin time (50 seconds) occurred within 48 hours. Treatment with deferoxamine was prompt and followed by empiric N-acetylcysteine once liver injury was apparent. The patient was minimally symptomatic and she eventually had a full recovery.

Key Words: acetylcysteine, deferoxamine, hepatotoxicity, iron intoxication, iron poisoning

**********

Iron is a known hepatotoxin. Chronic iron overload states (eg, hereditary hemochromatosis) are associated with chronic hepatitis and cirrhosis and usually manifest in adulthood. (1,2) Hepatotoxicity associated with acute iron exposure is unusual and largely confined to the pediatric age group, with only sporadic reports of adult cases. Also, hepatotoxicity has been associated only with serum iron levels greater than 1,700 [micro]g/dL. (3)

We report a case of an adult with severe hepatotoxicity after iron overdose despite relatively low serum iron levels. The case reminds physicians that severe iron hepatotoxicity in adults, although rare, can occur. Unlike acetaminophen levels, iron levels do not offer reliable risk stratification for liver injury.

Case Report

An 18-year-old woman was admitted to a community hospital, 4 to 5 hours after a one-time ingestion of approximately 100 ferrous sulfate pills (120 mg of elemental iron per kilogram body weight). The patient had had an altercation with a friend, and the overdose was with suicidal intent. Shortly thereafter, she had nausea and one episode of scant hematemesis with epigastric pain and one episode of watery diarrhea. The emergency medical service personnel brought her to the hospital after being called by the patient.

She reported no other ingestion. She was on no medications except occasional ibuprofen for headaches. She was unclear on why the iron tablets were in her home. The patient reported no use of tobacco, alcohol, recreational drugs, or alternative medicines. Her recent sexual contact was confined to her boyfriend. She had no recent travel history. She was not aware of any drug allergies. There was no family history of liver disease.

At admission, the patient vomited visible pill fragments. Her temperature was 98.9[degrees]F, blood pressure was 122/70 mm Hg, heart rate was 97 beats per minute, and respiratory rate was 22 per minute. Physical examination was within normal limits except for mild epigastric tenderness. Neurologic examination was unremarkable as was her mental status. Laboratory and other test results at the community hospital are as shown in Table 1 (day 1 of hospitalization). Urine and serum drug screens, including acetaminophen levels obtained at admission, were negative. An initial iron level obtained approximately 6 hours after ingestion was 340 [micro]g/dL.

The patient underwent gastric lavage with normal saline and was started on intravenous deferoxamine therapy at 10 mg/kg per hour for a total of 6,000 mg. Despite these interventions, serum alanine aminotransferase (ALT) rose rapidly to greater than 4,000 U/L on hospital day 2. A repeat course of intravenous deferoxamine was given and oral acetylcysteine therapy was started. The patient was given 140 mg of N-acetylcysteine (NAC) per kilogram body weight as a single oral loading dose followed by a dose of 70 mg of NAC per kilogram body weight every 4 hours. When prothrombin time and bilirubin began to increase, she was transferred to our facility in case liver transplantation became necessary.

On transfer to our intensive care unit, her vital signs remained stable. On physical examination, she was fatigued and drowsy but easily aroused and coherent. She had mild icterus and persistent epigastric tenderness with nausea and occasional emesis. She was negative for anti-hepatitis A virus IgM, hepatitis B surface antigen, and hepatitis C virus antibodies. Serum antinuclear antibody, antimitochondrial antibodies, and anti-smooth muscle antibodies were undetectable. Ceruloplasmin level was within normal limits, as was an [alpha]-1 antitrypsin level.

Over the next 4 days, the patient's clinical condition, including the epigastric tenderness and nausea, improved. Her liver enzymes, prothrombin time, and bilirubin also improved (see Table). The trend in serum aminotransferases and prothrombin time is shown in the Figure. She was discharged on hospital day 6. She was seen 1 week later in the clinic. Her laboratory values at this visit included a serum ALT of 362 U/L, international normalized ratio of 0.9, prothrombin time of 13.4 seconds, and the total bilirubin was 0.9 mg/dL. She was asymptomatic and following up with the psychiatry service.

Discussion

This case of acute iron poisoning with associated liver injury is of interest for two reasons: (1) acute iron hepatotoxicity is rare in adults and (2) the measured peak serum level of iron was well below that reported to cause liver injury. These points will be discussed as well as the pathophysiology of iron hepatotoxicity and general treatment measures with comments on the use of NAC in this setting.

Iron overdose typically occurs in children younger than 5 years of age, who mistake the iron pills for candy. (4) Therefore, literature on adult iron hepatotoxicity is scant. A literature search for adult cases of iron poisoning associated with hepatotoxicity yielded only three cases that were older than 15 years of age. (3,5) The oldest case reported is a 30-year-old pregnant woman in the 36th week of gestation. This patient took 5,000 mg of elemental iron (70 mg/kg of body weight), developed a peak ALT of 6,000 U/L, prothrombin time of 43 seconds, and total bilirubin of 18.7 mg/dL. The highest serum iron level measured was 492 [micro]g/dL after approximately 15 hours of ingestion. Deferoxamine was administered, but the patient progressed to hepatic coma and died 2 weeks later. (5) With little literature on and relative rarity of adult cases, the clinician may be taken off guard by iron hepatotoxicity or ignore the possibility beyond the time to consider transfer to a transplant center.

[FIGURE 1 OMITTED]

Even in the pediatric population, severe or fatal hepatotoxicity is quite unusual representing approximately 1% of all iron poisonings. The vast majority of iron overdoses follow a relatively benign course, with full recovery. Most iron toxicity involves the intestinal tract, ranging from nausea and vomiting to diarrhea and hematemesis. The gastrointestinal mucosa followed by the liver is the site of maximum concentration of iron in cases of acute iron poisoning. (6) Occasionally, the gastrointestinal symptoms are severe with shock, but overall only 20% of all poisonings require hospitalization. Perhaps 5% of those hospitalized will die or require liver transplantation.

Once hepatotoxicity occurs, the mortality rate can be as high as 50% or more (3,4); liver transplantation has been reported, but again, primarily in the pediatric setting. (7) Tenenbein et al reviewed 13 cases of acute iron poisoning associated with hepatotoxicity. In all of these 13 cases, hepatotoxicity developed within 48 hours of ingestion. Ten of these patients died and three patients survived, two of whom received chelation therapy, and one patient required liver transplantation. All of these patients were younger than 18 years of age.

The poorer prognosis associated with hepatic involvement in iron poisoning compared with acetaminophen poisoning may be due to different histologic patterns of injury. (3) Acetaminophen poisoning is associated with a more centrilobular involvement in contrast to iron poisoning, which is associated primarily with necrosis of the periportal areas, the site for hepatic regeneration. (8,9) Also, the periportal area receives the blood richest in oxygen and iron, both of which are substrates for free radical generation. Iron catalyzes hydroxyl radical formation, which initiates lipid peroxidation. In an experimental study of acute ferrous sulfate overdose in rabbits, mitochondrial swelling with loss of cristae beginning in the periportal region was observed. (10)

Previous cases reported in the literature suggest that the probability of hepatic injury increases with increasing serum iron levels. The lowest serum iron level associated with liver injury has been reported to be 1,700 [micro]g/dL (range of 1,700 to 21,680 [micro]g/dL, obtained 2 to 72 hours after ingestion). (3) The highest serum iron level measured in our case was only 340 [micro]g/dL, measured approximately 6 hours after reported ingestion. Peak serum iron levels occur approximately 6 hours after acute ingestion. Patient account of ingestion time is often suspect in overdose cases, but visualization of pill fragments on emesis in the emergency room offers some corroboration of the patient history. Indeed, early lavage of ingested pills would be in line with some success in intestinal decontamination and subsequent low serum iron levels. Deferoxamine therapy was prompt, and serum iron levels never again rose above 300 [micro]g/dL.

The reason for such hepatic susceptibility at such a low level of iron is unclear. We found no other biochemical or serologic markers to suggest an underlying chronic liver disorder or cirrhosis. Her liver function--related tests were normal on presentation. She had no physical examination findings suggestive of cirrhosis. The [alpha]-1 antitrypsin phenotype was heterozygous. ME, and the level was 109 mg/dL (normal range, 100 to 200 mg/dL). Although this phenotype confers a slightly higher risk for hepatic and pulmonary disease related to [alpha]-1 antitrypsin, this is only speculative in the absence of a liver biopsy. Most likely there is some component of an idiosyncratic reaction, as has been suggested for cases of acetaminophen toxicity occurring at lower than expected levels. (11,12) Hepatotoxicity based on idiosyncratic reactions is typically not dose-dependent or serum level--dependent and presumably stems from a hypersensitivity type of reaction. Indeed, cases of no hepatotoxicity despite very high serum levels are reported. (13,14) Liver biopsy may have been helpful here but was clinically unjustified. The presence of genetic hemochromatosis could have increased susceptibility as reports of acute iron induced hepatic necrosis have been reported with vitamin C ingestion, the vitamin C causing an acute rise in serum iron. The patient's young age argues against significant pre-event hepatic iron loading caused by genetic hemochromatosis. Unfortunately, no genetic testing was performed in our case.

Treatment of acute iron poisoning consists of early decontamination of the gut and chelation with parenteral deferoxamine therapy. Gastric lavage with a large-bore oral gastric tube using normal saline is recommended. Whole bowel irrigation may benefit cases in which the abdominal film reveals pills beyond the pylorus. (15) Deferoxamine is usually administered as a continuous intravenous infusion at 15 mg/kg per hour. Each 100 mg of deferoxamine binds approximately 9 mg of elemental iron producing the feroxamine-iron complex, which is excreted by the kidney, imparting a vin-rose discoloration to the urine. Disappearance of the vin-rose discoloration of the urine has been considered an end point for chelation therapy; however, normal urine color in the presence of high serum iron levels has been reported. Therefore, the decision to stop therapy should be based on a combination of clinical status, including evidence of end-organ toxicity, urine color, and serum iron levels. (16) Besides chelating iron, deferoxamine may prevent and even reverse lipid peroxidation. (17) Systemic hypotension and pulmonary toxicity are known adverse effects of deferoxamine therapy. Besides bowel decontamination and deferoxamine administration, diligent monitoring and supportive care are imperative. Management of metabolic acidosis and shock with constant monitoring for myocardial, hepatic, and renal involvement is necessary.

In our case, NAC therapy was given once significant hepatotoxicity was evident. There is little evidence to support this intervention. A MEDLINE and PubMed search using the terms NAC, Mucomyst, iron poisoning, iron toxicity, and iron-induced liver failure did not yield any report of the use of NAC in the treatment of iron-induced hepatic dysfunction. Nevertheless, animal studies have proposed a role for NAC in protection against iron-induced lipid peroxidation. (18) Moreover, NAC replenishes glutathione, which may provide general hepatoprotective effects for various causes of hepatic injury. There is evidence to suggest that NAC is able to reduce the severity of ischemia/reperfusion injury and improve postoperative graft function after liver transplantation. (19) Indeed, some centers give NAC for virtually all cases of fulminant hepatic failure. In a study of NAC in non-acetaminophen-induced acute liver failure, four of seven patients survived. (20) A prospective, randomized, double-blinded, placebo-controlled trial of NAC in acute liver failure not caused by acetaminophen overdose is currently being carried out by the US Acute Liver Failure Study Group (21). NAC has relatively few side effects, and the ability to give it intravenously has particular appeal for iron poisoning cases in which nausea and vomiting are often prominent. Whether NAC helped our patient avoid transplantation is unknown. Although it cannot be routinely recommended in all cases of iron poisoning associated hepatotoxicity, it is a therapy to be considered when the prothrombin time and bilirubin are rapidly rising.

Conclusion

Hepatotoxicity caused by iron poisoning remains relatively unusual in pediatric patients and distinctly rare in adults. Unlike acetaminophen levels, serum iron levels do not appear to correlate well with risk of hepatotoxicity. No such "action line" graphs of serum iron levels plotted against time after ingestion are available. There may be some idiosyncratic component to the hepatotoxicity. Our case had by far the lowest reported serum level reported in the literature (340 [micro]g/mL). Therefore, management with intestinal decontamination, deferoxamine, and close monitoring must be considered, even at peak serum levels less than 400 [micro]g/dL. Monitoring of liver enzymes, prothrombin time, bilirubin, and mental status is prudent in all cases. NAC may be considered if severe hepatotoxicity occurs.

Table. Acute liver failure caused by iron overdose: Laboratory data over
the course of hospitalization (a)

Laboratory parameter                          Day 1  Day 2    Day 3

Serum creatinine (normal range,
  0.5 - 1.5 mg/dL)                              0.7      0.7      0.6
Serum bicarbonate (normal range,
  23 - 27 mmol/L)                              20       19       19
Serum albumin (normal range, 3.5 - 4.5 g/dL)    3.9      3.0      3.0
Total bilirubin (normal range,
  0.0 - 1.0 mg/dL)                              0.3      3.1      2.3
Conjugated bilirubin (normal range,
  0.0 - 0.3 mg/dL)                              0.1      1.6      1.2
Alkaline phosphatase (normal range,
  35 - 129 U/L)                                59       87       84
Serum alanine aminotransferase (normal
  range, 4 - 41 U/L)                           14    4,048    3,618
Serum AST (normal range, 4 - 37 U/L)           17    2,417    1,993
PTT (normal range, 24.7 - 35.8 seconds)        34.4     35.3     39.5
Prothrombin time (normal range,
  12.8 - 14.8 seconds)                         14.5     46.3     51.5
International normalized ratio                  1.0      5.0      5.8
Platelets (normal range,
  150 - 400 k/[mm.sup.3])                     225      166      165
White blood cell count (normal range,
  3.5 - 10.5 k/[mm.sup.3])                      8.4      7.8      6.9
Iron (normal range, 37 - 158 [micro]g/dL)     340      136      219

Laboratory parameter                          Day 4    Day 5    Day 6

Serum creatinine (normal range,
  0.5 - 1.5 mg/dL)                                0.4      0.6      0.6
Serum bicarbonate (normal range,
  23 - 27 mmol/L)                                22       19       25
Serum albumin (normal range, 3.5 - 4.5 g/dL)      2.7      3.0      3.2
Total bilirubin (normal range,
  0.0 - 1.0 mg/dL)                                2.3      2.4      2.8
Conjugated bilirubin (normal range,
  0.0 - 0.3 mg/dL)                                1.5      1.6      1.5
Alkaline phosphatase (normal range,
  35 - 129 U/L)                                  85       85      107
Serum alanine aminotransferase (normal
  range, 4 - 41 U/L)                          2,642    1,657    1,438
Serum AST (normal range, 4 - 37 U/L)            630      188       97
PTT (normal range, 24.7 - 35.8 seconds)          33.8     32.9     29.2
Prothrombin time (normal range,
  12.8 - 14.8 seconds)                           38.8     29.0     15.1
International normalized ratio                    4.0      2.7      1.1
Platelets (normal range,
  150 - 400 k/[mm.sup.3])                       164      169      200
White blood cell count (normal range,
  3.5 - 10.5 k/[mm.sup.3])                        9.3      6.8      9.9
Iron (normal range, 37 - 158 [micro]g/dL)       207      162      140

(a) AST, aspartate amino transferase; PTT, partial thromboplastin time.

Accepted October 19, 2004.

References

1. Deugnier YM, Loreal O, Turlin B, et al. Liver pathology in genetic hemochromatosis: a review of 135 homozygous cases and their bioclinical correlations. Gastroenterology 1992;102:2050-2059.

2. Risdon RA, Barry M, Flynn DM. Transfusional iron overload: the relationship between tissue iron concentration and hepatic fibrosis in thalassemia. J Pathol 1975;116:83-95.

3. Tenenbein M. Hepatotoxicity in acute iron poisoning. J Toxicol Clin Toxicol 2001;39:721-726.

4. Zimmerman HJ. Syndromes of Environmental Hepatotoxicity, in Zimmerman HJ (ed): Hepatotoxicity. Philadelphia, Lippincott, Williams & Wilkins, 1999, ed 2, pp 346-347.

5. Olenmark M, Biber B, Dottori O, Rybo G. Fatal iron intoxication in late pregnancy. J Toxicol Clin Toxicol 1987;25:347-359.

6. Pestaner JP, Ishak KG, Mullick FG, Centeno JA. Ferrous sulfate toxicity: a review of autopsy findings. Biol Trace Element Res 1999;69:191-198.

7. Kozaki K, Egawa H, Garcia-Kennedy, R. et al. Hepatic Failure due to massive iron ingestion successfully treated with liver transplantation. Clin Transplant 1995;9:85-87.

8. Luongo MA, Bjornson SS. The liver in ferrous sulfate poisoning, a report of three fatal cases in children and an experimental study. N Engl J Med 1954;251:995-999.

9. Gebhardt R. Metabolic zonation of the liver: regulation and implications for liver function. Pharmacol Ther 1992;53:275-354.

10. Witzleben CL. An electron microscopic study of ferrous sulfate induced liver damage. Am J Pathol 1966;49:1053-1067.

11. Kaplowitz N. Acetaminophen hepatotoxicity: what do we know, what don't we know, and what do we do next? Hepatology 2004;40:23-26.

12. Fabris P, Dalla Palma M, de Lalla F. Idiosyncratic acute hepatitis caused by paracetamol in two patients with melanoma treated with high dose interferon-[alpha]. Ann Intern Med 2001;134:345.

13. Barr DG, Fraser DK. Acute iron poisoning in children, role of chelating agents. Br Med J 1968;1:737-741.

14. Henretig FM, Karl SR, Weintraub WH. Severe iron poisoning treated with enteral and intravenous deferoxamine. Ann Emerg Med 1983;12:306-309.

15. Schauben JL, Augenstein WL, Cox J, Sato R. Iron poisoning: report of three cases and a review of therapeutic intervention. J Emerg Med 1990;8:309-319.

16. Klein-Schwartz W, Oderda GM, Gorman RL, et al. Assessment of management guidelines in acute iron ingestion. Clin Pediatr 1990;29:316-321.

17. Link G, Pinson A, Hershko C. Heart cells in culture: a model of myocardial iron overload and chelation. J Lab Clin Med 1985;106:147-153.

18. Milchak LM, Douglas Bricker J. The effects of glutathione and vitamin E on iron toxicity in isolated rat hepatocytes. Toxicol Lett 2002;126:169-177.

19. Thies JC, Teklote J, Clauer U, et al. The efficacy of N-acetylcysteine as a hepatoprotective agent in liver transplantation. Transpl Int 1998;11(suppl 1):S390-S392.

20. Ben-Ari Z, Vaknin H, Tur-Kaspa R. N-acetylcysteine in acute hepatic failure (non-paracetamol induced). Hepatogastroenterology 2000;47:786-789.

21. National Institutes of Health website. Available at http:www.clinicaltrials.gov/ct/show/NCT00004467?order-1. Accessed on December 28, 2004.

RELATED ARTICLE: Key Points

* Acute liver failure from iron overdose is unusual in children and rare in adults.

* Peak serum levels of iron may not predict hepatotoxicity.

* One should monitor liver enzymes, bilirubin, international normalized ratio, and mental status in all iron overdose cases.

* Deferoxamine and early referral to a liver transplant center remain mainstay interventions.

Sumanth R. Daram, MD, and Paul H. Hayashi, MD

From the Division of Gastroenterology and Hepatology, Department of Internal Medicine, Saint Louis University School of Medicine, St Louis, MO.

The authors have not received any financial grants from any source whatsoever.

Reprint requests to Dr. Paul H. Hayashi, Division of Gastroenterology and Hepatology, Department of Internal Medicine, Saint Louis University School of Medicine, 3635 Vista Avenue, FDT-9S, St Louis, MO 63110. Email: [e-mail]