Susan C. Nicholson, M.D., C. Douglas Webb, Ph.D., and Robert C. Moellering, Jr., M.D.

Recently, the case history of a 44-year-old woman who experienced acute hepatitis subsequent to therapy for chronic sinusitis was reviewed. The patient sequentially was administered clarithromycin, levofloxacin, amoxicillin-clavulanate, and gatifloxacin. Her adverse events were attributed definitively to gatifloxacin, a surprising conclusion because many other possible causes of hepatitis existed in this case. Not ruled out as potential causes of the clinical and laboratory adverse events were hepatitis other than hepatitis A or B. Other antimicrobials administered were dismissed. In particular, extended treatment with amoxicillin-clavulanate has been clearly linked to hepatotoxic effects that may occur long after therapy begins. Thus,

while we agree that physicians must be aware of the potential for antimicrobial hepatotoxicity, we believe that this case study is not a solidly documented case of hepatitis attributable to gatifloxacin and overlooks other possible causes of acute hepatitis of which prescribers should be aware. (Pharmacotherapy 2002;22(6):794–797)

In a case study recently published in this journal,1 Drs. Henann and Zambie described hepatotoxic reactions in a 44-year-old woman with chronic sinusitis who was treated with two doses of clarithromycin 500 mg, levofloxacin 500 mg once/day for 10 days, amoxicillin 875 mg–clavulanate 125 mg twice/day for 35 days, and gatifloxacin 400 mg once/day for 6 days. The patient also took methscopolamine

2.5 mg–pseudoephedrine 120 mg, methscopolamine 2.5 mg–chlorpheniramine 8 mg, and a

homeopathic remedy that contained colloidal silver. Epigastric pain, nausea, and vomiting

occurred after 5 days of gatifloxacin, which was discontinued on day 7. Pain, lethargy, and nausea persisted. Three days after the last gatifloxacin dose, the patient was hospitalized with ultrasonographic evidence of mild cholelithiasis and elevated levels of total bilirubin (4.1 mg/dl), aspartate aminotransferase

(259 U/L), alanine aminotransferase (669 U/L), and alkaline phosphatase (243 U/L).1 Biliary

obstruction, intrahepatic duct abnormality, or hepatitis A or B infection was not evident. A

percutaneous liver biopsy showed eosinophilic infiltrates consistent with acute drug-induced

hepatitis. The patient remained weak, complained of right upper quadrant pain, and

became jaundiced. Liver function test results remained elevated 19 days after hospitalization

and returned to normal levels by 71 days. In light of the temporal relationship between

gatifloxacin treatment and the onset of adverse events, the low hepatotoxic potential of the

concomitant drugs, and the lack of any other readily apparent reason, gatifloxacin was

implicated as the likely cause of the acute hepatitis. 1 Although the data presented warrants including gatifloxacin-induced hepatitis as a possibility in this case, other information about the patient may call into question its definitiveness. For example, the authors ruled out viral infection as a cause for the patient’s symptoms because she had negative results on tests for hepatitis B surface antigen and antibody, hepatitis B core antibody, and hepatitis A antibody IgM. However, in the United States, approximately 20% of cases of acute hepatitis are due to hepatitis C virus (HCV), which has an incubation period of 7 weeks (range 4–20 wks) and is symptomatic in about one third of patients. 2 Serum transaminase levels can rise dramatically in acute HCV infection, as they did in the patient whose case was reviewed.2 Diagnosis of acute HCV infection on the basis of serologic markers is difficult because anti-HCV antibodies may never appear; however, viral RNA may be detected by polymerase chain reaction within days of infection.3 Moreover, hepatitis C can cause a wide range of hepatic histopathologic abnormalities, including inflammatory cell infiltrates (lymphocytes, eosinophils, neutrophils) making acute disease difficult to distinguish from

drug-induced hepatotoxic reactions based on biopsy results. Although none of the data other than the transaminase elevations indicate acute HCV infection, none appear to rule out this diagnosis. Acute hepatitis may be associated with hepatitis E infection and may follow infection with Epstein-Barr virus, cytomegalovirus, and herpes simplex virus—none of which were ruled out in this case.5, 6

The concomitant drugs were exculpated as potential causes for the observed hepatotoxicity because the patient apparently had been taking them well before the event without any ill effect. Antibiotic-associated hepatotoxicity, although its incidence is rare at approximately 1/10,000 individuals, has been reported for all of the antibiotics taken by the patient.7 Hepatotoxicity with cholestatic or mixed biologic characteristics

may occur 1–6 weeks after the initiation of clarithromycin, with liver function values typically returning to normal within 3 months after drug withdrawal.7–9 Levofloxacin also can cause hepatotoxic effects.10 Amoxicillin-clavulanate, the antibiotic administered immediately before gatifloxacin, likewise has

hepatotoxic potential.7 Typical symptoms of jaundice and pruritus usually occur 2–4 weeks after the start of oral or intravenous treatment,7 and a long course of therapy significantly increases the risk of this adverse event.11 For this reason, administration of amoxicillin-clavulanate for more than 10 days is not recommended.7 The reviewed patient received amoxicillin-clavulanate for 35 days just before starting gatifloxacin. Previous treatment with amoxicillin-clavulanate without incident does not lessen the possibility that this combination agent caused the hepatitis. Concern over quinolone-associated hepatotoxicity is prompted by the severe liver damage noted in patients treated with trovafloxacin.12

Nevertheless, it is uncertain whether the adverse hepatic effects of trovafloxacin are particular to this drug or represent an emerging class effect.13 Evidence to date supports a unique association of trova-floxacin

with severe hepatotoxicity at a frequency greater than that of the main quinolones in use.12–16

Trovafloxacin-associated hepatotoxicity may be due to hepatic metabolism that results in the formation of protein-adduct neoantigens that, in turn, induce a hypersensitivity reaction in the liver.14 Indeed, hepatotoxicity most often is observed for agents that are metabolized extensively in the liver.15, 16 Unlike trovafloxacin, gatifloxacin undergoes relatively little hepatic metabolism (< 1%), and more than 80% of the

drug is excreted unchanged in the urine.17 Reports of hepatitis with newer fluoroquinolones have varied. Sixteen cases of hepatitis attributed to moxifloxacin have been reported to the Food and Drug Administration’s Adverse Event Reporting System (AERS) as of December 14, 2001. 18 A similar number of patients with hepatitis who were receiving gatifloxacin has been reported to AERS. Insufficient data are

available through AERS to determine the relationship between drug and adverse event, and some effort is required to correct for possible duplication of reports. The incidence of hepatitis with levofloxacin has been reported at 1/100,000 patients in Japan and 1/109,000 patients worldwide.10 Case reports of hepatitis have been published for the older fluoroquinolones including ciprofloxacin, ofloxacin, and norfloxacin.1, 19–22

A single, well-documented case of reversible gatifloxacin-associated hepatitis was reported in a postmarketing study of 15,625 patients and is the only case reported to date in more than 44,000 patients receiving gatifloxacin in clinical trials.23 The case reported by Drs. Henann and Zambie 1 underscores the importance of careful monitoring of hepatic function in patients undergoing prolonged and complicated antimicrobial therapy. This is certainly sage advice in light of the common association of nearly all classes of antimicrobials with rare chemical-induced hepatotoxicity.7 Their review led to two conclusions: the hepatotoxic reaction probably was due to gatifloxacin, and hepatitis may occur with gatifloxacin.

Although gatifloxacin may have been involved in this patient’s hepatocellular dysfunction, clear

culpability cannot be established for two reasons. First, all possible viral etiologies of hepatitis were

not ruled out. Second, although amoxicillin-clavulanate was dismissed as the agent causing hepatitis, it is a highly likely cause in this case, particularly because it was administered for a long duration (35 days) temporally related to clinical illness. We respectfully suggest that the hepatitis case reported by Drs. Henann and Zambie 1 was confounded by several factors and thus does not represent a solidly documented

case of gatifloxacin-induced hepatitis. Prescribers should be mindful of the possibility of drug-induced

hepatitis with nearly all classes of antibiotics (including fluoroquinolones), which can occur as often as 1/10,000 patients.7 In addition, they should rule out viral hepatitis and consider noninfectious causes in any patient with suspected drug-induced hepatitis.

References

1. Henann NE, Zambie MF. Gatifloxacin-associated acute

hepatitis. Pharmacotherapy 2001;21:1579–82.

2. Hoofnagle JH. Hepatitis C: the clinical spectrum of disease.

Hepatology 1997;26(suppl 1):15S–20.

3. Schmilovitz-Weiss H, Levy M, Thompson N, Dusheiko G.

Viral markers in the treatment of hepatitis B and C. Gut

1993;34(suppl):S26–35.

4. Goodman ZD, Ishak KG. Histopathology of hepatitis C virus

infection. Semin Liver Dis 1995;15:70–81.

5. Krawczynski K, Aggarwal R, Kamili S. Hepatitis E. Infect Dis

Clin North Am 2000;14:669–87.

6. Carey WD, Patel G. Viral hepatitis in the 1990s, part III:

hepatitis C, hepatitis E, and other viruses. Cleve Clin J Med

1992;59:595–601.

7. Vial T, Biour M, Descotes J, Trepo C. Antibiotic-associated

hepatitis: update from 1990. Ann Pharmacother

1997;31:204–20.

8. Wallace RJ, Brown BA, Griffith DE. Drug intolerance to high-dose

clarithromycin among elderly patients. Diagn Microbiol

Infect Dis 1993;16:215–21.

9. Yew WW, Chau CH, Lee J, Leung CW. Cholestatic hepatitis in

a patient who received clarithromycin therapy for

Mycobacterium chelonae lung infection. Clin Infect Dis

1994;18:1025–6.

10. Yagawa K. Latest industry information on the safety profile of

levofloxacin in Japan. Chemotherapy 2001;47(suppl 3):38–43.

11. Thomson JA, Fairley CK, Ugoni AM, et al. Risk factors for the

development of amoxicillin-clavulanic acid associated jaundice.

Med J Aust 1995;162:638–40.

12. Bertino J Jr, Fish D. The safety profile of the fluoroquinolones.

Clin Ther 2000;22:798–817.

13. Ball P, Mandell L, Niki Y, Tillotson G. Comparative tolerability

of the newer fluoroquinolone antimicrobials. Drug Saf

1999;21:407–21.

14. Suchard J. Review: wherefore withdrawal? The science behind

recent drug withdrawals and warnings. Int J Med Toxicol

2001;4:15. Available from http://www.ijmt.net/4_2/4_2_15

.htm. Accessed December 16, 2001.

15. Lee WM. Drug-induced hepatotoxicity. N Engl J Med

1995;333:1118–27.

16. Owens RC, Ambrose PG. Clinical use of the fluoroquinolones.

Med Clin North Am 2000;84:1447–69.

17. Grasela DM. Clinical pharmacology of gatifloxacin: a new

fluoroquinolone. Clin Infect Dis 2000;31(suppl 2):S51–8.

18. Bayer Corporation. Avelox safety profile. Available from

http://www.avelox.com/Avelox/avx__iv__ safety.htm. Accessed

January 30, 2002.

19. Jones SF, Smith RH. Quinolones may induce hepatitis. Br Med

J 1997;314:869.

20. Hautekeete ML, Kockx MM, Naegels S, Holvoet JK, Hubens

H, Kloppel G. Cholestatic hepatitis related to quinolones: a

report of two cases. J Hepatol 1995;23:759–60.

21. Villeneuve JP, Davies C, Cote J. Suspected ciprofloxacin-induced

hepatotoxicity. Ann Pharmacother 1995;29:257–9.

22. Blum A. Ofloxacin-induced acute severe hepatitis [letter].

South Med J 1991;84:1158.

23. Von Seggern K, Russo R, Wikler MA. A novel approach to

postmarketing surveillance: the Tequin clinical experience

study [abstr]. In: Program and abstracts of the 40th

interscience conference on antimicrobial agents and

chemotherapy. Washington, D.C.: American Society for

Microbiology; 2000:468.

Authors’ Reply

We appreciate the comments from Drs. Nicholson, Webb, and Moellering; however, several points concerning their views need to be addressed. First, the possibility of hepatitis C virus causing this adverse event requires clarification. Hepatitis C is generally mild, with less than 25% of infected patients eveloping jaundice.1 Up to 90% of cases progress to persistent infection, with the majority of them developing chronic hepatitis.2 Those at risk for hepatitis C include intravenous drug users; health care workers;

individuals receiving blood products, hemo-dialysis, or tattoos; and people involved in high-risk

sexual behavior.1 In our case report,3 the patient did not have any risk factors, she developed jaundice with significant symptoms, and to date, she has shown no evidence of persistent or chronic manifestations of hepatitis. All liver function tests returned to the normal range 2.5 months after the onset of hepatitis and

remained within range at both 6- and 12-month follow-up. The patient’s symptoms also abated 5–6 weeks after the adverse event and did not return.

Second, hepatitis E was not addressed as it is endemic in Africa, Southeast and Central Asia, Mexico, and Central and South America.1 The patient had never visited these areas nor had any contact with anyone who has. Third, we did consider the associated hepatotoxicity with amoxicillin-clavulanate. Although we agree that the possibility does exist for implicating amoxicillin-clavulanate with this adverse event, that possibility certainly is lessened by the fact that the patient had taken the drug without any adverse effects on at least seven previous occasions. Moreover, she received a 10- day course of amoxicillin-clavulanate 6 months after the adverse effect, with no ill effects. The time frame in which the adverse event occurred—over 6 weeks (43 days) after the use of amoxicillin-clavulanate—was not consistent with the usual 2–4 weeks that has been observed but is consistent with gatifloxacin administration. Fourth, in their comments regarding the Adverse Event Reporting System (AERS), the authors state that "insufficient data are available through AERS to determine the relationship between drug and adverse event, and some effort

is required to correct for possible duplication of reports." In our review of the Food and Drug Administration (FDA) data, duplicate reports were identified and excluded. This data base, although not without flaws, serves as the best postmarketing account of reported adverse drug reactions. In spite of the authors’ cited shortcomings of the system, they use it to report the number of cases of hepatitis attributed to

moxifloxacin. Finally, the authors stated that our conclusion definitively attributed gatifloxacin as the cause of the reported adverse event. This was never stated nor implied in our report. Our conclusion was that the hepatotoxicity was probably due to gatifloxacin based on the time frame of onset and resolution of the event, and the assessment of other possible causes. Clear culpability was never established since drug rechallenge was not feasible. In addition, the commentary states that amoxicillin-clavulanate is a highly likely cause of the reaction in this case. There is no basis for their conclusion particularly when the patient

had received this agent before and after the acute hepatitis with no adverse effects. We reiterate our conclusion that there was a probable association between gatifloxacin and acute hepatitis in our reported case and that other cases have been reported to the FDA. The FDA issued label changes for gatifloxacin in

October 2001, reflecting the addition of hepatitis and several other adverse reactions to the agent’s

labeling. Clinicians should be mindful of this possible reaction.

References

1. Sharara AI, Hunt CM, Hamilton JD. Hepatitis C. Ann Intern

Med 1996;125:658–68.

2. Koziel MJ. Immunology of viral hepatitis. Am J Med

1996;100:98–109.

3. Henann NE, Zambie MF. Gatifloxacin-associated acute

hepatitis. Pharmacotherapy 2001;21:1579–82.

Neil E Henann, Pharm.D.

Michael F. Zambie, M.D.

The drug manufacturers rebuttal to this report is as follows:

Pharmacotherapy. 2002 Jun;22(6):794-6; discussion 796-7. Related Articles, Links

Comment on:
Pharmacotherapy. 2001 Dec;21(12):1579-82.

Antimicrobial-associated acute hepatitis.

Nicholson SC, Webb CD, Moellering RC Jr.

Infectious Diseases, Bristol-Myers Squibb, Plainsboro, New Jersey 08536, USA. Susan.Nicholson@BMS.com

Recently, the case history of a 44-year-old woman who experienced acute hepatitis subsequent to therapy for chronic sinusitis was reviewed. The patient sequentially was administered clarithromycin, levofloxacin, amoxicillin-clavulanate, and gatifloxacin. Her adverse events were attributed definitively to gatifloxacin, a surprising conclusion because many other possible causes of hepatitis existed in this case. Not ruled out as potential causes of the clinical and laboratory adverse events were hepatitis other than hepatitis A or B. Other antimicrobials administered were dismissed. In particular, extended treatment with amoxicillin-clavulanate has been clearly linked to hepatotoxic effects that may occur long after therapy begins. Thus, while we agree that physicians must be aware of the potential for antimicrobial hepatotoxicity, we believe that this case study is not a solidly documented case of hepatitis attributable to gatifloxacin and overlooks other possible causes of acute hepatitis of which prescribers should be aware.