Published: April 2003
IMMP logo

COX-2 Inhibitors and Hepatotoxicity

Prescriber Update 24(1): 2
April 2003

Dr David Coulter, IMMP Director, 
New Zealand Pharmacovigilance Centre, Dunedin

As with the non-specific non-steroidal anti-inflammatory agents, liver toxicity may occur with the cyclo-oxygenase-2 (COX-2) specific inhibitors, celecoxib and rofecoxib.  They may cause cholestatic, hepatocellular or mixed liver injury; all of which can be severe.  Practitioners should be alert to this possibility, particularly as the onset may be rapid.

Some case reports support causal association

As part of the Intensive Medicines Monitoring Programme (IMMP) monitoring of COX-2 inhibitors, 17 reports of hepatotoxicity have been received.  Most had an onset time of less than three months.  There are three case reports of significant liver injury occurring in association with rofecoxib.

  1. A woman aged 85 was being treated with rofecoxib 6.25mg daily for three weeks for inflammatory arthritis.  Peak enzyme values were ALP 239, ALT 1409 and AST 1545 units per litre indicating hepatocellular liver injury.  The total bilirubin rose to 43 µmol per litre.  No other medicines were listed. The liver function test (LFT) results had been normal prior to taking regular rofecoxib and returned to normal after it was withdrawn.
  2. A man aged 81 with controlled congestive heart failure and maturity onset diabetes was prescribed rofecoxib (dose not stated) for a sore neck.  He became unwell 'almost immediately' and seven days later was admitted to hospital with hepatocellular liver injury. His LFT results were ALP 115, ALT 1111 and bilirubin normal.  The only new medicine prescribed was rofecoxib.  There was no evidence of viral hepatitis.  Two weeks after withdrawal of rofecoxib the ALT level had returned to near normal without change in his other medicines.
  3. A man aged 61 was prescribed rofecoxib (dose not stated) for foot pain.  After about three months he developed severe cholestatic hepatitis, which was confirmed on liver biopsy.  His total bilirubin peaked at 501 µmol/litre.  At the same time he developed acute renal failure. He was taking no other medicines.  Recovery was complete two months after withdrawal of rofecoxib.

Other case reports included other known hepatotoxic medicines

While the clinical details concerning the above case reports were not complete and the investigations reported were not exhaustive, it is probable that these hepatic events were related to rofecoxib.  There were three other reports of similar severity involving celecoxib where the relationship was less clear due to concomitant hepatotoxic medicines including methotrexate and leflunomide.

In addition, there have been eight reports of mild liver function abnormalities with celecoxib and three with rofecoxib.  Two of these patients recovered following withdrawal of the COX-2 inhibitor, but the outcome of the others is unknown.

Hepatotoxicity has been reported infrequently in the literature

A literature search revealed a small number of reports of hepatic reactions to celecoxib1,2,3,4,5 and one with rofecoxib.6 One patient had an allergy to sulphonamides, and this may have been the risk factor that precipitated her cholestatic hepatitis because celecoxib has a sulphonamide moiety.2  The New Zealand data sheet for Celebrexâ„¢ (celecoxib) refers to borderline elevations of LFTs in clinical trials and notable elevations of ALT and AST occurring with no greater frequency than placebo.7  The Vioxxâ„¢ (rofecoxib) data sheet states that hepatic failure, hepatitis and jaundice have been reported, but with no comment about causality.8

Awareness and early recognition improves prognosis

Hepatotoxicity is known to occur infrequently with the non-specific non-steroidal anti-inflammatory agents (NSAIAs).  The IMMP reports of hepatotoxicity with COX-2 inhibitors suggest that this type of reaction is an uncommon class effect of all NSAIAs, both COX-2 specific and non-specific.  Patients using COX-2 inhibitors who have symptoms or signs suggestive of liver dysfunction (including an abnormal liver test result) should have the COX-2 inhibitor discontinued.

Competing interests (author): Unconditional programme grants have been received from various pharmaceutical companies, including Merck Research Laboratories USA.

Correspondence to Dr David Coulter, New Zealand Pharmacovigilance Centre, PO Box 913, Dunedin.

  1. Carrillo-Jimenez R, Nurnberger M. Celecoxib-induced acute pancreatitis and hepatitis: A case report. Arch Intern Med 2000;160:553-554.
  2. Galan MV, Gordon SC, Silverman AL. Celecoxib-induced cholestatic hepatitis [letter]. Ann Intern Med 2001;134:254.
  3. O'Beirne JP, Cairns SR. Cholestatic hepatitis in association with celecoxib. BMJ 2001;323:23.
  4. Nachimuthu S, Volfinzon L, Gopal L. Acute hepatocellular and cholestatic injury in a patient taking celecoxib. Postgrad Med J 2001;77:548-550.
  5. Alegria P, Lebre L, Chagas C. Celecoxib-induced cholestatic hepatotoxicity in a patient with cirrhosis [letter]. Ann Intern Med 2002; 137:E-75.
  6. Huster D, Schubert C, Berr F, et al. Rofecoxib-induced cholestatic hepatitis: treatment with molecular adsorbent recycling system (MARS) [letter]. J Hepatol 2002;37:413-414.
  7. Pharmacia. Celebrex Data Sheet 12 March 2002.
  8. Merck Sharp & Dohme (New Zealand) Limited. Vioxx Data Sheet 3 October 2002.