Article Text

Download PDFPDF

High incidence of hepatotoxicity of isoniazid treatment for tuberculosis chemoprophylaxis in patients with rheumatoid arthritis treated with methotrexate or sulfasalazine and anti-tumour necrosis factor inhibitors
  1. J Vanhoof1,
  2. S Landewe1,
  3. E Van Wijngaerden2,
  4. P Geusens3
  1. 1Clinical Research Centre for Bone and Joint Diseases, Biomedical Research Institute, Limburg University Centre, Diepenbeek, Belgium
  2. 2Department of Internal Medicine/Infectious Diseases, University Hospitals Leuven, Leuven, Belgium
  3. 3Department of Rheumatology, Academisch Ziekenhuis, Maastricht, The Netherlands
  1. Correspondence to:
    Dr P Geusens
    Biomedical Research Institute, Limburgs Universitair Centrum, 3590, Diepenbeek, Belgium;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Tumour necrosis factor (TNF) inhibition is a major breakthrough in the treatment of rheumatoid arthritis (RA). Infliximab, etanercept, and adalimumab are available TNF inhibitors that have been used and investigated extensively in RA.1–3 Reactivation of latent tuberculosis (TB) during anti-TNF treatment has been reported and therefore recommendations are made to screen patients for latent TB before starting TNF blocking agents.4

However, treatment with isoniazid (INH) can result in severe adverse effects. The two most important untoward effects of INH treatment are hepatitis and peripheral neuropathy. Peripheral neuritis associated with INH treatment can be reduced by the prophylactic administration of vitamin B6 (pyridoxine) 250 mg/week. INH associated hepatotoxicity can be severe and life threatening, with a death rate of 10%.5 Age seems to be the most important factor in determining the risk of important liver injury. Also a contributory role of alcohol consumption has been noted. The characteristic pathological process is bridging and multilobular necrosis. The severity of damage tends to increase if the drug is continued after symptoms of hepatic dysfunction have appeared. A reactive metabolite of acetylhydrazine, a metabolite of INH, may be responsible for liver injury, and patients who are rapid acetylators would be prone to such injury.

During the past year, the Biomedical Research Institute participated in several double blind randomised clinical trials to evaluate the effect and safety of treatment with TNF inhibitors in RA with methotrexate or sulfasalazine as the concomitant disease modifying drug (DMARD) treatment in 88 patients (table 1). Of these patients, 11 had a positive purified protein derivative (PPD) skin test (skin induration more than 5 mm) and two other patients had chest x ray abnormalities, possibly suggesting latent TB. In eight patients prophylactic treatment with INH 300 mg/day (+ pyridoxine 250 mg/week) was started. The other five patients were not given INH as the trial was ended for various reasons, shortly after performing the PPD skin test and chest x ray; one patient refused INH treatment. In four of eight patients treated with INH (and pyridoxine) at standard doses mild to severe hepatic dysfunction was noted. In three of these four patients discontinuation of INH resulted in a normalisation of the transaminase activity. The incidence of liver dysfunction in our patients considerably exceeds the reported incidence of hepatotoxicity of INH treatment.5–7 Interference with DMARD treatment, such as methotrexate and sulfasalazine, is probably the major reason for hepatotoxicity. In one patient excessive alcohol consumption might have had a contributory role, but it has to be noted that in this patient liver dysfunction was not seen before starting INH treatment. These results already justify a higher awareness of this adverse effect during treatment with tuberculosis chemoprophylaxis in patients with RA treated with sulfasalazine or methotrexate with or without anti-TNF treatment.8

The American Thoracic Society recommends that serum concentrations of aspartate aminotransferase and alanine aminotransferase are determined at baseline in patients over 35 years of age who are receiving INH for chemoprophylaxis, with monthly determinations thereafter. What to recommend if INH has to be stopped remains unclear; a rifampin plus pyrazinamide combination cannot be recommended because its hepatotoxicity rate is even higher than for INH treatment. A possible solution may be rifampin alone for four months after resolution of INH hepatotoxicity.5,7

Since the anti-TNF era, rheumatologists should take special care with the use and monitoring of long term INH treatment in combination with potentially hepatotoxic drugs such as sulfasalazine and methotrexate. Further studies and reports of monitoring and observation of hepatotoxicity of INH treatment in RA patients concomitantly treated with DMARDs during anti-TNF treatment are mandatory.8

Table 1

Characteristics of the total patient group and of patients developing liver dysfunction