Combination of Methotrexate and Isoniazid Appears Safe for Rheumatoid Arthritis Patients

The potential for hepatotoxicity is a well-recognized feature of therapy with a number of DMARDs in rheumatoid arthritis (RA), with methotrexate (MTX) being the most common utilized.  Concomitant use of other hepatotoxic agents has the potential to compound the risk of significant liver damage.  Among these, isoniazid (INH) is increasingly used in RA patients receiving MTX with evidence of latent tuberculosis infection (LTBI) or prior active tuberculosis for whom TNF inhibitor therapy is planned.  Here, Mor et al (Ann Rhem Dis 2007 published online Aug 29) explore the safety of this combination of potentially hepatotoxic agents.


Medical records of RA patients on methotrexate with a history of LTBI or untreated active pulmonary TB who were initiating prophylactic therapy with INH and attending clinics at the Bellevue Arthritis Clinic in New York City were examined.  Liver function test (LFT) elevation after starting INH was the primary outcome.  Additionally, outcomes in patients with prior treated active TB who initiated therapy with TNF inhibitors were explored.


Forty-four subjects who had received MTX in combination with INH were analyzed.  Approximately two-thirds were women with a mean age of 51 years.  The average dose of MTX was 17 mg per week and all subjects were receiving supplemental folate.  All subjects receiving INH had a positive PPD skin test (defined as > 5mm of induration after 72 hours).

Most (86%) received INH for LTBI and had normal chest radiographs.  Two-thirds of those treated with INH for LTBI were subsequently treated with TNF inhibitors in addition to INH.  After a mean duration of INH therapy of 7.5 months, 5 (11.3%) were noted to have abnormal LFTs; however, none had LFTs greater than two times the upper limit of normal.  In those who subsequently received TNF inhibitors, none were noted to have reactivation of LTBI after a mean follow-up of 29 months.

Among the 6 subjects with previous untreated active TB with radiographic evidence of pulmonary abnormality who received standard four drug therapy, all initiated therapy with a TNF inhibitor.  After a mean follow-up of 16 months, none had developed active TB.


The combination of INH and MTX appears to be safe when INH is used for the treatment of LTBI.  Subsequent initiation of TNF inhibitors after prophylactic or treatment regimens does not appear to be associated with reactivation of disease.

Editorial Comment:

Although the prevalence of untreated LTBI is generally low in the U.S. population, there are certain groups in which the prevalence is higher.  These include communities with a higher concentration of foreign-born individuals, which tend to be in urban areas.  In addition, as aggressive RA therapy with biologics spreads to Eastern Europe, Asia, Central and South America, and Africa, the safety of TNF inhibition in populations with a high prevalence of LTBI will be an issue of increasing concern.

These results agree with those from the START trial, in which subjects identified with LTBI and treated with INH showed no subsequent reactivation of TB.  However, there were subjects in that trial with false-negative screening for LTBI who were not treated and subsequently developed TB.  Thus, the focus of prevention of LTBI in RA patients treated with TNF inhibitors should focus on screening.  The finding that LFT abnormalities in RA patients receiving the combination of INH and MTX is important, and should discourage practitioners from reducing MTX dosing when initiating INH therapy.  It should be noted that these data do not confirm safety of combining INH with other DMARDs used in RA, with leflunomide being the most likely to result in hepatotoxicity when combined with INH.

Jon Giles, MD

About Jon Giles, MD