Current issues of ACP Journal Club are published in Annals of Internal Medicine


Low-dose methotrexate did not help taper steroid dosage, reduce symptoms, or improve pulmonary function in severe asthma

ACP J Club. 1991 July-Aug;115:3. doi:10.7326/ACPJC-1991-115-1-003

Source Citation

Erzurum SC, Leff JA, Cochran JE, et al. Lack of benefit of methotrexate in severe, steroid-dependent asthma. A double-blind, placebo-controlled study. Ann Intern Med. 1991 Mar 1;114:353-60. [PubMed ID: 1992876]



To assess the effect of methotrexate on corticosteroid requirements, symptoms, and pulmonary function in steroid-dependent, severely asthmatic patients.


Randomized, double-blind, placebo-controlled trial of 12 weeks' duration.


An asthma-care outpatient clinic.


19 of 46 patients referred with steroid-dependent asthma met American Thoracic Society criteria for asthma and had been treated with steroids for > 1 year with inability to taper prednisone doses below 15 mg/d. Current treatment included theophylline and regular, inhaled β-adrenergic agonists with current or past use of inhaled steroids and cromolyn. Patients with comorbid illnesses or histories of noncompliance were excluded. 17 patients completed the trial (1 patient was excluded for noncompliance, and 1 dropped out during steroid tapering with a severe exacerbation).


Intramuscular methotrexate (n = 8) or placebo (n = 9) was given weekly. The dose of methotrexate was 5 mg in week 1, 10 mg in week 2, then 15 mg weekly. Other medications were continued unchanged until steroid tapering began in week 7. A reduction of 5 mg of prednisone/wk was sought from weeks 7 to 12. Symptoms and peak expiratory flow rates were used to judge tolerance for reduction.

Main outcome measures

Mean steroid dose reduction, changes in daily symptom scores, and pulmonary function tests.

Main results

The methotrexate group decreased their steroid doses by a mean of 8.0 mg/d (40%, 95% CI 25% to 54%, P = 0.001) compared with 7.2 mg/d (40%, CI 18% to 62%, P = 0.003) for the placebo group, showing no difference in ability to taper. No differences existed between groups in mean steroid doses during the last 6 weeks of the trial (12.8 mg/d for the methotrexate group vs 12.6 mg/d for the placebo group), proportion of exacerbations requiring a temporary increase in steroids (P > 0.2), daily symptom scores, pulmonary function, or increased use of β-adrenergic agonists. Adverse effects (mostly gastrointestinal) occurred with similar frequency in the 2 groups (6 patients [67%] receiving methotrexate vs 5 patients [56%] receiving placebo). {Absolute risk difference 11%, 95% CI -32% to 50%, P = 0.6.}*


Weekly injections of low-dose methotrexate made no difference in the ability of severely asthmatic patients to taper steroid dosage, reduce symptoms, or improve pulmonary function.

Sources of funding: The National Jewish Center for Immunology and Respiratory Medicine and the National Institutes of Health.

Address for article reprint: Dr. G.R. Cott, Department of Medicine, National Jewish Center for Immunology and Respiratory Medicine, 1400 Jackson Street, Room K613, Denver, CO 80206, USA.

*Numbers calculated from data in article.


This study of patients with steroid-dependent asthma failed to find a significant benefit of 12 weeks of methotrexate treatment when compared with placebo. This result appears to refute the original observations of Mullarkey and colleagues, which suggested some benefit from methotrexate in this patient group (1). Several issues need to be considered, however, before methotrexate treatment is abandoned. All studies on this matter have had small sample sizes. Also, patients with severe asthma are not homogeneous with respect to environmental factors triggering their asthma, severity of airflow obstruction, degree of airway hyper-responsiveness, ability to perceive symptoms, or response to therapy. Deciding on the effectiveness of any single therapeutic intervention with such small sample sizes is therefore risky. Mullarkey and colleagues (1) found that methotrexate reduced steroid requirements by 40%; in this study by Erzurum and colleagues, both the placebo and methotrexate groups reduced their steroid needs by 40%, which may reflect rigid compliance of patients to conventional therapies, with frequent office visits and aggressive reduction of steroid dose.

Given the inherent difficulties in obtaining a large sample size, it is unlikely that any single-center study will provide the definitive answer about the effectiveness of methotrexate. Perhaps a meta-analysis of several smaller studies or a larger multicenter study will do so.

Paul O'Byrne , MD
McMaster UniversityHamilton, Ontario, Canada


1. Mullarkey MF, Blumenstein BA, Andrade WP, et al. Methotrexate in the treatment of corticosteroid-dependent asthma. A double-blind crossover study. N Engl J Med. 1988;318:603-7.