Skip navigation
Acute Pain | Chronic Pain | General

Second-line drugs for pain relief in rheumatoid arthritis, with toxicity analysis

Clinical bottom line:

We consider two systematic reviews here: Based on Capell et al, there does not appear to be any difference in the efficacy of gold, penicillamine, sulphasalazine or auranofin in the treatment of rheumatoid arthritis over a 12 month period. The Felson et al review considered more drugs, had a more comprehensive search strategy, with a larger dataset, but no pain outcome. Based on this review, in an efficacy/toxicity trade-off, methotrexate and antimalarial drugs had the highest efficacy relative to toxicity. Sulphasalazine also performed well, although was slightly more toxic. The least desirable drugs were intra-articular gold and auranofin.

Second-line or disease modifying drugs are widely used in the management of active rheumatoid arthritis, with a tendency for these drugs to be used relatively early on in the disease. One difficulty with long-term treatments in particular is the balance between efficacy and adverse effects.

Systematic review

Capell HA, Porter DR, Madhok R, Hunter JA. Second line (disease modifying) treatment in rheumatoid arthritis: which drug for which patient? Ann Rheum Dis. 1993; 52: 423-8.

With a summary of adverse effects data from:

Felson DT, Anderson JJ, Meenan RF. Use of short-term efficacy/toxicity tradeoffs to select second-line drugs in rheumatoid arthritis. Arthritis Rheum. 1992; 35: 1117-1125.

Inclusion criteria were of gold (intra-muscular), penicillamine, sulphasalazine or auranofin in patients with rheumatoid arthritis, active synovitis; rheumatoid arthritis inadequately controlled by non-steroidal anti-inflammatory drugs; no concomitant systemic corticosteroid use, maximum of three intra-articular steroid injection per six month period.

Included trials were those carried out by the review team, and does not represent the world literature in this field. Pain was measured using a visual analogue rating scale during assessment visits.

Reviewers assessed efficacy by comparing baseline scores with three, six and 12 month scores within each group. Relative efficacy was assessed by comparing each time point across all groups. These comparisons did not include dropouts. Placebo comparisons were not made.


Intra-muscular gold

Doses were 50 mg/week until a response was obtained, at which point injection frequency was reduced. Dropout rate was 32% (3 months), and 41% (6 months).

Of the 500 patients receiving gold, a proportion received gold plus hydroxychloroquine between six and 12 months. Reviewers state that the hydroxychloroquine patients did not alter outcome, and therefore included these patients in the analysis. The median percentage improvement from baseline in pain scores at three, six and 12 months was 23%, 32% and 33% respectively.


Doses were 125 mg/day, increasing by 125 mg each month until a response was obtained or adverse effects supervened or maximum dose of 1000 mg/day was reached. Dropout rate was 18% (3 months), and 36% (6 months). The median percentage improvement from baseline in pain scores at three, six and 12 months was 22%, 30% and 31% respectively.


Doses of enteric coated sulphasalazine were 0.5 mg/day until target dose of 40 mg/kg was reached or adverse events necessitated dose adjustment. There was also one dose-ranging trial. Dropout rate was 29% (3 months), and 39% (6 months). The median percentage improvement from baseline in pain scores at three, six and 12 months was 16%, 31% and 31% respectively.


Doses were 3 mg twice daily, increasing after six months to a maximum dose of 3 mg three times daily if necessary. Dropout rate was 31% (3 months), and 41% (6 months). The median percentage improvement from baseline in pain scores at three, six and 12 months was 11%, 28% and 35% respectively.

Overall, there were significant improvements over time on all measures including pain, and no one treatment was significantly better than another. Neither gender nor age had an influence on the response to treatment, but patients with a longer disease duration had a greater tendency to stop treatment.

Adverse effects

Felson et al plotted efficacy (a composite measure of improvement in tender joint count, grip strength and erythrocyte sedimentation rate) against toxicity (drop out rates due to drug toxicity, with each adverse effect weighted for seriousness - e.g. nausea less serious than organ toxicity). Over 6,500 patients from 79 trials were included in the analysis, and outcomes were taken between 12 months and one year.

In summary, using the composite efficacy measure and two further efficacy measures, the most effective drugs were methotrexate and sulphasalazine and penicillamine, and the least effective were antimalarial, and azathioprine. Using several measures of toxicity, this appeared to be highest with intra-articular gold and penicillamine, and the least toxic were antimalarial drugs, methotrexate and auranofin.

When different measures efficacy and toxicity were plotted against each other, the most effective and least toxic drugs were antimalarial drugs and methotrexate. Methotrexate was consistently one of the strongest drugs and had the most benign toxicity profile of the more efficacious drugs. The antimalarial drugs scored well, mainly due to relatively low toxicity rather than very high efficacy. sulphasalazine also scored relatively well, with good efficacy and modest toxicity. Intra-articular gold scored as the most toxic drug, and auranofin was one of the least effective.

Further reference

The following review is a preliminary meta-analysis of the Felson et al review summarised above:

Felson DT, Anderson JJ, Meenan RF. The comparative efficacy and toxicity of second-line drugs in rheumatoid arthritis. Arthritis Rheum. 1990; 33: 1449-61.

Related topics