Rheumatoid arthritis patients are frequently prescribed NSAIDs. A double-blind placebo-controlled trial comparing naproxen, celecoxib, and placebo demonstrated that the active agents were significantly more efficacious than placebo. There were fewer treatment failures in the active groups (approximately 25 percent) compared to placebo (45 percent). Adverse events were common in all groups, but few led to withdrawal from the study.80[II] A further double-blind randomized trial comparing cel-ecoxib with diclofenac over 24 weeks in rheumatoid arthritis came to a similar conclusion.81[II] However, it should be noted that whilst NSAIDs are useful in treating rheumatoid arthritis, others agents are often more effective.87[III]
In a survey assessing global preferences (effectiveness and side effects), which examined paracetamol and NSAIDs in patients with osteoarthritis (OA), rheumatoid arthritis, and fibromyalgia, 60 percent had a general preference for NSAIDs. The authors point out that this is a perception of effectiveness, which may differ from actual
Table 15.1 Review of evidence of effectiveness.
Low back pain Acute and chronic
Chronic back pain Subgroup assessment
Response to treatment
Survey of patient preference for benefit and side-effects Long-term comparison
NSAIDs systematic review NSAIDs versus placebo NSAIDs versus paracetamol NSAIDs versus placebo NSAIDs versus paracetamol Paracetamol/codeine versus tramadol NSAIDs, COX-2 antagonists and placebo Steroids versus NSAIDs
NSAIDs and paracetamol
Naproxen versus paracetamol Ibuprofen versus paracetamol NSAIDs
Limited evidence supports NSAIDs Statistical improvement for NSAID No difference
Unable to assess
Similar efficacy with combination being better tolerated Significant improvement compared with placebo; no difference between active agents Steroids show advantage in low dose, short-term use Preference for NSAIDs 60%, paracetamol 14%, and no preference 25% Similar efficacy with high drop-out rate suggesting neither is satisfactory Similar effects both better than placebo
Relative efficacy data not yet available 13[I], 14[I]
Probably no role to play
effectiveness. The results may be biased due to the beliefs and perceptions of both patients and physicians.83[IV]
In OA, the NSAIDs are commonly used. A two-year double-blind comparison of naproxen and paracetamol, using a model of OA of the knee, showed little difference between treatments for those completing the trial, the withdrawal rates were high (65 percent). The reasons for withdrawal were not significantly different between groups, although GI reactions were higher in the naproxen group.45[II] The Cochrane collaboration has reviewed the use of NSAIDs in OA of the knee.13[I] The reviewers conclude that despite a large number of publications, few are randomized control trials and many have substantial design faults. The authors were unable to demonstrate a difference in efficacy between agents or for withdrawal rates, and suggested that prescribing be based on relative safety, patient acceptability, and cost. Similar comments were made in a review which used OA of the hip as the model.14[I]
Pain can follow hip arthroplasty, as can heterotopic bone formation in the soft tissues surrounding the joint. A systematic review confirms that perioperative NSAIDs reduce the risk of heterotopic bone formation. The significance of this along with the short-term side effects is less clear and the long-term effect on pain and clinical outcome has not been fully elucidated.88[I], 89[I] There is also concern about the effect of NSAIDs and bone healing. Currently, there is insufficient evidence to make clear comments other than that further research is required.90
Neuropathic pain often proves difficult to manage and the evidence suggests that NSAIDs are probably ineffective in this condition.78[I]
Chronic pain in the elderly poses a major challenge for management. The population is aging and the elderly have a significantly higher incidence of chronic pain. Estimates from the USA suggest 70 million older people are prescribed regular analgesics and that the majority are NSAIDs. It is suggested that all NSAIDs should be used with caution and that high doses and long-term usage be avoided.91 It has also been suggested that NSAID toxicity is a problem in this population and ibuprofen should be the NSAID of first choice.92[V] There is a large individual variation with regard to the minimal effective and toxic doses. Dose titration is extremely valuable in this group and it is suggested that side effects are regularly monitored with long-term use.
Was this article helpful?