Osteoarthritis of the knee in primary care

Abstract

Topical NSAIDS are as effective as oral NSAIDs, and patients prefer them

Many older people have pain in one or both knees from time to time, and the most likely cause is osteoarthritis. In some people the symptoms are severe or intrusive enough to consider an intervention.

The National Institute for Health and Clinical Excellence (NICE) has just published its draft guideline on the management of osteoarthritis.¹ It lists five interventions regarded as “core treatments” for osteoarthritis of the knee—paracetamol; education and information; exercises; weight loss (if the patient is overweight); and topical non-steroidal anti-inflammatory drugs (NSAIDs). The guideline lists another 14 interventions, ranging from those that are safe (such as alterations to footwear or local heat and cold), to those that are potentially harmful (such as oral NSAIDs, opioids, and surgery). The first sentence of the draft guideline says, “Treatment and care should take into account the patients’ needs and preferences.” So what choices are available and how should people decide?

Two accompanying papers compare the value of a topical NSAID (ibuprofen gel) with oral use of the same drug for osteoarthritis of the knee.^{2 3} The first study by Underwood and colleagues describes two trials—a randomised controlled trial that compares advice to use topical ibuprofen with advice to use oral ibuprofen, and a preference trial offering the same options. The second paper by Carnes and colleagues is a nested qualitative study that explores the reasons for patients’ preferences. The randomised controlled trial was powered for equivalence, and it found no significant difference in the WOMAC osteoarthritis index or major and minor adverse effects at one year between people who used the topical preparation or the oral drug. Cynics might conclude that both interventions are useless, but other data indicate that topical and oral ibuprofen perform slightly better than placebo, at least in the short term.^{4 5}

The results from the preference data are fascinating. Firstly, more people chose the preference study than the randomised controlled trial, and nearly three times more of them opted for the topical preparation (n=224) than the oral preparation (n=79). Quantitative analyses showed that women and people with a lower socioeconomic status were more likely to choose the preference study. Another intriguing finding was that adverse events after oral ibuprofen occurred less often in participants who chose tablets than in those who were randomised to them. The qualitative data indicated that the choice between the topical or oral preparation depended on the severity of the pain, whether or not participants had pain at other sites, and their perceptions of likely adverse effects. So participants with more constant or severe pain and other painful sites (or both), and those who were more concerned about toxicity, opted for the topical gel. These choices seem reasonable.

So what does this mean? Firstly, it shows that, given options, patients will make sensible rational choices about how they want to be treated, and that their ability to choose may improve efficacy and reduce toxicity. Secondly, the data indicate that topical NSAIDs are a viable safe alternative to oral NSAIDs for the treatment of osteoarthritis of the knee. A systematic review of a different topical NSAID found similar results,⁶ and the NICE guideline suggests that topical agents are cost effective.¹ In view of the current distrust of oral NSAIDs among patients and professionals—because of problems with drugs like Vioxx and Prexige—this is important.

But will this change our practice, and will we switch our patients from oral drugs to topical ones? A variety of topical agents are available for osteoarthritis, ranging from old fashioned ointments, linaments, and balms that have been used for centuries,⁷ to topical NSAIDs, capsaicin,⁸ local anaesthetics, patches containing opioids or other analgesics, and topical preparations of seemingly ineffective agents such as glucosamine.⁹

The over the counter market for these preparations is huge. Why? Is it because of the efficacy of the drugs within them, or is it more about the age old practice of “rubbing it better?” In my view, placebo or context effects explain most of the value of topical agents in osteoarthritis.¹⁰ But for me to recommend a placebo it must be safe and be something that I believe in (so that I can prescribe it without damaging the trust between me and my patients). In addition, it is more likely to work if the patient believes in it.¹¹ Evidence based medicine and randomised controlled trials have sadly taken away the option of prescribing placebos even if, like topical NSAIDs for osteoarthritis, they are safe and useful. Perhaps it is just as well that the trials reported here did not include a placebo arm.