Hot Topics Blog
Rebecca is 68 and has proven OA of the knee. She otherwise has no significant past medical history and has been managing her knee pain reasonably well, but has been wary of ‘overdoing it’. She comes to see you to ask for a steroid injection as a friend of hers had one recently and said it was ‘wonderful’.
As a GP registrar (a good 10 plus years ago) it felt, rightly or wrongly, that the general consensus was to stick a needle, usually loaded with some steroid, into any musculoskeletal problem we could find. This was certainly the prevailing impression having recently come out of a six month rheumatology job which involved sessions in the injection clinic, where I focussed more on injection technique than the evidence base to support the procedure. Shoulders - ‘certainly Sir, would you like the anterior, lateral or posterior approach?’ Tennis elbow - ‘yes Madam, shall we try the perpendicular or fan technique?’ Plantar fasciitis - ‘this will hurt you more than me’…and what about knees? The satisfaction of draining a few 20ml syringes of synovial fluid out of an inflammatory effusion followed by an 80mg kenalog chaser was fairly substantial for both patient and clinician!
But times have changed and so has the evidence base, with increasing evidence that corticosteroid injections give, at best, only short term benefits for many musculoskeletal conditions, and in some cases worse longer term outcomes. But should we consider a corticosteroid injection for Rebecca for her knee OA? The evidence is equivocal but there seems to be an increasing ‘signal’ that we need to think twice before diving in with this intervention, at least as a ‘routine’ procedure. Corticosteroid injection for knee OA was the subject of a good review in the BMJ ‘uncertainties’ section in January 2020 (BMJ 2020;368:l6923). The latest Cochrane review from 2015 cited low quality evidence that corticosteroid injection may improve pain and function compared to placebo in the short term (up to 6 weeks), but there was no evidence to show improvement beyond 6 months. Subsequently a placebo controlled RCT in JAMA 2017 showed no clinical difference in those having steroid injections every 3 months over a 2 year period, compared to placebo. Worryingly this study also showed a small but statistically significant deterioration in cartilage depth on imaging in the steroid injection group - this is obviously only a surrogate marker and does not necessarily correlate to a deterioration in pain or function, but the BMJ article also cited some observational data suggesting worsening pain, stiffness and function in those having repeated steroid injections.
What about other options for knee OA? It is such a common presentation in General Practice and it feels like treatment options are increasingly being taken away from us. Paracetamol? Unlikely to help according to the latest Cochrane review in 2019 which cites high quality evidence that paracetamol does not give any meaningful benefits to pain or function in knee or hip OA. Opiates? NICE recommend them as an option, but given the latest Cochrane review in 2019 tramadol is another one off the list, with results showing overall minimal benefits and significant increased side effects in those taking tramadol. Topical NSAIDS can be effective but we are all too aware of the GI, CV and renal risks of oral NSAIDS. Knee replacements can be a highly effective intervention for those with more severe OA, but even that is not a panacea - we will all have many patients that have had fantastic outcomes from this procedure, but likewise a number who had significant complications or simply wish they had never had the operation in the first place.
So what, I hear you cry, can we do to help Rebecca? Well, we should not forget the importance of an honest and empathetic discussion about her condition and the importance of self management. There were some excellent insights in a 2018 Cochrane review looking at exercise interventions in patients with OA knee and hip. The quantitative data was inevitably of lowish quality, but there was enough data to suggest exercise interventions do improve pain, physical function and depression. Possibly of more importance was the review of the qualitative data which strongly suggests that challenging inappropriate health beliefs (e.g. that exercise will worsen joint damage) is a crucial barrier we can remove for patients by giving better information about the safety and value of exercise in OA.
So are steroid injections for knee OA off the menu? No, but I do think we need to be honest about the uncertainty of the evidence and any potential benefits; there will clearly still be situations where this might be a beneficial intervention - the younger patient that needs some improvement in symptoms to allow them to pursue exercise therapy and/or weight loss, or the frail patient that is not fit for surgery who might benefit from an injection to improve pain and function and reduce falls risk. As always there are no right answers in General Practice, but we do need to inform patients that knee steroid injections will only give short term relief and may possibly increase the risk of disease progression if repeatedly done, and that in the longer term it may be better to step away from the needle and focus more on exercise.
Find this blog useful? You can quickly add CPD to your account by writing a reflective note about the post you have read.
Log in to your NB Dashboard and use the 'Add Reflective Note' button at the bottom of a blog entry to add your note.