By: Dr. Geoffrey Modest
There have been a few articles in the past years which have questioned the utility of steroid joint injections. One recent one looked at knee injections in patients with osteoarthritis (OA) to see if injections improved outcomes in those about to enroll in physical therapy (see doi:10.1001/jamainternmed.2015.0461).
Details:
–100 patients (61% female, mean age 63, BMI 29) with knee pain on walking, evidence of localized knee inflammation on exam, and radiologic confirmation of OA were randomized to intra-articular 40mg of methylprednisolone plus 4cc of lidocaine 1% vs saline plus lidocaine, then 2 weeks later started a program with PT 3x/week for 12 weeks
–Primary outcome: pain relief on the KOOS questionnaire (knee injury and osteoarthritis outcome score). Secondary outcome included objective measures of physical function and inflammation. outcomes were measured at baseline, 2 weeks (when about to start PT), 14 weeks (at end of PT), and at 26 weeks
Results:
–no difference in primary or secondary outcomes with corticosteroid injections at any of the times measured
The authors do comment that they used doses of steroids on the lower range of the recommended dosage for knee joints, and that several older studies have found significant short-term benefit with steroid injections. But their findings do not support the common belief that giving steroids is useful to decrease pain in the short-term in order to improve the ability of the patient to perform and benefit from PT. They did not test either a higher dose of steroids, or the use of more than one injection (and some people, in my experience, do require a second injection several weeks later to get benefit).
I will use this article to make some tangential observations.
–The Natl Institute for Health and Care Excellence in the UK (NICE) have 2014 guidelines on osteroarthritis management (see here) which do support the use of intra-articular corticosteroid injections for the relief of moderate to severe pain in people with osteoarthritis. Unfortunately, this is just a set of recommendations, though the ones on intra-articular steroids refer back to their very in-depth and thorough review in their previous 316-page 2008 guidelines (see here), which basically support steroid injections pretty much anywhere OA is present
–A systematic review of 41 articles with 2672 patients looked at the data for steroid injections at various sites (see DOI:10.1016/S0140- 6736(10)61160-9) and found that steroid injections of lateral epicondylitis reduced pain compared with other interventions in the short-term. The data for rotator cuff tendinitis is mixed, though there was evidence that repeated injections are less likely to be beneficial
–An article on carpal tunnel syndrome compared steroid injection vs surgery (see DOI 10.1002/art.20767), finding that in carpal tunnel refractory to standard medical management, there was equal improvement with steroid injection vs surgery at 6 and 12 months, with more rapid relief with steroid injection.
So, as someone who has given literally many, many hundreds of injections, I would add:
–injections have a few huge benefits, including (certainly in my experience) rapid short-term relief of pain and gain in function, which is really important for my patients doing manual labor and needing to go to work, as well as many patients who just have terrible pain which interferes with their daily functioning. And it spares them the potential systemic toxicity of systemic NSAIDs, etc. That being said, many patients with less severe symptoms do very well with topical agents (capsaicin, topical NSAIDs such as diclofenac, topical lidocaine). And, over the decades I have had remarkably positive outcomes for knee injections for OA, gout, pseudogout, etc, as well as injections of other OA sites (eg AC joint, 1st MCP joint, etc), rotator cuff and bicipital tendonitis of the shoulder, lateral (but not medial) epicondylitis, trigger finger injections, DeQuervain’s tendinitis, intra-articular wrist injections for people with rheumatoid arthritis, carpal tunnel injections, an array of areas of bursitis (hip, knee, elbow, shoulder), and even muscle spasms of the lower back or trapezius, etc. Again, injections almost always give immediate relief, in many cases one injection is sufficient, and they spare the systemic effects of medical management. I also prefer injecting patients with gout (vs NSAIDs, colchicine, prednisone), with about 100% very positive results, even in the great toe where i would guess that >50% of the time I do not get into the joint itself.
–I typically suggest a different intervention, typically surgical, if the patient does not respond to 1 or 2 injections. And this applies to carpal tunnel, tendinopathies, or intraarticular injections (eg, the knee)
–but I also have many elderly with chronic knee OA who get good but short-term benefit (only a few months) from injections and are not interested in major surgical intervention, for whom I do repeated knee injections to preserve function, the ability to exercise and its attendant positive health effects, perform self-care, etc
–so, despite the negative studies, both the larger systematic reviews and my own experience has been pretty positive for intraarticular joint and other injections, with the caveat that there should usually be referral for more invasive treatments if the injections don’t work. and I do routinely suggest combining injections with other therapies, such as physical therapy, home exercises, medications when needed — even though there have been recent studies finding, for example, that acetaminophen does not work for back pain, or adding physical therapy for shoulder pain is not better than steroids alone.