Osteoarthritis (OA) is a progressive and often painful disease of the joints that affects millions. One of the most common treatments used to combat pain associated with OA is injection of corticosteroids into the affected joint. A recent article published in CBC news has quickly gotten the attention of the public with its claims that this treatment may be more harmful than it is effective. The article in question references several studies that supposedly expose the negative implications of such injections into the hip and knee. However, scientific research is complex and usually not entirely definitive, and is often misrepresented by the mainstream media to draw conclusions that may not be entirely valid.

In this discussion, we will review the studies referenced in the original article and discuss the other body of research that exists on the topic of effectiveness and potential harmful side-effects of corticosteroids. We will then summarize what the best currently available treatment options are for OA of the hip and knee. And finally, it will summarize the best treatment options currently available for osteoarthritis of the hip and knee.

Summary of the Research Referenced

This observational study examined 459 patients who were given corticosteroid injections into the hip or knee joints. It is important to note that in an observational study, there is no control group, meaning that there was no comparison made between patients who received the injection and those who did not. The authors looked at already available data from patients at one hospital to draw their conclusions. The conclusions were drawn based on X-rays and MRIs done before and after the injection, but not all patients had imaging available, in fact, only about half of the patients examined had post-injection images available.

The authors looked at the available information and found that some (about 8% of the 459) had significant complications, including stress fractures and bone death. The problem is that the authors do not provide pre- and post-injection comparisons between all those who suffered these adverse effects, so it is actually impossible to know whether any of these were present before the patients even had an injection. In fact, two of the patients whose images are discussed in the study did have pre-existing fracture and bone death, which got worse after treatment with corticosteroids. But this shouldn’t be surprising, as injection is not the standard treatment for these conditions.

We are given no information on the patients’ general health, lifestyle, and other treatments they may have gotten. In fact, the study was conducted in a hospital that provides care for underserved individuals, which means that these patients likely come for treatment once their disease is already in the later stages and they likely have not received quality medical care and guidance, so they are generally more likely to have poor outcomes. With the information provided, we cannot determine whether any of the adverse events observed were caused by the injection.

This study does highlight the need for some further investigation into the safety of corticosteroid injections as a standard treatment for pain due to osteoarthritis, but it definitely does not prove that corticosteroid injections cause severe adverse effects.

The article also references a double-blind randomized clinical trial which compared two groups of individuals: one group has corticosteroid injections into the knee every three months for two years, and the other had saline injections at the same frequency. They did find the corticosteroid group to have a greater rate of cartilage loss, but they actually found no difference in bone lesions, swelling and inflammation, serious adverse effects, or pain (we will discuss this in detail later). So yes, eight total injections of corticosteroid into the knee caused some cartilage loss in patients as seen on medical imaging, but did not actually result in any decrease in function, increase in pain, or any other changes within the joint (like bone death or fractures).

Another observational study (meaning no variables were controlled for, the authors took two sets of images and drew conclusions based on what they saw) looked at the images of two different groups of patients: patients with hip OA who chose to have corticosteroid injections due to high levels of pain and patients who had hip OA but did not choose to have an injection. The authors looked at before and after images of these patients and found that the group that had the injections had higher rates of OA progression and bone damage than those who did not get injections. But, no other variables were controlled for, and the authors themselves acknowledge that this study shows correlation, not causation. So it is possible that the patients who had chosen to get injections already had more severe pain and disability to begin with. Again, this does warrant some further investigation, but is not enough to draw a solid conclusion on the dangers of corticosteroid injections.

Finally, a review examining 27 different trials is referenced, and is said to have “found it was ‘unclear’ if there were actually any benefits to the treatment”. This is not entirely true, as the authors did find that “intra-articular corticosteroids may cause a moderate improvement in pain and a small improvement in physical function”, but they also found the quality of evidence to be low and therefore do not deem the treatment effective.

So based on this evidence, there may be some risks to corticosteroid injections into the joints for patient with OA, and there probably isn’t much benefit. But what does the larger body of research have to say?

Let’s take a look:

Corticosteroids and Pain

Corticosteroid injections are used to treat pain in the knee and the hip caused by arthritis, but do they actually work? The short answer is yes, but only for a brief period of time. Several systematic reviews have concluded that corticosteroids are more effective than placebo for pain reduction for up to 13 weeks, although most studies show that pain relief typically lasts about 3-4 weeks, and the greatest effect on pain is found at 1 week post injection. Evidence also shows that corticosteroid injections provide greater benefit to those with severe pain than those with mild to moderate pain levels. Studies have also found that there is a small improvement in function in patients receiving corticosteroid injections versus controls that lasts up to 6 weeks post injection, with the greatest effects shown at 1 week. So, there is a mild to moderate short term benefit in terms of pain relief and function, but that benefit is unlikely to last beyond about a month following the injection.

Are Corticosteroids Really Harmful to the Joint?

We already discussed this article comparing corticosteroids and control (saline) injections, which concluded that there was a greater loss of cartilage in the corticosteroid group. But it also concluded that there was no difference in terms of bone loss, pain, inflammation, or serious side-effects between the two groups. There was also no difference in symptoms or function, meaning that the loss of cartilage was largely not symptomatic. What this means is that those with greater cartilage loss may have to undergo a total joint replacement sooner than those without. It also brings into question the effectiveness of the injection for pain and function, but we already saw that this effect is rather small anyways.

A systematic review of 40 studies examining the effect of corticosteroid on cartilage health actually found that low doses and short durations of exposure to corticosteroid may have beneficial effects on cartilage. Higher doses and durations, however, have detrimental effects on cartilage. Based on this, the authors suggest that clinically, only the lowest effective dose should be used and treatment should be infrequent. This may explain the cartilage loss seen in the previously discusses study, as those patients received a total of 8 injections.

So if not Corticosteroids, what’s the Most Effective Treatment for Arthritis?

The one treatment which has been consistently shown to reduce pain and improve function and quality of life of patient with hip and knee OA, and one which was grossly under discussed in the CBC article, is exercise. There is a large amount of evidence showing that both aerobic (cardio) and strengthening exercises have a positive effect on patients’ pain and function, which can last up to 6 months even after the exercise has been stopped. There is also very little evidence for any adverse effects associated with exercise, and the negative consequences mentioned in the literature are minor, consisting of muscle cramping and mild joint discomfort. These can also be easily mitigated when exercise is supervised by a trained professional such as a physiotherapist or chiropractor. For best results, exercise should be taught and initially supervised by a professional to ensure it is catered specifically to the individual and carried out with proper form.

But the good news is that even simply walking has beneficial effects on pain and dysfunction caused by OA, so if you’re not ready to begin a structured exercise program or do not have access to an appropriate health care provider, you can begin a regular walking program to help decrease your pain and improve function.

What About Weight Loss?

Weight management or weight loss is only recommended for those patients who are overweight or obese and should be implemented with caution in anyone who suffers from certain comorbidities such as osteoporosis or osteopenia. It is best to speak to your health care provider to determine whether a weight loss program may be beneficial to you.

So What’s the Verdict, should I get an Injection?

With all the available evidence, it is safe to say that a corticosteroid injection alone is not a good form of treatment for hip and knee OA, and patients should first look to implement an exercise program into their daily routine to help mitigate pain and decreased function associated with OA. However, because corticosteroids do seem to provide some short-term pain relief, it is possible that they may be useful in helping some patients with high levels of pain get started on an exercise program.

One trial evaluating potential benefit of a corticosteroid injection before the start of an exercise program found no difference in pain, inflammation, or physical function upon completion of the program. This is good news, as it shows that the same level of function can be achieved without the help of corticosteroids, and that exercise may be superior to corticosteroids for pain relief, as patients in both groups saw substantial decreases in pain at 12 weeks. The authors did not look at patient motivation to begin an exercise program, but it is possible that those who got the injection were more comfortable starting the program due to the short-term reduction in pain provided by the injection.

It is likely that patients with high levels of pain who do get the pain relief from corticosteroids could be more motivated and thus more likely to seek out a more active lifestyle, which is what we often see in the clinic in practice.

The decision on whether or not to seek pharmacological intervention in the form of a corticosteroid injection should be an individualized one, made by the patient with the help of their health care provider. It should consider their pain levels, functional restrictions, and a multitude of other (biopsychosocial) factors. One thing we can say for sure is that an injection is not the solution to pain and dysfunction caused by OA, and the best long term treatment is (almost) any form of exercise.

References:

1. 1. Atchia, I., et al. “Efficacy of a Single Ultrasound-Guided Injection for the Treatment of Hip Osteoarthritis.” Annals of the Rheumatic Diseases, vol. 70, no. 1, 2011, pp. 110–116., doi:10.1136/ard.2009.127183.
   2. Baar, Margriet E. Van, et al. “Effectiveness of Exercise Therapy in Patients with Osteoarthritis of the Hip or Knee: A Systematic Review of Randomized Clinical Trials.” Arthritis & Rheumatism, vol. 42, no. 7, 1999, pp. 1361–1369., doi:10.1002/1529-0131(199907)42:73.0.co;2-9.
   3. Bannuru, R.r., et al. “OARSI Guidelines for the Non-Surgical Management of Knee, Hip, and Polyarticular Osteoarthritis.” Osteoarthritis and Cartilage, vol. 27, no. 11, 2019, pp. 1578–1589., doi:10.1016/j.joca.2019.06.011.
   4. Cibulka, Michael T., et al. “Hip Pain and Mobility Deficits—Hip Osteoarthritis: Revision 2017.” Journal of Orthopaedic & Sports Physical Therapy, vol. 47, no. 6, 2017, doi:10.2519/jospt.2017.0301.
   5. Fransen, Marlene, et al. “Exercise for Osteoarthritis of the Knee.” Cochrane Database of Systematic Reviews, 2015, doi:10.1002/14651858.cd004376.pub3.
   6. Godwin, Marshall, and Martin Dawes. “Intra-Articular Steroid Injections for Painful Knees Systematic Review with Meta-Analysis.” Canadian Family Physician, vol. 50, Feb. 2004, pp. 241–248.
   7. Henriksen, Marius, et al. “Evaluation of the Benefit of Corticosteroid Injection Before Exercise Therapy in Patients With Osteoarthritis of the Knee.” JAMA Internal Medicine, vol. 175, no. 6, 2015, p. 923., doi:10.1001/jamainternmed.2015.0461.
   8. Hepper, C. Tate, et al. “The Efficacy and Duration of Intra-Articular Corticosteroid Injection for Knee Osteoarthritis: A Systematic Review of Level I Studies.” Journal of the American Academy of Orthopaedic Surgeons, vol. 17, no. 10, 2009, pp. 638–646., doi:10.5435/00124635-200910000-00006.
   9. Hernández-Molina, Gabriela, et al. “Effect of Therapeutic Exercise for Hip Osteoarthritis Pain: Results of a Meta-Analysis.” Arthritis & Rheumatism, vol. 59, no. 9, 2008, pp. 1221–1228., doi:10.1002/art.24010.
   10. Jüni, Peter, et al. “Intra-Articular Corticosteroid for Knee Osteoarthritis.” Cochrane Database of Systematic Reviews, 2015, doi:10.1002/14651858.cd005328.pub3.
   11. Klocke, Rainer, et al. “Cartilage Turnover and Intra-Articular Corticosteroid Injections in Knee Osteoarthritis.” Rheumatology International, vol. 38, no. 3, 2018, pp. 455–459., doi:10.1007/s00296-018-3988-2.
   12. Kompel, Andrew J., et al. “Intra-Articular Corticosteroid Injections in the Hip and Knee: Perhaps Not as Safe as We Thought?” Radiology, 2019, p. 190341., doi:10.1148/radiol.2019190341.
   13. Mcalindon, Timothy E., et al. “Effect of Intra-Articular Triamcinolone vs Saline on Knee Cartilage Volume and Pain in Patients With Knee Osteoarthritis.” Jama, vol. 317, no. 19, 2017, p. 1967., doi:10.1001/jama.2017.5283.
   14. Mccabe, P.s., et al. “The Efficacy of Intra-Articular Steroids in Hip Osteoarthritis: a Systematic Review.” Osteoarthritis and Cartilage, vol. 24, no. 9, 2016, pp. 1509–1517., doi:10.1016/j.joca.2016.04.018.
   15. Middelkoop, M. Van, et al. “The OA Trial Bank: Meta-Analysis of Individual Patient Data from Knee and Hip Osteoarthritis Trials Show That Patients with Severe Pain Exhibit Greater Benefit from Intra-Articular Glucocorticoids.” Osteoarthritis and Cartilage, vol. 24, no. 7, 2016, pp. 1143–1152., doi:10.1016/j.joca.2016.01.983.
   16. Raynauld, Jean-Pierre, et al. “Safety and Efficacy of Long-Term Intraarticular Steroid Injections in Osteoarthritis of the Knee: A Randomized, Double-Blind, Placebo-Controlled Trial.” Arthritis & Rheumatism, vol. 48, no. 2, 2003, pp. 370–377., doi:10.1002/art.10777.
   17. Roddy, E. “Aerobic Walking or Strengthening Exercise for Osteoarthritis of the Knee? A Systematic Review.” Annals of the Rheumatic Diseases, vol. 64, no. 4, 2005, pp. 544–548., doi:10.1136/ard.2004.028746.
   18. Simeone, F. Joseph, et al. “Are Patients More Likely to Have Hip Osteoarthritis Progression and Femoral Head Collapse after Hip Steroid/Anesthetic Injections? A Retrospective Observational Study.” Skeletal Radiology, vol. 48, no. 9, 2019, pp. 1417–1426., doi:10.1007/s00256-019-03189-x.
   19. Tian, Kewei, et al. “Intra-Articular Injection of Methylprednisolone for Reducing Pain in Knee Osteoarthritis.” Medicine, vol. 97, no. 15, 20 Mar. 2018, doi:10.1097/md.0000000000010240.
   20. Wernecke, Chloe, et al. “The Effect of Intra-Articular Corticosteroids on Articular Cartilage.” Orthopaedic Journal of Sports Medicine, vol. 3, no. 5, 2015, p. 232596711558116., doi:10.1177/2325967115581163.