Steroid Injections May do Harm and do No Good for Joint Pain


Joint PainA 2017 study showing that steroid injections are not helpful for joint pain finally confirms what many researchers have been saying for years.

Like many Americans, I get joint pain from time to time.  Each time, doctors have recommended steroid injections.  When I say I’d rather just know whether it is better to apply heat, exercise, or rest my elbow (or hip or shoulder), the answer is always the same “Why don’t you want steroid injections?  That’s what you should do.”

I can only hope this new study finally convinces doctors all over the country to think twice before recommending steroid injections, especially when patients don’t want them.

There have been numerous studies of steroid injections over the years, and many have shown little or no benefit.  And steroids have risks.

The study published in May, 2017 in the American Medical Association’s Journal, JAMA, is the best one.  Researchers randomized 140 patients with knee osteoarthritis to receive steroid or a placebo (saline) injection every 12 weeks for 2 years. At each visit, the doctors asked the patients about pain and function. They also got an MRI so the doctor could examine the cartilage at the start of the study, after one year, and after 2 years.

The patients is the study all had osteoarthritis (the most common cause of joint pain) but also had evidence of a type of inflammation called synovitis, which was based on an ultrasound. If anyone is going to respond to steroids, these are the patients that would.

When the researchers compared the patients who received steroids with the ones that received placebo (saline), there was no significant difference in knee pain at any time during the two-year period.  Thanks to the placebo effect, pain did decrease a bit in all patients. There were no changes in knee functioning, stiffness, or walking.

However, one thing did change significantly between the groups: cartilage thickness. Cartilage helps to cushion the joints, and cartilage loss is the major cause of joint pain.  The patients who received steroid injections lost significantly more cartilage than those who got saline.  This is exactly the opposite of what patients and their doctors would want to happen.

Will this convince doctors and surgeons, who have been recommending steroid injections for years?  Some doctors might assume that one steroid injection could be helpful, even though injections every three months clearly isn’t.  But we now know there is no scientific evidence to show any benefit to steroid injections in the knee, and no reason to think it would be beneficial in other joints either.  Why not give patients a different placebo, such as a vitamin pill or glass of water described as “an arthritis treatment.”?

For more information, see this article on MedPage Today and the summary of the medical journal article in JAMA

All NCHR articles are reviewed and approved by Dr. Diana Zuckerman and other senior staff.