Q: I had a steroid injection for a painful rotator cuff shoulder problem. I was expecting relief from the pain fairly quickly but nothing happened. The surgeon wants to inject the shoulder again.  If it didn't work the first time why do it again?

A: Injection of a steroid (antiinflammatory) and a numbing agent into the subacromial bursa is a fairly common treatment for rotator cuff problems. An inflamed bursa (a fluid-filled sac between the muscle and bone) can cause pain when the arm is raised (forward or to the side) past 90 degrees. The rotator cuff tendon gets pinched between the bursa and the bone. Bringing the bursa back down to its normal size can help eliminate painful symptoms.

But the injection may fail for two main reasons. One, the problem wasn't a rotator cuff tendinopathy -- there may be something else causing the painful symptoms. This would require further testing and study. And two, failure to accurately inject the bursa can mean painful symptoms persist.

Studies show that accuracy of injection can range anywhere from 29 to 87 per cent. Accuracy is much improved when the surgeon uses ultrasound or fluoroscopy (a special type of X-ray) to guide needle placement and injection. Without these tools, accuracy can be very diminished.

And in a more recent study from the Sports Medicine Center at UC-Davis, it was discovered that accuracy of injection can also be affected by which direction the surgeon uses to give the injection. Injections to the subacromial bursa can be given from the anterior (front), lateral (side), or posterior (back).

In that study, one orthopedic surgeon injected 75 shoulders using these three pathways. There were three groups of patients and each group received one of the three types of injection. Patients were randomly assigned to the group they were in. The fluid injected included the steroid medication, a numbing agent, and a dye. The dye was part of the injection so that X-rays taken would show the accuracy of the injection (i.e., did the fluid actually end up inside the bursa?).

This surgeon found that injecting the shoulder from the front and side (anterior and lateral routes) gave better results than injecting from the back (posterior). This was especially true for women. It turns out that the posterior route was the least accurate when injecting the subacromial bursa in females.

Reference: Richard A. Marder, MD, et al. Injection of the Subacromial Bursa in Patients with Rotator Cuff Syndrome. In The Journal of Bone and Joint Surgery. August 15, 2012. Vol. 94A. No. 16. Pp. 1442-1447.