Spinal stenosis is a pain – both for those who suffer it, and the doctors who treat it.
As surgery is a potentially risky option with uncertain benefit, many doctors turn to glucocorticoid injections for their patients, to decrease pain and increase mobility. An estimated ten to eleven million such injections are performed in the US annually, numbers that have grown rapidly in recent years.
But do glucocorticoid injections actually help? A recent review from the North American Spine Society highlighted the paucity of data, concluding that there is insufficient evidence to make a recommendation either for or against. And this therapy isn’t always benign. While serious complications are rare, complications including paralysis and nerve damage have been reported.
With this background, Janna Friedly and colleagues set out to determine whether glucocorticoid injections actually benefit patients with spinal stenosis. Their results, published in this week’s issue of NEJM, suggest that the increasingly popular treatment is no better than their comparator, an injection of lidocaine alone.
The study investigators enrolled 400 patients with lumbar spinal stenosis, whose disease caused them moderate-to-severe leg pain and disability and who had been referred for steroid injections. All study participants were older than 50 and hadn’t ever undergone lumbar surgery, or received epidural steroid injections in the previous six months. Patients were randomly assigned to either epidural glucocorticoid injection with lidocaine, or lidocaine injection alone. The participants could receive a second injection three weeks later, at the patient’s discretion.
After six weeks, patients in both groups were asked to rate their average buttock, hip and leg pain in the previous week and to fill out a questionnaire that quantifies degree of pain and associated disability. They were also asked to respond to surveys of depression, anxiety and quality of life.
The results? Both groups of patients reported improvement in their pain and physical function at three and six weeks, whether or not their injections included glucocorticoids. At six weeks, there were no significant differences in pain or function ratings between the two groups. Of note, the patients randomly assigned to the glucocorticoid injections were more likely to have improvement in depressive symptoms. Those receiving glucocorticoid injections had a higher rate of adverse events, although the complications were generally mild.
In an accompanying editorial discussing these results, orthopedic surgeon Gunnar Anderson notes the difficulties inherent in trials of treatments for spinal stenosis. For one, spinal stenosis is a heterogeneous disease both in terms of cause – congenital versus degenerative – location, and extent. Additionally, Friedly’s trial did not include a control group that received sham injections, leaving open the question of whether the benefit seen in both groups could be due to the lidocaine injection itself, although this would be unlikely.
Despite these questions, Anderson concludes that the study “raises serious questions about benefits from epidural corticosteroid injections for spinal stenosis…Patients should be informed that currently best available data have not supported a significant long term clinical benefit overall, and that complications are possible.”