Steroid injections for a common form of back and leg pain known as spinal stenosis may have little or no benefit for patients, according to a new study by the UW School of Public Health and its research partners.

The study, published in the New England Journal of Medicine, was conducted at 16 hospitals across the U.S. with 400 patients age 50 or older who had at least moderate pain. Half the patients in the trial received epidural injections of the anesthetic lidocaine plus a corticosteroid, the other half received injections without corticosteroid.

"This is the first large, multicenter randomized trial to look at epidural steroid injections for spinal stenosis, and we found that injection with corticosteroids and lidocaine provided these patients with minimal or no additional benefit over lidocaine injections [alone]," said Dr. Janna Friedly, assistant professor in rehabilitation medicine at the University of Washington and a member of the Comparative Effectiveness, Cost and Outcomes Research Center, a joint effort of the Schools of Medicine and Public Health. "If patients are considering an epidural injection, they should talk to their doctor about the benefits and risks of the options available."

The researchers also found that patients combining corticosteroids and lidocaine were more likely to report side effects. Spinal stenosis occurs when the spinal canal narrows, putting pressure on the nerves. It is often caused by wear and tear and is the leading cause of back surgery in older adults, according to the Agency for Healthcare Research and Quality, which funded the research. Epidural injections for spinal stenosis have increased nearly 300 percent of the past two decades, costing millions of dollars.

Other School of Public Health faculty taking part in the research were Dr. Patrick Heagerty, professor of biostatistics; Dr. Sean Sullivan, professor of health services; Dr. Larry Kessler, chair and professor of health services; Dr. Jeffrey Jarvik, professor of health services; Dr. Brian Bresnahan, research assistant professor of health services; and Mr. Bryan Comstock, biostatistician.