Researchers say provocative discography may do more harm than good, but a prominent spine expert says it’s still a useful medical tool.

A common technique used to diagnosis lower back pain may actually be causing damage to patients’ spines.

That’s the conclusion of researchers from across the country who worked on a 10-year study of provocative lumbar discography, which was published recently in The Spine Journal.

The provocative discography procedure is performed on about 70,000 people in the U.S. every year to determine which disc in their lower back in causing pain.

The researchers concluded the technique causes more accelerated degeneration of discs as well as more frequent sciatica, back pain syndromes, surgeries, doctor visits, and loss of work.

“This long-term study proves what many spine specialists have suspected: provocative discography has serious adverse events over time,” Dr. Eugene J. Carragee, professor of orthopedic surgery at the Stanford University School of Medicine and one of the authors of the study, said in a press statement.

Their findings won a “best paper” designation at the 2015 International Society for the Study of the Lumbar Spine (ISSLS).

Others, however, aren’t convinced of the paper’s findings.

Dr. Conor O’Neill, the director of the nonoperative spine program at the University of California San Francisco Medical Center, said the study raises legitimate points, but its conclusions may be a bit strong.

“The bottom line on the study is it raises some concerns to be considered,” O’Neill told Healthline, “but it doesn’t provide convincing evidence that discography does this type of damage.”

Read More: 5 Strengthening Exercises for Lower Back Pain »

In a provocative discography, a needle punctures discs in the spine that are suspected of causing weakness or pain.

Fluid is then injected to pressurize the disc. This procedure is repeated in random discs until the patient feels pain in a disc, which is then considered the source of the patient’s back pain. Surgery or other treatment follows.

For their study, researchers recruited 75 patients with recent back pain problems between 1996 and 1998. These volunteers received an experimental discogram at three disc levels.

The researchers also enlisted a group of 75 volunteers with similar backgrounds with no back problems as a control group. These subjects were not given a discography.

All the volunteers were followed up after one, two, five, and 10 years. They had MRIs at the beginning of the experiment and after 10 years.

Read More: Get the Facts on the Spine »

The researchers said they noticed little difference in the two groups early on, but there were disparities by the end of the study period.

First, 16 of the patients who underwent discographies had lumbar surgery within 10 years compared with four in the control group.

In addition, 21 discography volunteers had new CT or MRI evaluations for back pain before the research was completed compared with 11 in the control group.

The researchers added the discography group had more doctor visits, more missed days of work, and more prolonged episodes of back pain. Those problems seemed to escalate in the later years of the experiment.

“Because the patients did not have back pain for years before the study began, the rapid disc degeneration and clinical deterioration of these working-age subjects is ominous,” Carragee, who is also an orthopedic surgeon and editor-in-chief of The Spine Journal, said. “Given the accuracy of discography testing has not been proven, the proof of definitive harm makes use of this test in most any circumstance questionable and most likely should not be used without high-level studies demonstrating a high usefulness of the test.”

The American Pain Society agrees. The organization’s recent guidelines recommend against using provocative discography to evaluate lower back pain.

Read More: Questions to Ask Your Doctor About Back Pain »

O’Neill, however, questioned some of the methodology used in the study.

He said the sample groups may not have been as representative of the general population as they could have been. He noted there was also a 20 to 30 percent dropout rate over the 10-year period that may have affected the study’s accuracy.

“It was a very limited population that may or may not represent the general public,” O’Neill said.

O’Neill has done thousands of discographies over the years and he said he has never come across a case where the procedure itself damaged a disc.

He said he uses discography only on patients who are being considered for lower back surgery and after he has studied their medical history, symptoms, and an MRI of their back.

O’Neill usually will puncture the suspected damaged discs of a patient and then needle stick a healthy disc as a control group.

However, he said, after reading this recent study he probably will stop puncturing healthy discs.

O’Neill acknowledged a discography is painful and should only be used on a very select group of patients.

“It’s a terrible test, but it’s the best one we have,” he said. “There’s no other way to diagnosis a painful disc. It’s better than guessing.”