Back pain? Here’s relief

This story reports on two studies published simultaneously about low back related problems. It’s worth noting that although both studies concern back surgery, the medical problem they address is not back pain. Back pain may attend these conditions, but the primary concern is relief from leg pain–the radiating symptoms that travel down the buttock, below the knee, and occasionally into the foot.

In addition to a headline and lead sentence that mislead on this score, the news story fails to accurately characterize the nature of the evidence in the two studies it reports. It also does not quantify the benefits of the treatments the studies explore. These lapses are especially important in the second study about treatments for people who had degenerative spondylolisthesis (DS) with associated spinal stenosis. One in a series of studies from the multicenter SPORT investigators, this study involved two different cohorts distinguished by their treatment preferences. One large group consisted of people who agreed to be randomly assigned to either surgery or nonsurgical care—either was fine with them. The other group consisted of people who chose their treatment. The news article fails to mention this. The randomized trial was essentially invalidated when 40% of patients assigned to nonsurgical care changed their minds and decided to have surgery. Though this might seem like a technicality that only a methodologist could love, “crossovers” make it very difficult to estimate the true effect of a treatment. Indeed, when the authors analyzed the randomized patients according to the group to which they were assigned (“intention-to-treat analysis”), they found no differences in the outcomes of the surgical and nonsurgical patients. When they analyzed patients according to the treatment they actually received (“as-treated analysis”), they saw a big difference. Richard A. Deyo, MD, suggests in his editorial that the truth probably lies somewhere in between. The news article also failed to mention that the people who received nonoperative care generally showed “moderate” improvement (a term quantified by the researchers in both relative and absolute estimates). If the article had explained more, readers might have come away with a more nuanced message about the efficacy of surgery for this unhappy condition.

The story’s succinct, single paragraph about the first study, comparing treatments for leg pain (sciatica) caused by a herniated disc, captures the essence of the study when it says that people suffering with sciatica “might want to wait and see” before deciding to have surgery. There is a good chance they will feel equally improved after a year regardless of whether they try surgery or nonoperative treatments. But when the story correctly points out that people who had surgery “got better faster,” it neglects to say how much better or how much faster—important information. For example, patients actually had worse disability scores at 4 weeks if they had disc surgery. The greatest advantage for surgery was between weeks 8 and 12, and surgery continued to speed recovery only through week 36. Those who waited to have surgery experienced similar improvement. As Deyo suggests, there needn’t be a rush to surgery. “There does not seem to be a therapeutic window that closes quickly.”

None of this would matter if all treatments were equally inexpensive and risk-free. But they aren’t—and the news story misses on this score as well. A simple mention that treatment decisions have trade-offs–in dollars or in lost productivity—would be helpful. A note on the potential harms would help, too. Were patients enrolled in the studies harmed by any of the treatments they received? Are any of the treatments known to have long-term harms that might not show up in these short-term studies? Is fusion benign? Fusion intentionally eliminates movement in the fused segment of the spine, which can add extra stress to the segments above and below. This added stress often creates its own problems and triggers additional fusions. People should know that, too.

HealthNewsReview.org often points out that articles do not need to be long to be complete. In this case, that may not be true. Degenerative spondylolisthesis with spinal stenosis is a complex problem, its treatment is potentially harmful, and the study designed to evaluate the condition was fairly complicated. Herniated discs are less complicated, surgery to treat them potentially less harmful, and the study evaluating them less complex. Though the article made a good faith effort to cover a lot of difficult ground, its 487 words may have been destined to fall short.