ARTICLE IN BRIEF
4 A randomized controlled trial found that while surgery – lumbar diskectomy – and non-surgical alternatives – physical therapy and exercise – significantly improved pain, patients who had surgery achieved slightly better results more quickly.
Most patients with acute back and leg pain from lumbar disk disease recover well with or without surgery, although surgery appears to be more helpful for severe disabling sciatica, according to a much-anticipated comparison of outcomes at 13 spine pain clinics in the United States.
Which intervention is most beneficial remains uncertain, however, because many randomized patients crossed over from the surgical to the non-surgical study group, and vice versa, and data on many subjects were missing, limiting interpretation of the results, the investigators and other spine experts said.
The multicenter Spine Patient Outcomes Research Trial (SPORT) found that while both techniques significantly improved pain, patients who underwent surgery achieved slightly better results more quickly. However, the differences were not statistically significant, except for measures of sciatica and self-rated improvement.
The results of SPORT, which included both a randomized and an observational study of patients who chose their own treatment, were published in the November 22/29 issue of the Journal of the American Medical Association (2006;296:2441–2450; 2451–2459).
Led by James N. Weinstein, DO, Chairman of the Department of Orthopedics at Dartmouth Medical School in Hanover, NH, the researchers evaluated surgical and non-surgical outcomes in 472 patients in 11 states. All were candidates for surgery, with imaging-confirmed lumbar intervertebral disk herniation and persistent pain, typically down a leg, for at least six weeks.
Inclusion criteria were uniform at enrollment and included radicular pain and evidence of nerve root irritation and tension or corresponding neurological problems such as asymmetrical depressed reflex, decreased sensation, or weakness.
All patients were candidates for surgery as indicated by advanced vertebral imaging showing disk herniation, and patients were excluded for prior lumbar surgery, cauda equina syndrome, scoliosis greater that fifteen degrees, segmental disability, vertebral fractures, infection or tumor, inflammatory spondyloarthropathy, or other comorbid conditions. Patients unwilling to undergo surgery within six months were also excluded.
Results were evaluated both by “intent-to-treat” analysis – outcomes by initial treatment group assignment regardless of whether the patient opted for a different intervention during the two-year study, and “as-treated” – outcomes analyzed and compared by actual intervention received. Primary outcomes included pain and physical function, while secondary outcomes included severity of sciatica, patient satisfaction, self-reported improvement, and employment status. Patients were evaluated clinically after three months, one, and two years.
The intent-to-treat analysis showed a small advantage with surgery for primary outcomes at each evaluation. From baseline, the surgical treatment effect as measured by the Sciatica Bothersomeness Index (SBI) fell 2.1 points at three months compared to nonsurgical care (−9.0 vs. −6.8); by −1.6 at one year (−10.3 vs. −8.7); and −2.6 after 2 years (−31.4 vs. −28.7). The SBI ranges from 0 to 24, with lower scores indicating less severity.
Markedly different results were reported in the as-treated analysis, which showed a strong, statistically significant advantage in favor of surgery at each follow-up. The treatment effect for surgery at three months as measured by the SF-36 Medical Outcomes evaluation of pain and function improved for the surgery versus non-surgical group by 14.8 points (40.9 vs. 26); at one year by 10.8 points (42.8 vs. 32); and at two years by 10.2 (42.6 vs. 32.4). The surgical treatment effect as gauged by the Oswestry Disability Index (ODI) fell 36.1 points at three months (−36.1 vs. −20.9); by 15.2 points at 1 year (−37.7 vs. −22.4); and by 13.4 points (−37.6 vs. −24.2 for no surgery) at two years.
Scores for the SF-36 range from 0 to 100, with higher scores indicating less severity, while the ODI ranges from 0 to 100, with lower scores indicating less severe symptoms.
Fifty percent of patients assigned to surgery did so within three months, as did 30 percent of those assigned to the conservative treatment arm, and between 40- and 60-percent of patients crossed over into a different study arm over the study's course, and data on 24 percent to 27 percent of patients were missing.
“While the study certainly had limitations, small effects were seen in the intention-to-treat analysis despite the high crossover rate, and large effects in the as-treated and observational cohort,” commented co-author Jon D. Lurie, MD, Associate Professor of Medicine at Dartmouth Medical School, in an e-mail.
At three months, surgical patients had more significant improvement in primary and secondary outcomes than did those who did not undergo procedures.
“These results and prior evidence … are probably strong enough that without significant changes in either surgical or non-operative treatments, another randomized trial … for intervertebral disc herniation is not needed,” he said.
“It's important to realize that the diagnosis is clinically well established; the amount of pain, the degree of disability, and the patient's ability to cope with and manage the pain are probably the key criteria for selecting surgery.”
BETTER TREATMENT DECISIONS?
John D. Loeser, MD, Professor of Neurosurgery and Anesthesiology at the University of Washington School of Medicine in Seattle, told Neurology Today in a telephone interview that the SPORT results simply confirm what he learned as a resident 30 years ago.
“Most patients with a ruptured disc improve without surgery, but if the pain is unendurable, and can't be successfully managed with medication, surgery can provide almost immediate relief,” he said. “The problems with patients crossing over into different treatment groups were also to be expected,” Dr. Loeser said. “This is by far the best study to date on the subject, but it is incredibly difficult to get people to be randomized when there is a surgical option for pain,” he commented.
In SPORT, only 5 percent of patients experienced post-operative complications from surgery, with 4 percent of patients undergoing re-operation within a year, half of which were for recurrent herniation in the same location.
Dr. Loeser said he believes the complication rate was considerably lower than in the general diskectomy population, likely reflecting the high level of expertise of the neurosurgeons involved rather than the true risk, although such data are scarce. For example, a study in older Medicare patients found almost twice the rate reported in SPORT.
“The main issue here is that there is a profoundly excessive amount of low back surgery in this country compared to the world at large. Patients learn they have a ruptured disc and picture Hiroshima in their spine, and that's just nonsense. But what the primary care physician says almost always determines which path the patient will take with regard to treatment, and unfortunately this is all too often to have surgery.”
He said he hopes the SPORT findings will convince primary care physicians to be more conservative when advising treatment. “All an operation does is shorted the duration of the pain. If you can do it with pain meds, the patient will be better within two years without surgical risks.”
In an accompanying editorial, orthopedic surgeon Eugene Carragee, MD, Director of the Stanford University Spine Center, said that the crossover rates make the SPORT findings difficult to evaluate, but the data should help doctors and patients make better decisions about treatment.
“The fear of many patients and many surgeons that not removing a large disk herniation will likely have catastrophic neurological consequences is simply not borne out,” Dr. Carragee said.
In a second editorial, David Flum, MD, a neurosurgeon at the University of Washington in Seattle, noted that the known placebo effect in surgical interventions makes any non-surgical comparisons very difficult, and that only through sham-controlled trials can the effect be quantified and placed in proper context.
Michael Wang, MD, Assistant Professor in the Department of Neurological Surgery and Spine Director at the Keck School of Medicine at the University of Southern California in Los Angeles, disagreed.
“This is an important study because we've had a paucity of good outcomes data,” he told Neurology Today. He said that he was surprised by the results.
“When the trial design was first announced there was some pushback from the surgical community because the study design was viewed as biased toward the conservative treatment arm,” he told Neurology Today in a telephone interview. “We expected the results to be yet another study showing no difference in treatment approaches, yet the data from the trial clearly favor surgery for debilitating sciatica and in overall patient-reported improvements.”
All patients deserve conservative measures first, but depending on the type of disorder, eight weeks to six months is long enough to determine whether surgery is warranted, he said.
“There are definitely some spinal operations that fall into that gray area where efficacy rates may be questioned, but this isn't one of them.”