Medical Device Daily Washington Editor
If Shakespeare had known Hippocrates, Hamlet might have tormented himself with the question: “To operate or not to operate?”
As is the case for a number of conditions, the debate over surgical versus non-surgical treatment — in particular for various forms of spinal disease — has swung the pendulum of consensus back and forth. And a new report from the Spine Patients Outcomes Research Trial (SPORT) study suggests that for two diseases of the back, the pendulum is swinging yet again toward surgery.
In the Nov. 27, 2006, issue of Medical Device Daily we reported on an article that appeared in the Journal of the American Medical Association. That article, authored by a team led by James Weinstein, DO, the chairman of the department of orthopedic surgery at Dartmouth-Hitchcock Medical Center (Lebanon, New Hampshire), indicated that the researchers took an iffy view of the superiority of surgery, based on the results of the first arm of the SPORT trial, which examined the comparative effectiveness of surgery versus therapy. The abstract of that article stated that “conclusions about the superiority or equivalence of the treatments are not warranted, based on the intent-to-treat analysis.”
However, a statement issued recently in connection with the second arm of the SPORT trial suggests that further analysis may be changing minds.
Those in the first trial who agreed to take the operation “experienced slightly more improvement than those who opted for non-surgical treatments,” according to the statement.
The second phase of the SPORT trial, a five-year, multi-center study supported in part by the National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS), part of the National Institutes of Health , has now looked at the combination disease of degenerative spondylolisthesis and symptomatic spinal stenosis in more than 600 patients.
Spondylolisthesis is a degeneration of spinal discs, which can cause a narrowing of the inside of the vertebrae in the spinal column, known as stenosis.
According to the numbers offered by NIAMS, 372 of the 601 enrollees opted for decompressive laminectomy while the remaining 235 chose other treatments, including physical therapy, injections of anti-inflammatory steroids and analgesic medications. (Neither the press release nor the abstract specified which medications were prescribed.)
Two-year data from the second of the SPORT trials suggest that patients who elected non-surgical management “reported modest improvement in their condition,” but those who went under the knife “reported significantly reduced pain and improved function.”
The latter group also reported that “relief from symptoms came quickly; some reported significant improvement as early as six weeks after the procedure.”
In keeping with the notion that no clinical trial ever comes off without some sort of hitch, this leg of the SPORT study ran into a substantial amount of patient crossover: 40% of all patients opted out of the control group and into the study group or vice-versa, which means the analysis loses the statistical rigor of an intent-to-treat trial.
Observational studies are generally treated as less rigorous than intent-to-treat trials at FDA advisory hearings, especially by biostatisticians, but that view is apparently not universally held.
In an article in the 2000 edition of Blood Purification, Robert Wolfe, PhD, a professor of biostatistics at the University of Michigan (Ann Arbor), took the position that “[i]f our objective is to spend a lot of money in order to achieve a slow and narrow advance in knowledge,” randomized controlled trials should be used “exclusively.”
However, Wolfe also said that while correlation “is almost always a marker” for causation, the direction of causation is not always clear and that “inferences from observation can be sharpened by isolating the effects of individual factors by changing one of them at a time in a controlled experiment.”
The report on the second arm of SPORT seems to echo that sentiment, stating that pooled results from the intent-to-treat study and a parallel observational study may end up creating “a more powerful osbservational study at the expense of information gained from the statistically rigorous study design originally planned.”
The patients in that observational cohort numbered 743, with 528 choosing surgery over therapeutic management.
According to an abstract from the article published on that study, also in the Nov. 22, 2006 edition of JAMA, the patients in this group “with persistent sciatica from lumbar disk herniation improved in both operated and usual care groups,” but that those who elected surgery “reported greater improvements than patients who elected non-operative care.”
The authors said, however, that “non-randomized comparisons of self-reported outcomes are subject to potential confounding and must be interpreted cautiously.”
Patient crossover was also an issue in the first arm of the SPORT trial, that data indicating that patients who underwent surgery for herniated discs experienced slightly more improvement than those who opted for non-surgical treatments. The mixed signals of these studies may be resolved (or not) when preliminary data from the third major SPORT study – to look at the effectiveness of surgery compared to non-surgical options for spinal stenosis without spondylolisthesis — are scheduled to be released later this year.
Weinstein was not available to respond to calls for comment on the articles.