Medicare Panel Debates Spinal Fusion Study Design
BY JOHN MCCORMICK, DECEMBER 1, 2006
Yesterday we attended the the Medicare Coverage Advisory Committee (MCAC) meeting on spinal fusion at the Centers for Medicare and Medicaid Services (CMS) in Baltimore, MD. The purpose of the meeting was to determine the type of data needed to construct the ideal spinal fusion study and whether or not the current studies are adequte. The outcome of the meeting? Inconclusive and somewhat confused. The outlook for the spinal implant industry? Benign. We do not think a non-coverage decision is imminent.
This MCAC was a panel of clinical experts in spine appointed by CMS to discuss and vote on a list of six questions and discussion points surrounding spinal fusion studies and study design. Before getting to the vote, it was a day of discussion and sometimes lively debate between the panelists about the treatment of low back pain, the adequacy of current studies and the definition of non-operative care, among other things.
Current Data
We found the first study presented by Dr. Douglas McCrory of Duke University to be among the more interesting presentations, because it was a pretty decent effort at producing a meta-data analysis (or literature review) of the recent universe of papers on spinal fusion. His first observation? That of the approximately 1,391 available citations, only 82 met the inclusion criteria for the literature search. Of these 82, only 4 applied to lumbar fusion for axial back pain. According to McCrory, these papers showed statistically significant reductions in the Oswestry Disability Index as a result of spinal fusion. So that means spinal fusion is good, right? You could say that, but the question of the relative performance of spinal fusion to non-operative care is where the kernel of the debate lies. These results appear to be statistically more ambiguous. In many studies, fusion shows superiority to non-operative care, but the statistical significance is questionable.
As a note aside, we were surprised that no members of the non-operative care community (such as chiropracters or rehab workers) came to the meeting to speak up and that there is a real paucity of data on conservative care and non-operative techniques.
Study Design Is the Issue
If yesterday's MCAC had a real takeaway concerning the current body of the available data on fusion, it would be that the current study designs are flawed. How so? The problem is that back pain is an elusive phenomenon. Osteochondral defects of the hip or knee are pretty easy to pinpoint, but pain generators in the back are extremely difficult to identify.
Some of the current study-design issues raised included the following:
- Non-operative care as a control is poorly defined in the studies
- Non-operative care data quality is poor at best
- Degenerative disc disease (DDD) may be accompanied by multiple pathologies. Isolating stand-alone DDD cases is difficult.
- Few studies have been done on patients over 65 years of age who are the prime medicare beneficiaries
- Workers comp cases bias results
- Very little is known about how much non-operative care patients received prior to undergoing fusion in the studies
- No studies have been done concerning fusion when non-operative techniques have failed
The Key Questions
The MCAC was asked to express their level of confidence concerning the following six questions through a vote at the end of the day. Here are the questions in summary form:
#1 - What level of confidence does the evidence provide in addressing the outcomes needed to determine the effectiveness of lumbar spine fusion for low back pain due to lumbar degenerative disc disease? Is pain relief the appropriate primary outcome or should it be restoration of function, return to work or something else?
#2 - What level of confidence does the evidence provide for characterizing the complications, adverse events, and other harms from lumbar spinal fusion for DDD, for both the short term (2 yrs or less) and long term (> 2 yrs)?
#3 - Based on the evidence presented, how likely is it that lumbar spine fusion for lumbar DDD improves clinical outcomes as compared with conservative treatment in the short term and long term?
#4 - Based on the evidence presented, how likely is it that the various fusion procedures improve health outcomes for lumbar DDD? Panel members voted on a list of procedures including ALIF, PLIF, etc.
#5 - What level of confidence does the evidence provide that radiographic interpretations are correlated with clinical outcomes for lumbar spine fusion due to lumbar DDD? Is there a uniform agreement regarding terminology for radiographic interpretations?
#6 - Based on the evidence presented, how likely is it that the results generalize to the Medicare population for relief of pain and for complications, adverse events and other harms?
Votes tended toward inconclusive and the MCAC was divided over whether data from various clinical trials apply to elderly and disabled patients who are Medicare beneficiaries.
In short, fusion on an stand-alone basis has been shown to improve pain and improve the patient's life. In order to prove that fusion has superiority to non-operative care, better and more granular clinical trials need to be developed. In our view, the meeting did not cause concern that CMS would issue a non-coverage decision. Rather, this was an exploration on how to better design studies.
