NASS responds to New England Journal of Medicine article, " Spinal-fusion surgery - the case for restraint."
(Original article published February 12, 2004; response published July 25, 2004)
By Thomas J. Errico, MD [1], Robert J. Gatchel, PhD [2], Jerome Schofferman, MD ]3], Edward C. Benzel, MD [4], Thomas Faciszewski, MD [5], Marjorie Eskay-Auerbach, MD, JD [6], and Jeffrey C. Wang, MD [7]
[1] NYU-HJD Department of Orthopaedic and Neurological Surgery, NYU School of Medicine, New York, NY
[2] Department of Psychology, College of Science, The University of Texas at Arlington, TX [3] SpineCare Medical Group, San Francisco Spine Institute, Daly City, CA
[4] Department of Neurosurgery, the Cleveland Clinic Foundation, Cleveland, OH
[5] Marshfield Clinic, Marshfield Center, Marshfield, WI
[6] Center for Spine Care, PC, Tucson, AZ
[7] Orthopaedic Spine Service, University of California at Los Angeles, Los Angeles, CA
Sentiments expressed herein were unanimously approved by the NASS Board of Directors.
FDA device/ drug status: Approved for this indication: pedicle screw; artificial disc replacement.
Author Disclosures, which may indirectly relate to the subject of this manuscript:
- Author ECB acknowledges a financial relationship (Research grant from Depuy-Acromed, Nuvasive Spinal Concepts, Mazor, Surgical Technologies Ltd Scientific Advisory Board; consultant for Depuy-Acromed, Medtronic, Midas Rex, Proncuron, Axiomed, Spine Wave, Mazor, Surgical Technologies Ltd Scientific Advisory Board; stock shareholder for Depuy-Acromed, Axiomed, Nuvasive, Orthovita, Spine Wave; honorarium from Depuy-Acromed, Mazor, Surgical Technologies Ltd Scientific Advisory Board; royalties from Depuy-Acromed, Spincal Concepts, Axiomed, Mazor).
- Author TJE acknowledges a financial relationship (Fellowship support from Synthes Spine, Medtronic Sofamor Danek; grant research support from Spine Solutions, SpineCore; consultant for Fastenetix, SpineCore; stockholder and board member for SpineCore).
- Author JS acknowledges a financial relationship (Speaker’s Bureau for Merck, Pfizer, Medtronic)
- Authors MEA, RJG, TF and JCW: Nothing of value received from a commercial entity related to this research.
Members of the North American Spine Society (NASS) have expressed concern regarding recent commentary on spine surgery by Deyo, Nachemson and Mirza, published in the New England Journal of Medicine [1]. These authors question the value of performing spine fusion surgery for pain from degenerative conditions and immediately following cervical discectomy. They also criticize the benefit of surgical implants, and imply that spinal fusion is rarely, if ever, indicated for conditions other than for severe scoliosis, spondylolisthesis, spinal tuberculosis and fractures. There is a need for a balanced presentation of these and other important spine care topics. The commentary by Deyo et al. [1] prompted a voluntary multidisciplinary group from the Board of the North American Spine Society to assess the commentary in detail, including its clinical critiques of spine surgery and all of the references accompanying the article. This multidisciplinary panel of authors practice within the fields of clinical psychology, internal medicine, neurosurgery and orthopaedic spine surgery. As concerned members of the spine care community and practicing health care professionals, we would are obligated to highlight some of the specific weaknesses in the Deyo et al. [1] commentary and add important caveats regarding their reasoning or oversights that resulted in some of their misleading conclusions on spine care. Specifically, the article had numerous stated or implied observations on spine fusion surgery with regard to:
- the “overuse” of spine fusion surgery
- comparisons of total hip replacement surgical rates to spinal fusion surgery rates
- geographic variation of the prevalence rate of surgical fusion procedures
- the controversy over fusion for discogenic pain
- the “overuse” of spinal instrumentation
- lack of justification for the use of instrumentation to aid fusion
- complication rates with the use of spinal instrumentation
- the published literature for the evaluation of spinal fusion surgery
- cervical fusion after cervical discectomy
- vision loss following spinal surgery
- placebo-controlled trials: sham surgery
- evolving medical technologies and the “learning curve” of surgeons
The claims of the article were supported by a select literature review and attempted to narrow the role of fusion in spine surgery to conditions limited to those proven by evidence based standards. On the face, this appears to be a reasonable suggestion. The article, however, presented a nonsystematic evaluation of selected literature and did not employ a meta-analysis or other objective scoring criteria by which the methodologies of the reference studies could be impartially rated. Our comments are organized around the above mentioned issues.
Overuse of Spine Fusion Surgery
These authors suggest the procedure of fusion may be overused. Furthermore, they state that much of the increase in fusion rates results from a rise in surgeries on older adults with spinal stenosis, as well as from an increasing rate of fusion surgeries for discogenic pain. Deyo et al. [1] strongly suggest that fusion operations are performed in excess and lay the groundwork for implicating a possible profit motive as the reason. There are no data, however, to prove that the increase in spinal fusion is not legitimate on its own terms. It is certainly possible that spine fusion was underutilized prior to 1997 rather than overutilized today. As opposed to ten or twenty years ago, spine specialists now have superior diagnostic tools, better understanding of the structural causes of lower back pain (LBP) and realize the dangers of introducing instability after wide decompression for severe foraminal stenosis. As well, we have vastly improved procedural techniques for fusion [2-18].
As our population ages, it is natural to expect that surgical rates related to degenerative disorders will increase commensurately. This includes both spine fusion and decompressive procedures. While spine surgery rates for the elderly have increased, this is paralleled by a marked rise in the average age of elderly patients undergoing surgery for other medical conditions such as replacement of the aortic valve, repair of abdominal aortic aneurysm, removal of cancerous bladder, coronary artery bypass and removal of a portion of pancreas [19].
It is also reasonable to recognize, as a factor, the public’s gradual acceptance of spine surgery---and thereby, patient acceptance. This acceptance came as spine surgery improved in safety and with the evolution of technologies that provided increasing success, returning patients to their prior lifestyles and activities from a virtually inactive life. This type of transformation was rare to unheard-of twenty years ago and suggests the field of spine care is maturing from the treatment of only catastrophic injury or disability to include treating more moderate disabilities of activities of daily living. Today’s spine surgery patients often do not consider surgery as the last resort, but as the best solution after failure of conservative therapy.
Any and all of these factors are reasonable to consider as legitimate contributions to observed increases in fusion rates in recent decades. The extent of contribution can only be known by carefully considered multidisciplinary research.
Comparisons of Total Hip Replacement Surgical Rates to Spinal Fusion Surgery Rates
Deyo et al. [1] also compared the rates of spinal fusion to total hip replacement (THR), noting the significantly higher percentage growth in spinal fusion operations over the growth rate of total hip replacement. Certainly, total hip replacement represents a mature technology with already widespread penetration. In contrast, the spinal fusion technology is newer, more diverse in nature, and at a much earlier growth phase than THR. It is also noteworthy that while there are good prospective outcome studies, there is a distinct lack of prospective randomized controlled trials of THR that meet the evidence based medicine (EBM) standards Deyo et al. [1] would require of spinal fusion surgery. Total joint replacement surgery is somehow afforded the luxury of being “grandfathered in.” A recently presented prospective study comparing the outcomes of spine surgery to total hip replacement and total knee replacement showed that the operated spine patients had an astonishingly low preoperative quality of life compared to total joint replacement patients. Spine surgery dramatically improved their quality of life. Interestingly, within the spine surgery group, a higher outcome score was found in the posterior fusion subgroup than in the “less controversial” laminectomy or discectomy subgroups. In fact, the posterior lumbar fusion subgroup had the same final outcomes at one year as the total knee replacement (TKR) group. The authors felt the utility of spine surgery was equal to or better than that for THR or TKR in a one-year perspective [20].
The Geographic Variations of the Prevalence Rates of Surgical Fusions
The geographic variations of the prevalence rate of surgical procedures are reported by Deyo et al. [1] to result from uncertainty in diagnosis and varying beliefs regarding appropriate indications for surgery; both suggest a lack of consensus. Geographic variations do occur, and for some conditions, do represent a need for further research. However, the author of the original study [21] concedes that this variation may be a result of factors such as differences in the epidemiology of the disease, sociodemographic factors and various types of medical care--seeking behavior across different regions. Varying treatment for the same condition may be a result of the prevalence of small hospitals in some of the regions and the absence of local training programs. Variations in other countries may also be a result of insurance issues and access to medical care that significantly differs from country to country (eg, socialized vs. nonsocialized medicine environments). Furthermore, databases used for analysis have limitations that are sometimes not placed in proper context during research. For example ICD-9CM diagnosis and perioperative complication coding for spine procedures include much information in the form of "NOS" codes, or "not otherwise specified" codes---when only spine procedures are being evaluated. As many as 30% of codes contained in ICD-9CM databases, such as those used for research purposes similar to Deyo et al. contain NOS codes [22-23]. This lack of information, as well as misinformation, may mislead the reader and can lead to erroneous statements about the benefits of spine procedures in terms of comorbidities, outcome and complications.