Skip Navigation LinksHome > Blogs > The Spine Blog
The Spine Blog
Friday, November 27, 2015

The spine literature is replete with papers documenting poor post-operative outcomes in both worker’s compensation (WC) patients and those on chronic opioid therapy (COT). The intersection of these two groups has been less well-studied, yet these two characteristics seem to go hand in hand. As such, Dr. Anderson and his colleagues from Case Western Reserve decided to query the Ohio WC database to determine the role of COT in WC patients undergoing lumbar fusion. They identified 1,002 WC patients undergoing lumbar fusion for degenerative disk disease (DDD) from 1993-2003 based on ICD-9 and CPT codes. Over 60% of these patients also underwent some type of decompressive procedure, suggesting that they had a nerve compression syndrome in addition to DDD. Forty-nine percent of these patients were on pre-operative COT (defined as opioid use for over one year), and 57% were on COT post-operatively. Not surprisingly, the strongest predictor of post-operative COT was pre-operative COT. Depression, a post-operative diagnosis of failed back syndrome, additional lumbar surgery following the index procedure, and a prolonged duration out of work pre-operatively were all significant independent predictors of post-operative COT. In patients without any of these risk factors, only 11% went onto post-operative COT. However, only 8% of this population had none of these characteristics. Only 23% of patients returned to work following fusion, including 11% of those on post-operative COT and 38% of the non-COT group. Overall, this study paints a relatively bleak picture for WC patients undergoing lumbar fusion, especially those on pre-operative COT.


While the findings of this study are not surprising to those who care for WC and COT patients, the authors should be congratulated for distilling a mountain of data down to some very salient points. Worker’s compensation patients on COT who are considering lumbar fusion should be informed that it is highly unlikely that they will either return to work or get off narcotics following a lumbar fusion. The authors concluded that “the outcomes of this study could suggest a more limited role for discogenic fusion among patients receiving WC.” While this may be true, the limitations inherent in this type of study preclude a strong conclusion to that effect. The population included in this study is likely heterogeneous in that nearly two thirds of the patients underwent some type of decompression, indicating that many of the patients were not having surgery for DDD alone. Additionally, no patient reported outcomes were available in this database, so no firm conclusions can be reached about the degree of improvement in pain or function. Finally, there is no non-operative comparison group, so it is unclear if these patients get a benefit from fusion relative to other treatment options. The Spine Patient Outcomes Research Trial (SPORT) did compare surgical and non-operative outcomes for lumbar disk herniation and spinal stenosis and found that while WC and COT patients had worse surgical and non-operative outcomes, both groups still had a significant benefit from surgery.1,2 If this is true for WC patients on COT undergoing fusion for DDD will remain unknown until a high quality comparative study is done in this population. What this study does make clear is that WC patients on COT will probably not return to work or get off opioids following a fusion. If these patients are aware of this during their decision making process, they may be less likely to opt for a fusion. The most important message from this paper is probably that these patients should never be started on opioid therapy following their work-related back injury given the poor outcomes this portends.


Please read Dr. Anderson’s article on this topic in the November 15 issue. Does it change how you view the role of opioids and lumbar fusion in the WC population? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor




1.            Pearson A, Lurie J, Tosteson T, et al. Who should have surgery for an intervertebral disc herniation? Comparative effectiveness evidence from the spine patient outcomes research trial. Spine 2012;37:140-9.

2.            Pearson A, Lurie J, Tosteson T, Zhao W, Abdu W, Weinstein JN. Who should have surgery for spinal stenosis? Treatment effect predictors in SPORT. Spine 2012;37:1791-802.


Friday, November 20, 2015

The benefits, risks, and indications for the use of bone morphogenetic protein (BMP) remain poorly defined. While there was great initial enthusiasm for a molecule that could improve fusion rates and decrease the need for autograft harvest, utilization of BMP for spinal fusion has decreased since the 2011 publication of a critical review by Carragee et al. that questioned the safety and efficacy of BMP.1 Since then, multiple other studies have been published evaluating data from the FDA trials on BMP and from other sources, and these have yielded conflicting results regarding the association between BMP and cancer.2-5 Given that this question remains unanswered, Dr. Malham and his colleagues from Australia elected to retrospectively review their data on 527 consecutive patients who underwent lumbar fusion using BMP in order to evaluate fusion rates and risk of cancer. They determined fusion rates using CT scans from 6-24 months postoperatively, and cancer risk was evaluated using a questionnaire and a statewide cancer registry. Cancer rates in the fusion patients were compared to standardized incidences determined using the registry data based on age and gender. Similar to prior studies, the authors reported high fusion rates (90% overall, 94% for interbody fusion, 85% for posterolateral instrumented fusion). Twenty-seven patients (5%) developed cancer over an average follow-up of 4.4 years, a rate that was 16% lower than expected for an age and gender matched cohort. Given the relatively low numbers involved, subgroup analyses based on type of cancer was not possible.


While this study adds data to the discussion about the association between BMP and cancer, the question remains unanswered. Given that BMP is a strong growth factor, there is a biologically plausible relationship between BMP and cancer. However, no strong data exists showing this to be true. The data from the FDA trials shows an increased risk of cancer among patients treated with high dose BMP-2, however, of the 20 cancers diagnosed in the BMP-2 group, 10 were skin cancers, and 4 of these were squamous cell carcinomas in one patient.2 If that single patient had been randomized to the non-BMP group and developed the four skin cancers, the difference would likely no longer be significant. A large Medicare database analysis showed no increased risk of cancer in lumbar fusion patients who received BMP compared to lumbar fusion patients who had not.4 None of these studies, including the current study which was underpowered and has only a database control group, was appropriately designed to answer the question. A very large RCT comparing cancer rates between lumbar fusion patients treated with and without BMP with strict surveillance for many years after surgery would be necessary to definitively determine if an association exists between BMP and cancer. In order to detect a 50% increase in cancer rate (i.e. 7.5% vs. 5%) with 95% confidence and 80% power, this would require approximately 3,000 patients to be randomized. It seems unlikely that such a trial will ever be performed. High quality registry data would likely provide the best evidence possible, though observational data is limited by potential selection bias and confounding. At this point, surgeons need to weigh the potential risks, benefits, and costs when considering the use of BMP for lumbar fusion. Fusion rates tend to be quite high for one and two level instrumented lumbar fusions using local bone graft alone, suggesting that little is gained by the use of BMP in this situation.6,7 It seems like the risks and costs of BMP might be justified in challenging fusion environments such as long thoracolumbar fusions in adult deformity patients or in established non-unions.


Please read Dr. Malham’s paper on this topic in the November 15 issue. Does this change how you view the relationship between BMP and cancer? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor





1.            Carragee EJ, Hurwitz EL, Weiner BK. A critical review of recombinant human bone morphogenetic protein-2 trials in spinal surgery: emerging safety concerns and lessons learned. The spine journal : official journal of the North American Spine Society 2011;11:471-91.

2.            Carragee EJ, Chu G, Rohatgi R, et al. Cancer risk after use of recombinant bone morphogenetic protein-2 for spinal arthrodesis. The Journal of bone and joint surgery American volume 2013;95:1537-45.

3.            Fu R, Selph S, McDonagh M, et al. Effectiveness and harms of recombinant human bone morphogenetic protein-2 in spine fusion: a systematic review and meta-analysis. Annals of internal medicine 2013;158:890-902.

4.            Kelly MP, Savage JW, Bentzen SM, Hsu WK, Ellison SA, Anderson PA. Cancer risk from bone morphogenetic protein exposure in spinal arthrodesis. J Bone Joint Surg Am 2014;96:1417-22.

5.            Simmonds MC, Brown JV, Heirs MK, et al. Safety and effectiveness of recombinant human bone morphogenetic protein-2 for spinal fusion: a meta-analysis of individual-participant data. Annals of internal medicine 2013;158:877-89.

6.            Inage K, Ohtori S, Koshi T, et al. One, two-, and three-level instrumented posterolateral fusion of the lumbar spine with a local bone graft: a prospective study with a 2-year follow-up. Spine 2011;36:1392-6.

7.            Sengupta DK, Truumees E, Patel CK, et al. Outcome of local bone versus autogenous iliac crest bone graft in the instrumented posterolateral fusion of the lumbar spine. Spine (Phila Pa 1976) 2006;31:985-91.



Friday, November 13, 2015

Three column fractures through an ankylosed spine are known to be highly unstable and associated with high rates of neurological deficit and mortality. Ankylosing spondylitis (AS) and diffuse idiopathic skeletal hyperostosis (DISH) are the two most common causes of spinal ankylosis, and they both tend to present in older patients. These elderly, frail patients are at increased risk of falls leading to fractures, and this population has a higher rate of mortality following spinal cord injury than younger, healthier patients. While prior studies have reported high rates of paralysis and death associated with spinal fractures in AS, most of these were relatively small case series. Given the lack of a large scale study on the topic, Dr. Robinson and colleagues from Sweden queried the Swedish Patient Registy to identify AS patients who were treated for a spinal fracture from 1987 to 2011. They then cross-referenced the Swedish mortality register to determine survival for their cohort of 990 AS fracture patients (including a total of 1,131 fractures). The three month mortality was 17% for all patients. The rate of surgery for the injury increased from just over 30% at in 1987 to approximately 70% by 2011. The authors created a Cox Proportional Hazards model to determine individual characteristics associated with mortality, and they found that increasing age, spinal cord injury, male gender, increasing comorbidity burden, and non-operative treatment were all associated with higher mortality. Surgical treatment decreased the risk of death by 21%. Reassuringly, the mortality rate decreased significantly over the years of the study.


It is uncommon to see a published paper in orthopaedics or spine surgery reporting a significant improvement in mortality. However, for conditions such as unstable fractures through an ankylosed spine in an older population, this is not surprising. If surgical treatment is able to prevent paralyisis in some cases, that will likely lead to improvements in mortality. Before jumping conclusions beyond the scope of the data, the limitations of this paper need to be considered. It shares the same limitations inherent in all retrospective database studies, the most important being the risk for selection bias. While the two patient groups were relatively similar on paper, no randomization occurred, and surgeons may have been selecting healthier patients with a lower risk of mortality for surgery. Additionally, the database did not include information on the type of fracture, which also likely influenced the treatment decision. These limitations are important, but the message that stabilizing three column fractures in AS likely leads to better outcomes is clear. Non-operative treatment—essentially bedrest--is typically tolerated poorly by elderly patients and is generally contraindicated. This study adds to the literature that is now becoming quite consistent on this topic. The authors should be congratulated on assembling such a large cohort that they could follow with high quality mortality data.

Please read Dr. Robinson’s article on this topic in the November 1 issue. Does this change your view on how unstable fractures in AS should be treated? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor

Friday, November 06, 2015

Inappropriate early use of lumbar MRI for acute low back pain (LBP) in the absence of “red flags” has been well-documented and is associated with higher rates of invasive treatment such as injections and surgery. High levels of geographic variation have been shown for “preference-sensitive” medical care, and the decision about ordering a lumbar MRI early in the course of LBP represents a preference-sensitive decision. Geographic variation in the use of lumbar MRI for acute LBP has not been well-studied, though one would expect high levels of variation. A more interesting question is what factors could be driving this variation, and Dr. Pransky and his colleagues from Massachusetts elected to explore this. They used the nationwide Liberty Mutual worker’s compensation (WC) database to identify WC patients who received a disability payment related to an episode of acute LBP. Patients who underwent lumbar MRI within 30 days of the date of injury were identified, after excluding patients who had diagnoses associated with “red flags” (i.e. cancer, fracture, infection, cauda equina syndrome, etc.) The rate of early MRI ranged from 6% in Vermont to 58% in Arkansas. After stratifying patients according to severity of LBP, they found an even greater degree of variation in the MRI rate among the “low severity” patients, ranging from 2% in Vermont to 53% in Arkansas. While a high level of variation was expected, this magnitude of variation is striking. In the most interesting part of the analysis, the authors created a multivariate model to determine which factors were associated with the variation in MRI rate between the 6 states with the highest rates and the 6 states with the lowest rates. Their final model included median state income, proportion of MRI scanners owned privately, and severity of LBP, with the model explaining 84% of the interstate variation.


Most clinical practice guidelines suggest that the majority of lumbar MRIs performed within 30 days of the onset of acute LBP in the absence of red flags are inappropriate, especially in the “low severity” cases, which represented the majority of cases in the current study. While geographic variation has been studied extensively across most medical conditions, this is one of the first geographic variation studies to focus on what essentially amounts to inappropriate care. While high levels of geographic variation are the norm rather than the exception for preference sensitive care, a 25-fold variation across states for inappropriate early MRI for “low severity” LBP merits attention. The variation is likely even higher across smaller units of analysis such as hospital referral regions. The factors associated with the variation were interesting, and the authors suggested that less sophisticated patients (using state median income as a proxy for this) and physicians with a financial interest in MRI scanners (using proportion of scanners owned privately as a marker for this) were likely the drivers of the higher rate of inappropriate early MRI use. While this may be true, median income and rate of privately owned MRI scanners may be markers for other factors driving the use of MRI. Patients and providers in states with low median income and high rate of physician-owned MRI scanners live and work in medical cultures that are markedly different from the states with low MRI utilization rates. As a spine surgeon who cares for WC patients with LBP in two of the lowest use states (NH and VT), I still observe what seems to be a fairly high rate of early MRI use and inappropriate early use of narcotics, injections, and even surgery in this population. I imagine the medical culture in places like Arkansas and Florida is significantly different than in Northern New England. Geographic variation studies tend to conclude with statements about the need to better understand the variation and how better evidence is required to help reduce unwarranted variation. In this case, there is strong evidence that early MRI is not indicated for acute LBP, and the current study suggests some risk factors for inappropriate use of early MRI. Given that physicians do not seem to respond to current clinical practice guidelines, it is not surprising that payors are creating policies to prevent inappropriate use of medical resources. Unfortunately, their approach is to create bureaucratic hurdles to obtaining care, which frequently impedes the ability of patients and physicians from obtaining care even when it is appropriate.

Please read Dr. Pransky’s article on this topic in the November 1 issue. Does this change how you view the use of early MRI in acute LBP? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor

Friday, October 30, 2015

The spine surgery literature has traditionally viewed degenerative spondylolisthesis (DS) patients as a homogenous population. However, all spine care physicians recognize the substantial heterogeneity in both patient and disease characteristics that exist in this group. There is wide variation in age, medical comorbidities, mobility at the affected level, the degree of slip, the location of stenosis (central vs. foraminal), disk height, facet morphology, and pain location (back vs. leg). Since the classic Herkowitz and Kurz paper showing better outcomes for patients who underwent decompression and fusion versus decompression alone, the former has become the standard treatment for DS.1 While this may be appropriate for the “average” DS patient, there are likely some patients who will do just as well with decompression alone, while others might benefit from an interbody fusion. Unfortunately, the literature offers no guidance on how to select the most appropriate surgical treatment. Given this knowledge void, Dr. Schroeder and his colleagues in Philadelphia performed a survey of LSRS and AOSpine members to ascertain how they take individual patient characteristics into account while determining a surgical treatment plan for DS. They queried the 223 respondents about the role of age, intervertebral motion, slip magnitude, pain location, and disk height in treatment selection. For a 75 year old patient with neuogenic claudication, minimal back pain, a low grade slip, and no motion on flexion-extension x-rays, 53% selected a decompression without fusion. For the same scenario in a 50 year old patient, only 29% selected a decompression alone, indicating that age plays a significant role in treatment selection. For cases in which fusion was selected, an interbody fusion (PLIF or TLIF) tended to be the most popular treatment option, with posterolateral instrumented fusion the next most common response, and uninstrumented fusion being selected by fewer than 10% of respondents for any scenario. The biggest driver for selecting interbody fusion was the need for complete facetectomy, while a collapsed disk space was the strongest contra-indication to interbody fusion.


The authors should be applauded for drawing attention to an important topic that has gained scant attention. While surgeons seeing DS patients know that most of them are best served with a decompression and fusion, there is little to no guidance from the literature about how to perform that fusion or determine who might do well with a decompression alone. This survey revealed that a PLIF/TLIF is the most popular fusion technique for DS, and a prior review of the American Board of Orthopaedic Surgery database also demonstrated this.2 However, there is little or no evidence supporting interbody fusion for DS. The Spine Patient Outcomes Research Trial did show some advantages for interbody and instrumented fusion over uninstrumented fusion for the first two years, but by four years the patient-reported outcomes were essentially the same regardless of fusion technique.3 This study did show a higher fusion rate with instrumented or interbody fusion. Despite the lack of evidence supporting interbody fusion for DS, surgeons have widely adopted this technique. It is unclear what drives this decision-making, though it could be related to surgeon belief that higher fusion rates lead to better outcomes. A more cynical observer might suggest that it is related to increased reimbursement for interbody fusion. This study does a nice job shining a light on the wide variation in surgical decision-making for DS, which reflects the lack of strong evidence supporting specific surgical techniques based on individual patient factors. This variation is likely to persist until further research is done to answer the question.


Please read Dr. Schroeder’s article on this topic in the November 1 issue. Does this change how you choose a surgical technique for DS? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor



1.            Herkowitz HN, Kurz LT. Degenerative lumbar spondylolisthesis with spinal stenosis. A prospective study comparing decompression with decompression and intertransverse process arthrodesis. The Journal of bone and joint surgery 1991;73:802-8.

2.            Kepler CK, Vaccaro AR, Hilibrand AS, et al. National trends in the use of fusion techniques to treat degenerative spondylolisthesis. Spine (Phila Pa 1976) 2014;39:1584-9.

3.            Abdu WA, Lurie JD, Spratt KF, et al. Degenerative spondylolisthesis: does fusion method influence outcome? Four-year results of the spine patient outcomes research trial. Spine (Phila Pa 1976) 2009;34:2351-60.


About the Blog

Spine Journal
This Blog provides a forum for discussion about high impact articles published in Spine, including the bi-annual publication of "Evidenced-Based Recommendations for Spine Surgery." Website users can use this forum to discuss how the articles have affected their practice and query the authors about their findings and recommendations.