The Spine Blog

Friday, September 15, 2017

Debate about the need for fusion in degenerative spondylolisthesis (DS) has continued since Herkowitz and Kurz’s seminal 1991 paper reporting better outcomes in DS patients who underwent decompression and fusion compared to decompression alone.1 Two recent RCTs randomized DS patients to decompression and fusion or decompression alone and reached essentially opposite conclusions.  The Swedish study found no differences in patient reported outcomes or reoperation rates, while the American study reported significantly better patient reported outcomes and a lower reoperation rate in the decompression and fusion group.2,3 In order to add more data on this topic to the literature, Dr. Ulrich and colleagues from Switzerland used data from the Lumbar Stenosis Outcome Study to compare outcomes between DS patients treated with decompression and fusion and decompression alone.  They identified a cohort of 131 DS patients who had undergone surgery and had follow-up for at least 12 months (46 had outcomes out to 3 years). The decision to perform fusion in addition to decompression was made by the treating surgeon, and approximately 1/3 of patients underwent a posterolateral instrumented fusion with or without interbody fusion. The fusion patients were significantly younger and more likely to undergo a single level procedure. Unadjusted analyses revealed significantly more improvement on the Spinal Stenosis Measure for the fusion group, but after adjustment for baseline characteristics, the differences were no longer significant. The reoperation rate was 4.3% (n=2) in the fusion group and 9.4% (n=9) in the decompression only group. The vast majority of reoperations (7/9) in the decompression alone group were for restenosis at the index level.

This paper adds some interesting data to the debate about the role of fusion in DS. However, the limitations of a small, non-randomized study need to be considered before drawing any strong conclusions. There were major differences in baseline characteristics in the two cohorts, namely that the decompression only patients were older and had more multilevel disease. The authors presented no data about radiographic characteristics like disk height, facet alignment relative to the sagittal and coronal planes, motion on flexion-extension x-rays, and degree of disk degeneration that may have influenced surgeons’ decision making about the need to perform fusion. As such, there were likely many unmeasured confounders that could not be taken into account with statistical analysis. Additionally, very few patients reached 3 year follow-up, and the precision of models controlling for multiple covariates decreases as the number of patients decreases. Overall, this paper suggests that many DS patients do quite well with a decompression alone, and it is possible that the surgeons may have been able to select those patients based on factors not reported in this paper. Similarly, certain patients may do better with a fusion, and performing a fusion seems to reduce the risk of reoperation at the index level. At this point, the scientific literature does not provide guidance on how to select which DS patients will benefit from fusion in addition to decompression, and surgeons will continue to rely on their clinical judgment to make this decision. In the United States, most surgeons continue to perform decompression and fusion for DS, possibly due to the desire to avoid difficult revision surgery at a previously decompressed level. Hopefully future research will provide evidence to help surgeons select the right operation for DS patients based on their individual characteristics.

Please read Dr. Ulrich’s article on this topic in the September 15 issue. Does this change how you view the need for fusion in 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.            Forsth P, Olafsson G, Carlsson T, et al. A Randomized, Controlled Trial of Fusion Surgery for Lumbar Spinal Stenosis. The New England journal of medicine 2016;374:1413-23.

3.            Ghogawala Z, Dziura J, Butler WE, et al. Laminectomy plus Fusion versus Laminectomy Alone for Lumbar Spondylolisthesis. The New England journal of medicine 2016;374:1424-34.


Friday, September 8, 2017

Spinal osteomyelitis in the absence of neurological deficit, structural instability, or uncontrolled sepsis is typically treated with antibiotics. Some patients ultimately fail to clear the infection with antibiotics alone and require surgical debridement. When to proceed with surgical treatment for spinal osteomyelitis is frequently a judgment call in the absence of the aforementioned indications, and the benefits of debridement need to be weighed against the risks, including wound healing complications, iatrogenic meningitis from a dural tear, instability if fusion is not performed, and the physiological insult of surgery in a compromised patient. Being able to predict which patients are likely to fail antibiotic treatment could provide surgeons with more information while making the decision about the need for surgical debridement. In order to address this question, a team from Massachusetts General Hospital performed a chart review of all patients diagnosed with spinal osteomyelitis from 2001 through 2014. They identified 215 patients who were initially treated with antibiotics. They defined treatment failure as undergoing surgical debridement or death from infection and reported a 29% failure rate, including two patients who died of sepsis. In a multivariate analysis, they found that diabetes, osteomyelitis at another site, and presence of an epidural abscess were all independent predictors of antibiotic treatment failure. The strongest predictor was osteomyelitis at another site, with a hazard ratio of 8.1. However, only three patients presented with osteomyelitis at another site, so the confidence around that estimate was quite large.

The authors have done a nice job developing a large retrospective cohort of over 200 patients with spinal osteomyelitis. This likely represents one of the largest case series assembled for this diagnosis, and it allowed them to perform some statistical analysis that is not possible with smaller case series. Their independent risk factors are all well known risk factors for persistent infection. About 23% of patients in the series had an epidural abscess. While the exact number of patients with epidural abscess who failed antibiotic treatment is not reported, the hazard ratio is about 2, indicating approximately twice the risk of antibiotic failure than those without epidural abscess. Nonetheless, this paper suggests that neurologically intact patients with epidural abscess can frequently be treated successfully without surgery, so epidural abscess is not an absolute indication for debridement. The major limitation of this paper is that the decision to proceed with surgical debridement was not based on specific criteria, so the definition of failure is somewhat loose. It is possible that some of the patients treated surgically could have been successfully treated with further antibiotics. However, this would be a difficult topic to study in a prospective fashion due to the low numbers involved, and it does not lend itself to randomization. While this paper does not provide surgeons with clear cut criteria that can guide the decision to proceed with debridement of a spinal infection, it does offer information about the patient and disease characteristics that increase the risk of antibiotic failure.

Please read this paper by de Graeff et al. in the September 1 issue. Does this change your perspective on risk factors for antibiotic treatment failure in spinal osteomyelitis? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor

Friday, September 1, 2017

Sagittal imbalance results in worse pre-operative and post-operative patient reported outcomes, and deformity surgeons have been more aggressive about correcting sagittal parameters over the past decade.1 Unfortunately, larger magnitude procedures including osteotomies and long fusions to the pelvis are frequently required to achieve sagittal balance, and these procedures are associated with greater risk of complications both in the perioperative period as well as in the long-term. Initial efforts at correcting sagittal imbalance were focused on restoring parameters to norms derived from a relatively young, asymptomatic population. As experience with these procedures has developed, surgeons have noted that overaggressive correction of sagittal imbalance in the elderly population was frequently associated with mechanical failure of the fusion or proximal junctional kyphosis (PJK).2 As such, age-appropriate correction targets were developed to take into account that kyphosis normally increases with aging and is not necessarily pathologic.3 To better understand the effect of aggressive correction on PJK, Lafage et al. analyzed the International Spine Study Group’s database of adult deformity cases that included 679 patients. All patients had either a coronal Cobb angle greater than 20 degrees, sagittal vertical axis (SVA) greater than 5 cm, pelvic tilt (PT) greater than 25 degrees, or thoracic kyphosis greater than 60 degrees. All patients had fusion to the pelvis and had follow-up radiographs for at least one year post-operatively. They defined PJK as a kyphotic angle greater than 10 degrees between the upper instrumented vertebra (UIV) and the vertebra two levels cranial to the UIV and an increase in that angle of at least 10 degrees from the initial post-operative film to the follow-up image. Overall, they found that 45% of patients developed PJK. The PJK patients had a significantly lower pelvic incidence – lumbar lordosis (PI – LL) mismatch than the non-PJK patients (2 degrees vs. 8 degrees) and also had a lower SVA (32 mm vs. 41 mm). The PI – LL mismatch was significantly lower for PJK patients across all age groups. They also found that PJK patients were “overcorrected” when compared to age appropriate PI – LL mismatch and SVA.

This paper adds to the literature suggesting that overcorrection of sagittal imbalance, especially in the elderly population, leads to a higher rate of PJK and likely failure. This paper included only radiographic data and no patient reported outcomes, so it is not clear if the patients who were “undercorrected” did as well clinically as those who were corrected to a greater degree. Additionally, the authors did not report reoperation rates, so it is unclear how many of these patients underwent revision surgery to address the PJK or how many developed neurological deficits. Nonetheless, radiographic PJK is likely a reasonable surrogate for worsening symptoms and implies an increased risk of pain, neurological deterioration, and reoperation. In terms of the need to aggressively correct elderly patients to “normal” sagittal alignment, the pendulum seems to be swinging back towards achieving an acceptable alignment without putting unreasonable demands on hardware and the patient’s skeleton. Future studies that look at the relationship between degree of correction and patient reported outcomes stratified by age will further inform surgeons about the degree of correction they should seek to obtain in elderly patients.

Please read Dr. Lafage’s article on this topic in the September 1 issue. Does this change how you view the need for aggressive correction of sagittal imbalance in the elderly population? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor



1.            Glassman SD, Bridwell K, Dimar JR, Horton W, Berven S, Schwab F. The impact of positive sagittal balance in adult spinal deformity. Spine 2005;30:2024-9.

2.            Hallager DW, Karstensen S, Bukhari N, Gehrchen M, Dahl B. Radiographic Predictors for Mechanical Failure After Adult Spinal Deformity Surgery: A Retrospective Cohort Study in 138 Patients. Spine (Phila Pa 1976) 2017;42:E855-E63.

3.            Lafage R, Schwab F, Challier V, et al. Defining Spino-Pelvic Alignment Thresholds: Should Operative Goals in Adult Spinal Deformity Surgery Account for Age? Spine (Phila Pa 1976) 2016;41:62-8.


Friday, August 25, 2017

While it is clear that the rate of lumbar fusion for degenerative conditions has increased significantly over the past two decades, considerable uncertainty remains about the indications for lumbar fusion, especially at the individual patient level. Most surgeons agree that spondylolisthesis, scoliosis, and adjacent segment disease are reasonable indications for fusion, though how to take other patient factors like smoking, obesity, and psychosocial comorbidities into consideration is less clear. In an effort to ensure that their fusion patients meet indications for surgery and have received appropriate non-operative treatment, the spine group at Virginia Mason Medical Center in Seattle convened a multidisciplinary conference including physiatrists, pain specialists, neurosurgeons, orthopedic surgeons, and nursing staff to determine if patients recommended for fusion are good candidates. For this study, they evaluated 100 consecutive patients who had been referred to their institution for a second opinion after an outside surgeon had recommended a lumbar fusion including one to three levels. All patients were presented at the multidisciplinary conference, and the panel then voted to determine if fusion was indicated. The exact breakdown of diagnoses is somewhat unclear, though they reported that 20% had spondylolisthesis, 18% spinal stenosis without listhesis, 10% degenerative disk disease with axial back pain, 6% adjacent segment degeneration, 4% pseudarthrosis, and 4% were misdiagnosed and had hip arthritis as the actual cause of their symptoms. Of the 100 patients, the panel recommended that 58 receive no lumbar surgery. Reasons for avoiding surgery included misdiagnosis in 4, morbid obesity in 10, smoking in 10, need for further PT in 22, recommendation for epidural steroid injection in 6, and recommendation for vertebroplasty in 2. Of the 42 who the panel felt met the indications for surgery, 6 received a decompression alone, and 36 underwent fusion with or without decompression. The authors concluded that such a panel “may provide appropriate care, ultimately improving the value of spine care by removing or reducing the cost of surgery and by improving clinical patient outcomes.”

The authors describe a novel approach to determining which patients are appropriate candidates for lumbar fusion, a decision that is traditionally made by a spine surgeon in conjunction with the patient. They have shown clearly that the panel and the outside surgeon disagreed about the indications for fusion in the majority of cases. The exact reasons behind the disagreements get obscured by combining the data for the purposes of analysis and presentation, but the panel automatically ruled out surgery for morbidly obese patients and smokers. The majority of patients the panel determined were not indicated for surgery were not morbidly obese or smokers, and the clinical scenarios for the patients recommended for PT or injection are not described. The authors concluded that a multidisciplinary panel can reduce fusion rates, though the more accurate conclusion is that the panel can reduce fusion rates at their institution. It seems likely that many of the patients who were deemed inappropriate fusion candidates ultimately had surgery at another institution, though these data are not reported. The conclusion that such a panel might improve the value of spine care is a long logical reach as this study included no patient reported outcomes. To demonstrate that such a panel improves value would require an RCT comparing costs and outcomes between spine patients evaluated by such a panel and those treated in the traditional model where decision-making is between the patient and surgeon. Simply reducing the rate of fusion does not necessarily improve value, as some patients who may have benefited from fusion could continue to receive expensive, ineffective non-operative care. While the spine group at Virginia Mason appears to have more stringent indications for lumbar fusion than surgeons at other institutions, it is not yet clear that use of a “fusion panel” improves the value of spine care.

Please read this paper in the September 1 issue. Does this change your view of how surgical decision making should occur? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor

Friday, August 18, 2017

Rates of lumbar fusion vary significantly across geographic areas, though the underlying causes of this variation remain poorly understood.1 The relationship between surgeon characteristics and fusion rates at the individual surgeon level has not been extensively studied. As such, Michael Schallmo and colleagues from Chicago performed a study to determine the relationship between surgeon demographic characteristics and individual surgeon fusion rates based on the Medicare database. They identified approximately 4,000 spine surgeons who performed at least 11 lumbar fusion procedures on Medicare beneficiaries between 2011 and 2013. Using publicly available information on the internet, they then determined each surgeon’s specialty (orthopedics vs. neurosurgery), practice type (academic vs. private), gender, years in practice, geographic location (Northeast, South, Midwest, or West), degree (MD vs. DO), and medical school location (US vs. foreign). They used the Medicare database to determine the individual surgeon’s fusion rate, defined as the number of lumbar fusions performed divided by the number of unique Medicare beneficiaries seen during the 3 year period. There was a near equal distribution of orthopedic spine surgeons and neurosurgeons (48% vs. 52%), over 90% were in private practice (also including non-academic hospital employed surgeons), 96% were MDs, 91% graduated from a US medical school, and 98% were men. The overall fusion rate was 7.5%. The greatest variation in fusion rate was for academic compared to private practice surgeons, with academic surgeons having a 10% fusion rate compared to 7.2% for private surgeons. Surgeons in the West had a significantly higher fusion rate than those in other regions (8.5% West, 7.8% Midwest, 7.3% Northeast, and 6.9% South). Neurosurgeons had a slightly higher fusion rate (7.6% vs. 7.4%) compared to orthopedic surgeons. Surgeons in practice for less than 27 years had a somewhat higher fusion rate than their older colleagues. Other factors were not significantly associated with fusion rate.

The authors should be congratulated for performing what was likely a very arduous study that evaluated fusion rate at the individual surgeon level. The similarities across demographic groups are probably as impressive as the differences, with the only major difference observed between academic and private practice surgeons. The authors noted that this may be due to academic surgeons being referred more complex cases that have already had extensive non-operative treatment and frequently require more complex, reconstructive procedures. The study did not stratify the analysis based on diagnosis, and it is likely that academic surgeons care for more patients with fractures and tumors, diagnoses that are more likely to require fusion surgery than degenerative conditions. The geographic variation was less striking in this case as the country was divided into 4 large regions for this analysis compared to hospital referral regions, which were used in prior studies of geographic variation.1 At this larger level of analysis, much of the variation among hospitals is lost by averaging across many hospital referral regions. The major limitation of this study is that the outcome, “fusion rate”, is highly dependent on the denominator (i.e. number of Medicare patients seen). Given that different surgeons have very different practices, some may see many non-operative patients or patients with non-spinal conditions. As such, surgeons with identical indications for surgery may have very different “fusion rates”. This study is helpful as it shows that demographic characteristics do not appear to drive most of the observed variation in fusion rates. Studying exactly what drives the variation in the decision to perform a lumbar fusion is difficult, and it likely depends on surgeon and patient factors that cannot be measured from databases. The “right” rate of lumbar fusion remains unknown and will likely remain difficult to define until there is evidence-based consensus on the appropriate indications for fusion.

Please read this article in the August 15 issue. Does this change how you think about the relationship between surgeon factors and the decision to perform lumbar fusion? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor



1.            Weinstein JN, Bronner KK, Morgan TS, Wennberg JE. Trends and geographic variations in major surgery for degenerative diseases of the hip, knee, and spine. Health Aff (Millwood) 2004;Suppl Web Exclusives:VAR81-9.