Skip Navigation LinksHome > Blogs > The Spine Blog
The Spine Blog
Saturday, December 27, 2014

Over the past 25 years, certain indications for lumbar fusion such as spondylolisthesis and scoliosis have become reasonably well-established in the literature. Simultaneously, the number of fusion techniques available to spine surgeons has exploded and now includes uninstrumented posterolateral fusion, instrumented posterolateral fusion, ALIF, PLIF, TLIF, XLIF, and AP fusion. Likewise, bone graft options have expanded from traditional iliac crest bone graft (ICBG) to also include local bone graft, allograft, demineralized bone matrix, inorganic matrices, bone marrow aspirate, and BMP-2. The most recent addition to the spine surgeon’s armamentarium is minimally invasive fusion. The various combinations of approach, instrumentation, and bone graft material yield dozens of fusion “techniques” that could be considered. Given this rapid expansion of techniques, it is not surprising that studies comparing fusion techniques comprised a substantial amount of the spine surgery literature this year. In February, an RCT comparing the use of silicate-substituted calcium phosphate (Si-CaP) to BMP-2 for MIS-TLIF demonstrated a 35% pseudarthrosis rate for Si-CaP to 8% for BMP-2.1 Respondents to our Quick Poll demonstrated a clear preference for local bone graft (86%) for MIS-TLIF, with 0% reporting that they used BMP-2. In September, a study looking at fusion technique trends for degenerative spondylolisthesis in the American Board of Orthopaedic Surgery database showed that the rate of uninstrumented fusion dropped from 36% in 2003 to 8% in 2011, while the rate of posterior interbody fusion (including PLIF and TLIF) increased from 14% to 37% over the same time period.2 The Quick Poll results matched the study results, with 60% of respondents reporting they favor TLIF for degenerative spondylolisthesis, 20% XLIF, 15% posterolateral instrumented fusion, and 5% posterolateral uninstrumented fusion. Another September study using the National Surgical Quality Improvement Program (NSQIP) database reported that ICBG harvest moderately increased hospital stay, operating room time, and transfusion rate compared to fusions using alternative bone graft sources.3 Concerns over the use of ICBG were clearly on the minds of Quick Poll respondents, with only 10% reporting they used ICBG for fusions of three or more levels, with 50% reporting the use of local bone graft + extenders and 40% local bone graft + BMP-2. An October study analyzed the Kaiser Permanente database that included over 9,000 fusion cases and found that the use of BMP-2 did not reduce the two year re-operation rate for pseudarthrosis.4 This finding supports the thinking of the majority of Quick Poll respondents, 2/3 of whom answered that BMP-2 does not reduce the re-operation rate for single level posterolateral instrumented fusion. A recent RCT published in November reported no clinical or radiographic benefit associated with the addition of posterior interbody fusion to posterolateral instrumented fusion for isthmic spondylolisthesis.5 Despite their apparent enthusiasm for interbody fusion, 50% of Quick Poll respondents answered that the addition of interbody fusion did not improve outcomes for isthmic spondylolisthesis.


Does the 2014 literature focused on fusion techniques help to guide actual surgical decision-making? Given the apparent contradictions between clinical practice and the literature, the answer seems to be a resounding “No”.  Despite the lack of evidence supporting even the use of pedicle screw instrumentation, let alone interbody fusion, only 8% of fusions for degenerative spondylolisthesis in the 2011 ABOS database were uninstrumented. Part of the disconnect between the literature and clinical practice may be due to the low level of evidence on the topic. Many of the recent studies have been database analyses, which are always retrospective and usually at high risk for selection bias and confounding that are difficult to control for statistically. While many spine topics are challenging to study with a Level 1 RCT, studies comparing fusion techniques are well-suited to this study design given that randomization can occur in the OR, making crossover impossible. Blinding is somewhat more difficult, but can be accomplished. Hopefully 2015 will bring more Level 1 studies that help guide us in choosing the fusion technique that leads to the best outcome for each individual patient while also being safe and cost-effective.

Happy New Year from the Spine Blog!


Adam Pearson, MD, MS
Associate Web Editor



1.            Nandyala SV, Marquez-Lara A, Fineberg SJ, Pelton M, Singh K. Prospective, randomized, controlled trial of silicate-substituted calcium phosphate versus rhBMP-2 in a minimally invasive transforaminal lumbar interbody fusion. Spine (Phila Pa 1976) 2014;39:185-91.

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.            Gruskay JA, Basques BA, Bohl DD, Webb ML, Grauer JN. Short-term adverse events, length of stay, and readmission after iliac crest bone graft for spinal fusion. Spine (Phila Pa 1976) 2014;39:1718-24.

4.            Guppy KH, Paxton EW, Harris J, Alvarez J, Bernbeck J. Does bone morphogenetic protein change the operative nonunion rates in spine fusions? Spine (Phila Pa 1976) 2014;39:1831-9.

5.            Lee GW, Lee SM, Ahn MW, Kim HJ, Yeom JS. Comparison of posterolateral lumbar fusion and posterior lumbar interbody fusion for patients younger than 60 years with isthmic spondylolisthesis. Spine (Phila Pa 1976) 2014;39:E1475-80.



Friday, December 19, 2014

Anterior cervical discectomy and fusion is considered the gold-standard operation for cervical radiculopathy, and most surgeons consider it to have a high rate of success and a low rate of complications and need for revision surgery. The only large, consecutive series of ACDF with long-term follow-up is that from Henry Bohlman’s group from the 1990s, a time when the use of autograft and no plate was the standard. Given the lack of a modern case series, Dr. van Eck and her colleagues from Pittsburgh elected to review their series of ACDFs from 2000-2010 to determine revision rates for pseudarthrosis and adjacent segment disease and compare these with historical data. They included 672 patients who had an average follow-up of 31 months. Over 50% were smokers, and the majority underwent two-level procedures. All patients received an anterior cervical plate, and 95% were treated with tricortical iliac crest autograft. They reported an overall reoperation rate of 15%, with 7% undergoing revision for pseudarthrosis and 7% for ASD. Somewhat surprisingly, there were no significant baseline differences between the solid union group and the pseudarthrosis revision surgery group. Age, smoking status, and number of levels fused did not predict pseudarthrosis. Similarly, there were no baseline differences in age, number of levels fused, plate to adjacent disk distance, and smoking status between the patients undergoing revision for ASD and those who did not. The distance between the plate and the cranial adjacent disk space was 0.5 mm less in the ASD revision group compared to the non-revision group, but this difference was neither  statistically nor clinically significant.


The overall 15% revision rate for ACDF probably strikes most surgeons as relatively high, but it is similar to the historical rate of 14% published by Hilibrand et al. Most surgeons would also probably be surprised that the addition of an anterior plate affected neither the reoperation rate for pseudarthrosis nor the revision rate for ASD. Even more striking is the lack of associations between previously reported risk factors for pseudarthrosis (i.e. smoking, multiple level surgery) and ASD (i.e. plate to disk distance, younger age, few levels fused) and re-operation. There are a few potential explanations for this, including that re-operation is a very subjective outcome, and the decision to pursue revision surgery depends on the wishes of the patient and surgeon as much as on radiographic appearance or clinical findings. Additionally, the actual number of patients with complications was relatively low, so subgroup analyses were likely underpowered. While traditional risk factors such as smoking and multilevel fusion did not appear to be more common in the pseudarthrosis revision group, we should not change how we view these risk factors given the limitations of the current study. These data on revision rates can be used to counsel patients, establish benchmarks, and compare results with other treatment approaches such as disk replacement. Please read Dr. vanEck’s paper on this topic in the December 19 issue. Does this article change how you view revision surgery after ACDF? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS
Associate Web Editor

Friday, December 12, 2014

Global sagittal balance has been studied extensively and plays a major role in surgical decision making when correcting thoracolumbar deformity. Cervical sagittal balance has only recently been the focus of investigation, with some data indicating that cervical sagittal imbalance leads to worse outcomes with posterior surgery. In an effort to better understand the effect of sagittal balance on outcomes in cervical spondylotic myelopathy (CSM), Dr. Roguski and her colleagues performed a prospective observational study in 49 CSM patients undergoing anterior or posterior decompression and fusion for multilevel stenosis. All cases were reviewed by a panel of spine surgeons who agreed that they were candidates for either anterior or posterior surgery. Approximately 60% of patients underwent a posterior laminectomy and fusion, with the remaining 40% undergoing anterior procedures. The posterior group included more men, had a C2-7 sagittal vertical axis (CSVA) about 1 cm greater, and 6 degrees more lordosis compared to the anterior group. Baseline patient reported outcomes were similar between the anterior and posterior groups. Overall, outcomes tended to be better in the anterior group, though most of the differences failed to reach significance, likely due to the small size of the groups. The post-operative CSVA was found to correlate with post-operative SF-36 physical component scores (PCS) and mJOA scores as well as changes from baseline in the PCS and Neck Disability Index (NDI). Subgroup analysis showed that these correlations were only found in the posterior group and were not present in the anterior group. Baseline CSVA was found to be the strongest predictor of post-operative CSVA, and surgical approach was not strongly correlated with post-operative CSVA after controlling for baseline CSVA.


Cervical sagittal balance is a relatively new concept, and this is one of the first papers to look at its role in CSM outcomes. Given the observational nature of this study, the association between surgical approach and outcomes is not clear, and the anterior and posterior groups were significantly different at baseline. Parsing out the independent effects of surgical approach and sagittal balance is difficult in that sagittal balance seemed to play more of a role for patients undergoing posterior surgery, but the CSVA was significantly greater both pre- and post-operatively in the posterior group. All of the correlation coefficients describing the relationship between CSVA and outcomes were less than 0.5, indicating relatively weak correlations. Most of the correlations were not statistically significant, and it is unclear if CSVA was not related to most outcomes or if the study was simply underpowered to detect the weak correlations. It would have been helpful if the authors provided baseline characteristics for the CSVA > 40 mm and CSVA < 40 mm groups rather than simply the correlations between CSVA and baseline scores. It seems likely that the relationship between CSVA and symptoms is not linear, and there is likely a threshold above which sagittal imbalance is more likely to be symptomatic. This study certainly raises awareness about cervical sagittal balance, but given that many other factors such as age, comorbidities, severity of cord compression, duration and severity of symptoms, number of levels involved, degree of lordosis/kyphosis, and surgical approach also affect outcomes, this study does not allow for strong conclusions to be made about the role of cervical sagittal balance in CSM. Hopefully future studies with more patients and multivariate analysis will allow for better clarification of its role in determining outcomes.


Please read Dr. Roguski’s article on this topic in the December 1 issue. Does this change how you view the role of cervical sagittal balance? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS
Associate Web Editor

Friday, December 05, 2014

Most spine surgeons who perform anterior cervical fusions have had the experience of obtaining an intra-operative or post-operative AP x-ray and found that their cervical plate is malaligned. With a rigid plate, most probably do not worry as long as the screws have good purchase and are not so lateral as to put the vertebral artery at risk. With the advent of dynamic plates, there is concern that malalignment in the coronal plane could compromise the ability of the plate to allow for graft subsidence and result in less load sharing and possibly increased rates of pseudarthrosis and hardware failure. In order to answer this question, Dr. Lawrence and his colleagues from Utah performed a biomechanical analysis in which a dynamic anterior cervical plate was applied to a cadaveric model with a C5 corpectomy with the plate aligned appropriately or in 20 degrees of malalignment in the coronal plane. A spacer containing a load cell was placed in the corpectomy defect, and range of motion testing was performed in all three planes. Axial load testing was also performed to assess load sharing between the plate and the spacer. After the initial round of testing, the load cell was shortened by 10% to represent subsidence, and the tests were repeated. The tests were initially performed with the plate appropriately aligned followed by the malaligned condition. The authors found no significant differences in the changes in range of motion or stiffness following shortening of the spacer between the correctly aligned and malaligned plating groups. Additionally, there were no differences in load sharing between the groups in either the full-length or shortened spacer conditions. Based on these results, the authors recommended against changing the alignment of the plate if it is malaligned up to 20 degrees.


Surgeons who read this paper will rest easier knowing that moderate degrees of plate alignment will not compromise its biomechanical performance. While greater degrees of malalignment would likely compromise the plate’s ability to allow subsidence—a plate rotated 90 degrees would allow no subsidence—the authors noted that it would be difficult to put a plate on with much greater than 20 degrees of malalignment. Like all biomechanical studies, this one has some limitations. The malaligned plate testing was always performed after the appropriately-aligned tests, so the bone-screw interface may have been less robust in the malaligned condition. However, the authors looked at changes in range of motion and stiffness after shortening the spacer rather than absolute values, which likely mitigated this effect. Additionally, it is always challenging to extrapolate from the biomechanics lab to the clinical arena. Given that the benefit of dynamic plating is relatively modest, it seems unlikely that moderate degrees of malalignment would lead to clinically detectable differences.1 When it comes to deciding whether to accept a malaligned plate detected by intra-operative AP radiograph or fluoroscopy, it seems as though the risk of losing fixation by changing the alignment far outweighs any potential biomechanical benefit of better alignment. It may simply be better to get the AP x-ray in the recovery room to eliminate any temptation to get the plate perfectly aligned.


Please read Dr. Lawrence’s article on this topic in the December 1 issue. Does this change how you view cervical plate malalignment? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor



1.            Campos RR, Botelho RV. Systematic review of the effect of dynamic fixation systems compared with rigid fixation in the anterior cervical spine. Eur Spine J 2014;23:298-304.



Friday, November 28, 2014

The benefit of epidural steroid injections (LESI) for various indications in the lumbar spine has been debated, with most evidence suggesting they provide a short-term benefit for lumbar radiculopathy, while there is probably minimal benefit for axial low back pain or neurogenic claudication.1 There is less data on the cost-effectiveness of LESI. Given this lack of economic data, Dr. Spijker-Huiges and colleagues from the Netherlands performed a “pragmatic RCT” in which they compared patient-reported and economic outcomes for 63 patients with two to four weeks of lumbar radiculopathy who were randomized to LESI or “standard” care (which appears to have included pain medication and admonitions to continue regular activities). Diagnosis was based on the general practitioner’s impression and did not require any imaging. The two treatment groups had significant differences at baseline, with the LESI group having more severe symptoms on multiple patient-reported measures. After controlling for these differences, the authors reported that the LESI group had significantly better outcomes on the numerical rating scale for low back pain--the primary outcome measure--as well as other secondary outcome measures out to one year. However, the patient-reported outcome measures were not explicitly reported in this paper, and it seems that a separate paper from the trial on the patient reported outcomes remains in the publication process. Cost analysis indicated that LESI was associated with about $1000 less in costs, primarily due to approximately $2000 lower costs resulting from less missed work in the LESI group.


This is an interesting study that suggests the use of LESI in acute radiculopathy may actually save money due to less missed work. While this is an intriguing idea, certain study limitations make it difficult to strongly support that conclusion. The patient population is somewhat nebulous as no imaging study was required for inclusion. While the diagnosis of lumbar radiculopathy is oftentimes clear based on history and physical exam, there are multiple other conditions in the differential diagnosis, and advanced imaging is typically necessary to confirm the diagnosis and location of the lesion. In the United States, most practitioners performing injections require cross-sectional imaging prior to LESI to confirm the diagnosis and rule out rare, more serious spinal pathology that can present with radicular symptoms (i.e. tumors, infection, etc). Additionally, the accounting approach to costs due to missed work was not explained in this paper. This is not always a straightforward process, and the costs can vary markedly depending on the methods used as well across nations. The cost of an injection ($259) is very low in this study, which certainly drives down the costs for the LESI group. In the United States, Medicare reimbursement for LESI was estimated to be $637, which represents the low end of payment for this procedure.2 This study did not use traditional cost-effectiveness analysis, which yields an incremental cost effectiveness ratio (ICER) in dollars per quality adjusted life year. Given that the intervention was both more effective and less expensive, the units for the ICER are probably not that important. While it has certain methodological limitations, this paper certainly raises the question of whether early LESI for lumbar radiculopathy in patients who are out of work might actually be a money-saving intervention, at least from the societal perspective.

Please read Dr. Spijker-Huiges’s article on this topic in the November 15 issue. Does this paper change how you view the cost-effectiveness of LESI? Let us know by leaving a comment on The Spine Blog.


Adam Pearson, MD, MS

Associate Web Editor


1.            Chou R, Atlas SJ, Stanos SP, Rosenquist RW. Nonsurgical interventional therapies for low back pain: a review of the evidence for an American Pain Society clinical practice guideline. Spine (Phila Pa 1976) 2009;34:1078-93.

2.            Bresnahan BW, Rundell SD, Dagadakis MC, et al. A systematic review to assess comparative effectiveness studies in epidural steroid injections for lumbar spinal stenosis and to estimate reimbursement amounts. PM R 2013;5:705-14.


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.