Skip Navigation LinksHome > Blogs > The Spine Blog
The Spine Blog
Saturday, August 29, 2015

The use of lumbar epidural steroid injections (LESI) for spinal stenosis (SpS) is controversial, with a recent RCT and meta-analysis concluding that LESI provided minimal or no benefit in SpS.1,2 A post-hoc subgroup analysis of the Spine Patient Outcomes Research Trial (SPORT) demonstrated worse surgical and non-operative outcomes for patients who received an LESI within 3 months of enrollment. Given the widespread use of LESI in SpS without much evidence supporting its efficacy and concerns that it could have a negative effect on outcomes, Dr. Fekete and colleagues in Switzerland analyzed data from the Lumbar Stenosis Outcome Study (LSOS) to compare surgical and non-operative outcomes between patients who had received at least one LESI within the year prior to enrollment and those who had not. Patients receiving an LESI after enrollment were excluded. The study was an observational cohort study in which patients and surgeons determined treatment. The authors found no significant differences in post-operative outcomes between the LESI and no-LESI surgical patients on the Spinal Stenosis Measure (SSM). Among the non-operative patients, the LESI patients had somewhat worse outcomes on some of the SSM subscales.


This well-done observational study suggests that LESI does not lead to worse surgical outcomes for SpS. These findings are contrary to the SPORT subgroup analysis, though the patient populations were substantially different. In the SPORT study, patients who had received LESI prior to enrollment were excluded. That study compared results between patients who received LESI within 3 months of enrollment and those who did not. This is in contrast to the current study in which the LESI patients were defined as those receiving an injection in the year prior to enrollment, while those receiving an LESI after enrollment were excluded. The outcome measures were also different, with SPORT using the SF-36 and Oswestry Disability Index while the LSOS used the SSM as the primary outcome measure. In SPORT, patients who received an LESI and subsequently underwent surgery had a baseline preference for non-operative treatment, which ultimately failed. As such, these may have represented patients with unmeasured characteristics associated with treatment failure—whether surgical or non-operative. In the LSOS, patients had received the LESI prior to enrollment. Given that it was an observational trial, the LSOS patients who underwent surgery may have had a stronger baseline preference for surgery than the LESI patients in SPORT who only underwent surgery after failing non-operative treatment. The SPORT study also reported longer OR times, higher blood loss, a higher rate of dural tear and longer length of stay in the LESI patients, which raised the possibility that LESI resulted in adhesions or other changes in the epidural space that made surgery more difficult. Since the injections were likely more remote in the LSOS (all injections were prior to enrollment), if LESI had such an effect in the epidural space, it may have diminished over time. Given the differences in study design and patient populations, it is not surprising that SPORT and LSOS reached different conclusions on this topic. While it is interesting to consider the effect of LESI on subsequent surgical outcomes, the more pressing question is whether or not LESI is indicated in SpS given that it has been shown to provide minimal benefit.


Please read Dr. Fekete’s article on this topic in the August 15 issue. Does this change your view on how LESI affects outcomes in SpS? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor




1.            Chou R, Hashimoto R, Friedly J, et al. Epidural Corticosteroid Injections for Radiculopathy and Spinal Stenosis: A Systematic Review and Meta-analysis. Ann Intern Med 2015.

2.            Friedly JL, Comstock BA, Turner JA, et al. A randomized trial of epidural glucocorticoid injections for spinal stenosis. N Engl J Med 2014;371:11-21.


Friday, August 21, 2015

Upright lateral flexion-extension (FE) x-rays are the gold standard to detect spinal “instability”. While the definition and implications of “instability” are highly controversial, most spine specialists agree that FE x-rays are necessary to detect it. It is well known that a listhesis can be missed on a supine MRI due to reduction of the slip in that position and that an upright lateral radiograph is a significantly more sensitive study.1 However, the amount of translation observed between the supine MRI and an upright radiograph (US, i.e. upright-supine) has not been compared to the amount of translation measured on FE x-rays. As such, Dr. Liu and his colleagues from Boston compared translation on US to FE to determine if FE x-rays add any additional information beyond what can be obtained from a supine MRI and a single upright lateral radiograph. They analyzed radiographs and MRIs of 68 patients with degenerative and isthmic spondylolisthesis and calculated translation as a percentage of the anterior-posterior dimension of the superior vertebra in the listhetic motion segment. They found the US demonstrated significantly more motion (i.e. 7.6% vs. 4.9%) than FE radiographs, with 71% of patients having greater motion on US with the remaining 29% having more motion on FE. Using 8% translation as the threshold of “instability”, they found “instability” in 43% of patient on US and 18% on FE x-rays.


In the current environment focusing on cost containment as well as limiting radiation exposure, this study questions the utility of FE radiographs in an era where nearly all spondylolisthesis patients being evaluated by a spine specialist have an MRI and upright radiograph. While most spine surgeons agree that an upright radiograph adds information to an MRI, this study implies that there may not be a need to get FE x-rays in addition to a standing lateral film for spondylolisthesis patients. This paper does not answer the question of how many patients have spondylolisthesis on a flexion film that is not present on a standing lateral film. At the same time, it is also unclear if such a patient should be considered either “unstable” or carry the diagnosis of spondylolisthesis. Even less clear is whether or not a subtle slip on a flexion film alone should be the reason for a fusion. This study does a nice job demonstrating that a listhesis tends to reduce more with laying supine than with extension, and that FE x-rays are not necessary to determine the amount of translation at a listhetic level. The difficulty with all measures of “instability” is that they correlate poorly with clinical findings, and it is unclear if the presence of “instability” should alter surgical treatment decisions. This paper reports that “unstable”, or possibly better termed hypermobile, patients tend to have less severe back pain. Another paper showed that hypermobile degenerative spondylolisthesis patients improved more with non-operative treatment than their “stable” counterparts.2  Some surgeons suggest that “unstable” spondylolisthesis patients need more rigid fusion constructs (i.e. pedicle screw instrumentation or interbody support), though there is no high quality evidence supporting this viewpoint. While debates about “instability” will likely persist, this paper adds to the literature suggesting that flexion-extension radiographs might not add much to an MRI and a single upright lateral x-ray.

Please read Dr. Liu’s article on this topic in the August 15 issue. Does this paper change your view of flexion-extension radiographs? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor





1.            Lattig F, Fekete TF, Grob D, Kleinstuck FS, Jeszenszky D, Mannion AF. Lumbar facet joint effusion in MRI: a sign of instability in degenerative spondylolisthesis? European spine journal : official publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society 2012;21:276-81.

2.            Pearson AM, Lurie JD, Blood EA, et al. Spine patient outcomes research trial: radiographic predictors of clinical outcomes after operative or nonoperative treatment of degenerative spondylolisthesis. Spine (Phila Pa 1976) 2008;33:2759-66.


Friday, August 14, 2015

Surgical site infection (SSI) following instrumented spinal fusion is a serious complication that negatively affects outcomes and increases healthcare costs. Infection in the presence of hardware is difficult to eradicate and frequently requires multiple trips to the operating room, placement of a wound vac, soft-tissue coverage by plastic surgery, and intravenous and oral antibiotics. The most significant recent advance in SSI control is the use of intra-wound vancomycin powder, which has been shown to significantly reduce the rate of SSI in instrumented posterior spinal fusion. This represents a low risk, low cost intervention to reduce infection rate, and other low cost, low risk measures are needed. The authors of the current study wanted to investigate if covering pedicle screws with a sterile towel once they have been opened would reduce the rate of screw contamination. The authors randomized 42 patients to either the covered screw group or the control group in which the screws were not covered intra-operatively. A screw from the set was placed in culture medium when the screws were opened and every thirty minutes thereafter. The authors found that the rate of contaminated screws was significantly higher for all time points in the uncovered control group. In fact, after 120 minutes , 55% of screws in the control group were contaminated compared to 18% in the covered group , a three-fold increase. While SSI rates were not reported, the authors did note that the infection rate was not reduced in the covered group.


This paper demonstrates how a simple, low cost intervention (i.e. covering implants with a sterile towel) can reduce contamination of surgical instrumentation. The results are in-line with prior studies and are consistent with what one would predict. This small study has significant limitations that need to be considered, including a low number of patients, no reporting of the SSI rates, and no discussion of patient or surgical characteristics. These data alone do not suggest that covering implants actually reduces infection rates, but it is a simple, low cost, no risk intervention that cannot hurt and might help. Even the most skeptical reader would have a difficult time arguing against adopting the practice. Demonstrating its effectiveness in a large RCT would be difficult as infection is a relatively rare event and covering implants likely has only a modest effect compared to other factors. This seems like one of those topics for which Level 1 evidence may not be necessary to adopt a new clinical practice.

Please read Dr. Menekse’s   article on this topic in the August 15 issue. Do you think you will start to cover implant trays based on these data? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS
Associate Web Editor

Friday, August 07, 2015

Adult spinal deformity (ASD) surgery has high complication rates, and many factors such as patient age, medical comorbidities, obesity, smoking status, and complexity of surgery are associated with complications. Surgeon experience and skill, factors which one would assume are associated with outcomes and complications, have not been evaluated as possible predictors of complication rates. As such, Dr. Skovrlj and colleagues from the Scoliosis Research Society (SRS) analyzed the SRS database that included over 5,000 ASD cases to determine if complication rates were associated with surgeon experience. While the database does not include years of surgical experience, members are classified as either candidate or active members, with the assumption that candidate members have less experience than active members. The authors reported an overall complication rate of 13.3%, with no significant difference between the two surgeon cohorts. Compared to other ASD literature, this is a relatively low complication rate, possibly related to it being self-reported data. In looking at specific complications, the authors noted that the candidate members reported spinal cord complications (1.1% vs. 0.55%) and superficial surgical site infections (1.8% vs. 0.9%) at twice the rate of active members, differences that were statistically significant. Conversely, the active members reported significantly more pulmonary complications (1.2% vs. 0.45%). Rates of other complications were not significantly different.


This is an interesting study in that it evaluated surgeon experience as a predictor of complication rate, a common-sense association but one that has not been formally evaluated in ASD surgery. While this study suggests experience might affect the likelihood of specific complications, in general, there was not a strong relationship between experience and complications. However, the limitations of the study, which are substantial, need to be considered before drawing any conclusions. Given that the database relies on self-reporting, many complications are probably not captured. Additionally, it is also possible that candidate members were more likely to report complications than active members. This is a retrospective observational study using a prospectively recorded database, so it is at risk of confounding by other factors as well. If the candidate members had sicker patients undergoing more complex surgery, one would expect a higher complication rate. Unfortunately, these data were not recorded, so it is difficult to determine how much of the increased complication rate was due to inexperience versus other factors. Outcomes such as OR time and blood loss may have been more strongly associated with experience, yet these were not recorded. While somewhat preliminary, these data are not surprising and do suggest that inexperience may be associated with some complications in complex spine surgery. However, all experienced surgeons were inexperienced once, so it is difficult to know how to act on data such as these. As the saying goes, “Good judgment comes from experience, and experience comes from bad judgment.” Hopefully inexperienced surgeons tackling challenging ASD cases have mentors that can guide their decision-making to some degree. Papers like this also should serve to remind inexperienced surgeons to have heightened awareness for potential complications and to seek out assistance for cases that challenge their skill set.

Please read Dr. Skovrlj’s article on this topic in the August 1 issue. Does this change how you view how surgeon experience can affect outcomes? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor

Friday, July 31, 2015

Disk degeneration (DD) and Modic changes (MC) have been shown to be associated with low back pain (LBP) in many, but not all, population-based studies. These disparate findings could have been related to different ways of measuring low back pain and disability and different patient populations under study. In order to clarify the association between DD, MC, and LBP, Dr. Maatta and colleagues used data from the TwinsUK database, that included T2-weighted MRI images on 823 patients at baseline, with 429 having a follow-up MRI at an average of 10 years after the initial scan. The vast majority of the population was women (96%), and the average age at baseline was 54. Unlike most prior studies, they defined “disabling low back pain” as one or more episodes of LBP lasting for at least 1 month, which limited basic activities such as walking around the house, standing for more than 15 minutes, getting in and out of a car, or going up and down stairs. Disk degeneration was categorized on a 0-20 point scale based on disk height loss, signal intensity change, posterior disk bulge, and anterior osteophytes. In univariate analysis, they found that age, BMI, DD, and MC were all associated with having an episode of LBP, with MC patients having more than double the risk at baseline compared to the non-MC group (35% vs. 16%). Multivariate analysis demonstrated that age, DD, and MC remained independent predictors of disabling LBP, with the presence of MC increasing the odds of LBP by 57%. Not surprisingly, there were strong correlations between DD and MC, with disk height loss and disk signal intensity change being associated with baseline MC, while disk height loss and posterior disk bulge predicted the development of MC.


The authors should be congratulated on their ISSLS award winning paper  demonstrating that MC is a strong, independent risk factor for having experienced disabling LBP. Compared to prior studies that usually defined LBP based on pain scales, this study was unique in looking at episodes of disabling LBP. Given that LBP can wax and wane, a patient’s pain rating on a given day may not reflect their long-term back condition, so a strict definition of having suffered an episode of disabling LBP is probably more meaningful. Interpreting the statistics is somewhat difficult as DD and MC are reported to be independent predictors of disabling LBP, yet their presence is strongly correlated. This implies they are part of the same degenerative process, but the presence of MC increases the risk of disabling LBP beyond the presence of DD alone. The two major limitations of this paper are the inability to characterize the type of MC (no T1 images were available for analysis) and the homogenous patient population (middle-aged women in the UK). This makes it difficult to know how different types of MC are associated with symptoms, though Type I MC are the most common and the type most associated with pain in prior studies. The strength of association implies that the results are probably generalizable, but it is important to repeat this study in other groups to confirm this. What this paper does not tell us is how the presence or absence of MC should affect how we treat the LBP population. Does the presence of MC predict better or worse outcomes with medication, PT, surgery or even antibiotics? Now that multiple studies have confirmed that MC is associated with pain and disability, future studies should determine if their presence or absence can help guide treatment for this challenging and common clinical problem.

Please read Dr. Maatta’s article in the August 1 issue. Does this change how you interpret Modic changes? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor    

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.