Home Current Issue Previous Issues Published Ahead-of-Print Collections In The News Blog For Authors Journal Info
Skip Navigation LinksHome > Blogs > The Spine Blog
The Spine Blog
Sunday, April 20, 2014

Studies of lumbar radiculopathy have typically focused on pain and function as outcomes, with very little reported on the time course and degree of improvement of numbness. While pain tends to be the most bothersome radicular symptom for most patients, numbness and paresthesia can also be troublesome. Based on anecdotal experience, surgeons have traditionally told patients that numbness improves slowly and unpredictably, though there has not been much data supporting this belief. In an effort to fill the knowledge gap surrounding numbness after surgery for lumbar radiculopathy, Drs. Huang and Sengupta reviewed the pain diagrams of 85 surgical patients with lumbar radiculopathy due to disk herniation and/or spinal stenosis. In addition to pain, patients were also asked to shade areas on the pain diagram corresponding to numbness and paresthesia. All patients had baseline pain diagrams as well as at least two more post-operative pain diagrams over one year of follow-up. Additionally, patients were stratified based on whether they reported radicular symptoms for more or less than 6 months. The authors reported that pain improved fastest and to the greatest degree, with marked improvement in the first 6 weeks, beyond which there was minimal improvement. Numbness and paresthesia never improved significantly from baseline, though there were trends towards improvement over the course of an entire year. The Oswestry Disability Index and SF-36 physical function scores improved significantly out to three months and then plateaued. Patients with a longer duration of symptoms tended to improve slower than patients with symptoms for less than 6 months. These findings effectively confirmed what surgeons knew from anecdotal experience—pain improved the most rapidly and completely while numbness and paresthesia tended to improve slowly and incompletely.


The authors have made an excellent first step in quantifying difficult to measure outcomes such as numbness and paresthesia. No prior study has done such a complete analysis of these symptoms, and the results of this study will be valuable to surgeons and other spine providers counseling patients about their likely outcomes following surgery. The current method of quantifying the number of pixels on a computer-based pain diagram is novel, and future studies should evaluate exactly what this is measuring. Further stratification based on the underlying diagnosis (i.e. disk herniation vs. stenosis) would also be interesting as the current subgroup analysis focused on duration of symptoms. Given that all of the short term symptom patients had disk herniation while the long term patients were a mixture of disk herniation and stenosis, some of the differences observed could be driven by differences in pathology rather than in duration of symptoms. Weakness is another bothersome symptom that has not been well-studied, and future rigorous studies evaluating the time course and degree of improvement of weakness in lumbar radiculopathy would also be important. Data on the natural history of numbness in patients treated non-operatively would also be interesting. The current paper offers an important evidence base for discussions about how pain and numbness will likely change following surgery for lumbar radiculopathy, and such discussions should help create realistic patient expectations.

Please read Drs. Huang’s and Sengupta’s article on this topic in the April 15 issue. Does this affect how you counsel your lumbar radiculopathy patients pre-operatively? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor

Thursday, April 10, 2014

Level 2 studies have shown that subfascial vancomycin powder is effective in reducing surgical site infections in posterior fusions in the cervical, thoracic, and lumbar spine, with the most pronounced benefit observed in high risk trauma patients. While vancomycin powder likely reduces the rate of surgical site infection, it does not eradicate the problem, and there is concern that its use could alter the microbiology observed when infection does occur. In an attempt to determine the effect of subfascial vancomycin powder on the microbiological spectrum of surgical site infection, Dr. Prasad and his colleagues from Philadelphia reviewed 981 consecutive spine surgery patients treated with vancomycin powder to evaluate the microbiology of surgical site infection in this population. They defined deep infection as the presence of clinical signs and symptoms consistent with surgical site infection leading to a return to the OR for irrigation and debridement, regardless of culture results. Based on this definition, 6.7% of patients had a surgical site infection, and 5.2% had positive wound cultures. Consistent with prior studies, 86% of patients with positive cultures were infected with gram positive organisms, the majority of which were Staph aureus. However, 60% of patients also had gram negative organisms grow from intra-operative cultures, compared to a historical control of 21% at the same institution. Based on these findings, the authors concluded that vancomycin powder could increase the rate of gram negative or polymicrobial infection.


This paper is the first large study to look in-depth at the microbiology of surgical site infection following the application of subfascial vancomycin powder. The data do indicate a markedly increased rate of gram negative infection compared to historical controls. However, the historical controls also had a lower overall rate of infection. This indicates that the historical controls were likely a very different population compared to patients in the current study, who received vancomycin based on surgeon choice. Approximately 90% of patients in the current study underwent fusion, while at least 20% of the patients in the historical controls underwent decompression alone. Given the likely difference in the two populations, it is hard to know if the use of vancomycin powder was driving the higher rate of gram negative infection or if patients receiving vancomycin were at higher risk of gram negative and polymicrobial infection. It is interesting that the proportion of patients with gram positive infection remained about the same despite the use of vancomycin. This paper makes an important contribution and should alert clinicians to the possibility of gram negative infection following the use of vancomycin powder. In patients who have received vancomycin powder and present with a surgical site infection, empirical broad spectrum antibiotics should be considered until culture data are back.


Please read Dr. Prasad’s article on this topic in the April 1 issue. Does this change your view on the use of subfascial vancomycin powder? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor

Friday, April 04, 2014

The use of opioids in the treatment of chronic low back pain (CLBP) is controversial due to a lack of evidence regarding the efficacy and adverse effects associated with their use for this indication. Initially, opioids tended to be used only for patients with cancer pain or those with acute pain related to surgery or trauma, but the pendulum swung towards their use for chronic non-cancer pain in the 1990s. More recently, concern over long-term adverse effects such as tolerance, addiction, and overdose, not to mention abuse and diversion, has caused the pendulum to swing back against their use for chronic non-cancer pain. In an effort to add an evidence base to the discussion, Dr. Chaparro and an accomplished group of co-authors updated the Cochrane Review on the topic of opioids for CLBP, which was published in the April 1 issue. They included blinded RCTs comparing opioids to placebo or other medication in an outpatient setting for patients reporting CLBP (defined as greater than 12 weeks) with a follow-up period of at least 1 month. They included fifteen trials with a total of 5,540 patients that studied “strong opioids” (i.e. morphine, oxycodone, oxymorphone, hydromorphone, tapentadol), tramadol, and transdermal buprenorphine. Their meta-analyses indicated that there was very low to moderate quality evidence that strong opioids, tramadol, and transdermal buprenorphine led to more improvement in pain and function than placebo, with the best quality evidence in support of strong opioids and tramadol. There were few studies comparing opioids to an active control, and these did not show any significant benefit for opioids compared to celecoxib or antidepressants. Not surprisingly, opioids had a worse side effect profile than placebo and were associated with nausea, somnolence, dizziness, and dry mouth. No serious adverse events were reported.


The main conclusion of this review, that opioids were more effective than placebo for short-term treatment of CLBP, is not surprising. Opioids have strong analgesic properties in the short-term, and patient blinding may not have been very effective given the analgesic benefit from opioids as well as the significant side effect profile. The more pressing question is not if opioids are more effective than placebo in treating CLBP in the short term but whether or not opioids are more effective than alternative treatments in the long term. Given that long term opioid use is associated both with tolerance and decreased efficacy as well as potential adverse effects, it seems unlikely that the short term benefit and absence of serious adverse effects reported here would persist in the long term. The real question is whether or not opioids provide better long term outcomes than other potentially less risky alternatives such as anti-inflammatory medication and physical therapy. Now that there are serious concerns about the negative long term effects of opioid therapy, it is unclear whether a study comparing the long term outcomes between opioids and alternative therapies is ethical. Rather than performing such a study, it may simply be time for the spine community to abandon the use of strong opioid medication for CLBP and focus on alternative strategies that can hopefully be more effective and less dangerous.

Please read Dr. Chaparro’s article in the April 1 issue. Does this change you view about the role of opioids in the treatment of CLBP? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor

Friday, March 28, 2014

“There are good fast surgeons and bad fast surgeons, but there’s no such thing as a good slow surgeon.” This aphorism seems to be supported by Dr. Kim and colleagues’ recent paper looking at the correlation between duration of surgery for one level lumbar fusion and complications. While duration of surgery has been found to be a risk factor for complications in many surgical disciplines, it had not been evaluated in spine surgery. As such, this group from Chicago used the National Surgical Quality Improvement Program (NSQIP) database to determine the association between operative duration and complications in one level lumbar fusion. They included over 4500 patients undergoing one level lumbar fusion, including posterolateral fusion, XLIF, ALIF, PLIF or TLIF, with or without instrumentation. Patients undergoing multilevel fusion, revision fusion, or pelvic instrumentation were excluded. The authors did not comment about decompression, but there is no indication that patients undergoing decompression as well as fusion were excluded. The duration of surgery was classified as under 2 hours, 2-3 hours, 3-4 hours, 4-5 hours, and greater than 5 hours, and the correlation between duration of surgery and complications within 30 days of surgery was determined. The univariate analysis indicated that duration of surgery was associated with increased overall complications, medical complications, and surgical complications. In the multivariate analyses—the details of which are somewhat unclear—the authors reported that increasing operative duration was associated with overall complication rate, medical complications, superficial infection, and DVT. The other complications tended to only be significantly related to operative time in cases over 5 hours in length. Based on these data, the authors concluded that operative duration is an independent risk factor for complications and could be considered as a quality metric.


Common sense indicates that minimizing operative time while still performing safe and effective surgery should always be a goal, and this and other studies support this. However, the data presented in this database analysis does not strongly support the conclusion that operative duration is a strong, independent predictor of complications. Not surprisingly, increased operative time was associated with age, BMI, gender, inpatient admission, infected or contaminated cases, medical comorbidity, and increased complexity of cases. While the authors performed a multivariate analysis that likely controlled for some of these factors, which covariates were included in the models was not reported. Additionally, posterolateral fusion, XLIF, ALIF, PLIF, and TLIF were also analyzed together, and fusion technique is clearly related to both operative time and complications. Other factors such as performing a simultaneous decompression, hospital or surgeon volume, indications for surgery, having had prior lumbar decompression, and having a perioperative complication (i.e. durotomy, misplaced hardware, etc.) were not included in the analysis, and all of these factors can affect both surgical duration and complications. These and other unmeasured confounders likely played a greater causative role in generating complications than did the actual duration of surgery. Given these issues, the authors’ conclusion that duration of surgery is a useful quality metric is probably not supported by the data. A more thorough analysis controlling for the many potential confounders would be necessary to make such a conclusion, and these other data points are likely not included in the NSQIP database. For now, surgeons should strive to operate as efficiently as is safe, and duration of surgery should not be considered a quality metric on which surgeons are evaluated. If surgeons became incentivized to operate faster, it seems unlikely that outcomes would improve.


Please read Dr. Kim’s article on this topic in the March 15 issue. Does this article change how you perceive the need for efficiency in the OR? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor

Saturday, March 22, 2014

Post-operative analgesia is extremely important after spine surgery, especially as patients expect shorter hospital stays and more procedures are done on an outpatient basis. Opioids and NSAIDs have been the mainstays of treatment, though both have serious side effects and many surgeons feel NSAIDs are contra-indicated after fusion. As such, efforts have been made to develop multimodal analgesia regimens, and neuromodulating medicines such as gabapentin and pregabalin have been shown to be effective adjuncts to opioids and NSAIDs. Additionally, spine surgery can be associated with longer-term neuropathic pain, and it is possible that early use of neuromodulating medications can reduce the risk of this. It is on this background that Dr. Khurana and colleagues from India peformed a double-blind RCT comparing the use of gabapentin, pregabalin, and placebo as part of a multimodal analgesia protocol following lumbar discectomy. They randomized 30 patients to each group, with each patient receiving the allocated treatment pre-operatively and then every eight hours post-operatively for 7 days. In addition, each patient received an IV dose of diclofenac and also received IV tramadol as a rescue medication while hospitalized. Outcome measures included a Visual Analog Scale (VAS) for pain at rest and with motion, the Oswestry Disability Index (ODI), and the Prolo economic and functional score. These were evaluated frequently over the first 72 hours and then followed out to 3 months. In the acute post-operative period, gabapentin and pregabalin were associated with significantly lower VAS scores and less tramadol use compared to placebo. At 3 months, the gabapentin and pregabalin groups had significantly better VAS, ODI, and Prolo scores compared to the placebo group as well. Additionally, the pregabalin group had a moderate but significant advantage over gabapentin on the VAS, ODI, and Prolo scores at 3 months. Side effects and complications related to the medications were mild and included primarily sedation and nausea.


This study represents a high quality study that clearly demonstrates an advantage to gabapentin and pregabalin compared to placebo, albeit a relatively modest one. Given that the side effect profiles of the medications at the dosage used in the study was acceptable—about 10% of patients reported sedation—these medications should be considered as part of a multimodal analgesic protocol as they improved patient pain and function and reduced the need for opioids. Additionally, the longer-term benefit at 3 months is interesting and may represent how early modulation of certain pain pathways can prevent the development of chronic pain in some patients. This study has some limitations, namely that the duration of hospitalization and post-hospitalization use of analgesics was not recorded or reported. In the United States, most lumbar diskectomies are performed on an outpatient basis, so the use of IV narcotics is very limited. It seems as though the patients in this study were admitted for a period of time, and it is unclear if they were prescribed oral opioids or NSAIDs upon discharge. Most multimodal analgesic protocols would include oral NSAIDs and opioids, and it is unclear to what degree these were used in this study. Another limitation is the authors’ decision to convert the ODI to a categorical outcome, which obscures the magnitude of differences between the groups. While the pregabalin group had a greater proportion of patients with an ODI score of less than 20% at 3 months compared to the gabapentin and placebo groups, the actual mean scores were not reported and the differences may have been small. Despite these limitations, this high quality RCT provides relatively strong evidence that gabapentin and pregabalin modestly decrease pain and improve function following lumbar discectomy, even out to 3 months. Hopefully a similar but larger study with a better defined post-discharge protocol will be performed in the future to confirm these results.

Please read Dr. Khurana’s paper on this article in the March 15 issue. Does this change how you view the perioperative use of gabapentinoids in lumbar discectomy? 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.