Friday, June 24, 2016
The current opioid epidemic has raised awareness about the negative consequences of long-term narcotic use. Many patients who become opioid addicts were first exposed to narcotics following surgery or an injury. While opioids will likely continue to have an important role in post-operative pain control, ways to minimize their use need to be considered. The short-term negative effects of narcotics are also problematic for spine surgeons and their patients. With efforts to move spine surgery to the outpatient setting and minimize the length of inpatient stays, side-effects such as nausea, vomiting, ileus, and delirium all present impediments to rapid discharge. As such, interest in multimodal analgesia (MMA) has increased. This approach seeks to capitalize on the synergistic effect of different classes of pain medications, including local anesthetics, neuromodulators, muscle relaxants, anti-inflammatories, acetaminophen, and narcotics. MMA allows for the dose of each medication to be minimized, thus reducing side effects. It has been shown to be effective in total joint arthroplasty and lumbar surgery, though it may have an even larger role in outpatient or short stay procedures. As such, Dr. Bohl and his colleagues from Rush University Medical Center in Chicago recently published their experience with MMA for ACDF. While they initially attempted to perform an RCT, poor protocol adherence resulted in them deciding to abandon the RCT and perform an observational cohort study. They enrolled 239 patients undergoing ACDF by a single surgeon over a 5 year period. Patients were allowed to select their analgesia approach, with 23% selecting MMA and 77% choosing patient-controlled analgesia (PCA). Both protocols were similar and included pre-operative oxycodone, pregabalin, IV acetaminophen, and a local anesthetic injection. The main difference was that the PCA group received a morphine PCA pump on post-op day 0 along with IV fentanyl prn, while the MMA group received tramadol, hydrocodone, and pregabalin on post-op day 0. Starting on post-op day 1, all patients received hydrocodone and cyclobenzaprine. Since this was not a randomized study, the groups were somewhat different, with the PCA group having significantly more comorbidities and a higher rate of smokers. The MMA group used significantly fewer oral morphine equivalents per hour while in the hospital (2.5 mg/hr vs. 5.8 mg/hr), had less nausea/vomiting (6% vs. 38%), and a shorter length of stay (27 hrs vs. 40 hrs). There were no differences in visual analog pain scores in the hospital or the rate of narcotic use at 6 or 12 weeks.
The authors have done a nice job reporting their experience with MMA for ACDF, an outpatient or short-stay procedure at most institutions. They demonstrated a 30% reduction in the length of stay, which may have been due to reducing the negative side effects associated with PCA use. While these results are promising, the reader must remember that this was not a randomized trial and that there were significant baseline differences between the groups that may have been driving the differences. The PCA group had more medical comorbidities and a higher number of smokers, which may have resulted in some of the observed differences (though some prior studies have shown a shorter length of stay for smokers, who want to leave the hospital to smoke). The other interesting result was that over three quarters of patients chose the PCA over MMA. Studies such as this one can be used as evidence to educate patients about the negative short-term effects of narcotics and possibly change their expectation to receive a PCA pump following surgery. As more spine surgery transitions to outpatient or short-stay procedures, the importance of effective analgesia without the side-effects associated with high dose narcotics becomes more important. Hopefully future studies on MMA will employ randomized designs in order to definitively evaluate the benefits of different analgesic regimens.
Please read Dr. Bohl's article on this topic in the June 15 issue. Does this change your approach to peri-operative analgesia? Let us know by leaving a comment on The Spine Blog.
Adam Pearson, MD, MS
Associate Web Editor
Friday, June 17, 2016
Most investigational device exemption (IDE) studies evaluating cervical disc arthroplasty (CDA) devices have showed similar outcomes to ACDF, with some suggesting a lower reoperation rate with CDA. The previous meta-analyses on the topic have combined results from a variety of CDA devices, which increased the power of the studies but injected a degree of heterogeneity to the results. In an effort to pool results from multiple studies of the same device, Dr. Zhu and his colleagues from China performed a meta-analysis of RCTs comparing the Bryan CDA device to ACDF. They identified 8 RCTs, 6 of which were IDE studies. The studies included over 1,800 patients, and most included 2 or 3 years of follow-up (classified as short-term). Only two studies had follow-up to 4 years (classified as medium-term). The Neck Disability Index (NDI) was the only patient reported outcome measure reported, with 2 short-term and 1 medium-term studies reporting NDI outcomes. There was no difference between CDA and ACDF in the pooled short-term studies, while the medium term study showed an advantage for CDA. Three studies showed less operating time for ACDF, with surgical time for ACDF ranging from 15-35 minutes quicker. Adverse events were lumped together as one outcome measure, and ACDF had a significantly higher adverse event rate, primarily due to dysphagia that was recorded in 40% of patients. There were no significant differences in reoperation rates at either the index or adjacent levels.
This study represents another meta-analysis comparing CDA to ACDF that suggests similar overall outcomes. While some prior studies have suggested a higher reoperation rate for ACDF, this study did not confirm that finding. The reoperation rate over 2-4 years was in the 5% range for both interventions, which is substantially lower than the reoperation rate of over 7-12% for ACDF cited in prior meta-analyses.1,2 While this study offers the advantage of evaluating only one CDA device, this also limited the number of studies that could be included and limited the power of the meta-analysis. With only one or two studies with medium-term follow-up for most outcome measures, this study really does not represent a meta-analysis for the medium-term outcomes. Like most other systematic reviews on this topic, this one demonstrated similar outcomes for CDA and ACDF over the short term.3 Long-term RCT data is needed to determine if CDA offers a true advantage in decreasing clinically relevant adjacent segment disease and reducing long-term reoperation rates. Additionally, longer-term studies are needed to provide data on the failure rates of the CDA devices. Given that clinical outcomes in the short and medium term are similar for the two surgical techniques and the long-term outcomes for CDA remain unknown, most surgeons have been slow to adopt the technology.
Please read Dr. Zhu's article on this topic in the June 15 issue. Does this change how you view cervical disc arthroplasty? Let us know by leaving a comment on The Spine Blog.
Adam Pearson, MD, MS
Associate Web Editor
1. Gao Y, Liu M, Li T, Huang F, Tang T, Xiang Z. A meta-analysis comparing the results of cervical disc arthroplasty with anterior cervical discectomy and fusion (ACDF) for the treatment of symptomatic cervical disc disease. J Bone Joint Surg Am 2013;95:555-61.
2. Zhong ZM, Zhu SY, Zhuang JS, Wu Q, Chen JT. Reoperation After Cervical Disc Arthroplasty Versus Anterior Cervical Discectomy and Fusion: A Meta-analysis. Clin Orthop Relat Res 2016;474:1307-16.
3. Boselie TF, Willems PC, van Mameren H, de Bie R, Benzel EC, van Santbrink H. Arthroplasty versus fusion in single-level cervical degenerative disc disease. Cochrane Database Syst Rev 2012:CD009173.
Saturday, June 11, 2016
What to measure to define outcomes following spine surgery has been debated for decades. The early spine literature focused on radiographic parameters (i.e. fusion status), progressed to surgeon reported outcomes (i.e. excellent, good, fair, poor), and finally to patient reported outcomes (i.e. SF-36, ODI, pain visual analog scales, etc.) More recently, there has been increased interest in patient satisfaction, in part related to how hospitals and doctors are rated and, in some cases, paid. The literature has shown relatively weak correlations between these different types of outcome measures, indicating that studies may have markedly different results depending on what they choose for their outcome measures. While some payors and rating agencies have focused on patient satisfaction as the most important outcome, surgeons frequently feel as though they have little control over this. Additionally, there may be little correlation between patient satisfaction and traditional concepts about the quality of care provided. Given the lack of consensus about how to create a meaningful, patient-centered outcome measure for spine surgery, Dr. Mancuso and her colleagues at the Hospital for Special Surgery created expectation scales for lumbar and cervical spine surgery. These scales include 20 questions about what improvement patients expect from surgery in domains including symptoms, function, and emotional well-being. This survey was provided to patients after they decided to go ahead with lumbar (n=366) or cervical (n=133) surgery along with the ODI or NDI, SF-12, and pain visual analog scales. The patients were then contacted 2 years post-surgery, and they again filled out the expectations survey, global satisfaction questions, and the other questionnaires they filled out at baseline. Twenty-four percent of lumbar patients and 31% of cervical patients had all of their expectations met or exceeded, with an average of 66% of lumbar expectations and 78% of cervical expectations being met. There was a relatively strong correlation between the proportion of realized expectations and general satisfaction (r=0.73 for lumbar, 0.62 cervical), while there were no significant correlations between realized expectations and the improvement of leg or arm pain or with improvement in the SF-12 mental component summary score. There were moderate correlations between expectations met and improvement in ODI and NDI (r=0.52 for both). This study did show that expectations about return to work, resolution of axial pain, and improvement of numbness were the least likely to be met.
The authors should be congratulated for creating a truly patient-centered outcome measure for spine surgery. By definition, this outcome measures the factors that patients expect to improve and that are also likely important to the patient. While surgeons may focus on outcomes that they believe surgery can achieve (i.e. resolution of radicular pain, improved walking ability, etc.), these may not be the outcomes that patients care about. This tool should help patients define their expectations pre-operatively and then guide a discussion with the surgeon about the likelihood of surgery achieving those expectations. Patients with unrealistic expectations can be counseled to either adjust their expectations or avoid surgery if it is unlikely to achieve the goals that are important for them. For example, patients who have been out of work for many years and expect to return to work after surgery can be advised that surgery is unlikely to help them to achieve that goal, and they may be better treated with vocational rehabilitation if return to work is their most important goal. One of the limitations of this type of outcome measure is that surgeons may have very limited control over patients realizing their expectations, and the quality of care they provide may not be reflected in the proportion of expectations met. This is similar to patient satisfaction. Surgeons frequently feel as though they have achieved their goals if they can eliminate a patient's radicular pain or claudication, though this study showed poor correlation between the proportion of expectations met and improvement of radiculopathy. While this type of outcome measure may best reflect the success of an operation from the patient's perspective, it is probably not helpful in determining if surgeons are providing high quality care. Future work on outcome measures will need to consider the perspective from which the outcome measure is being considered (i.e. patient, surgeon, payor) and take into account that these measures are frequently measuring very different things.
Please read Dr. Mancuso's article on this topic in the June 1 issue. Does this change your view about patient reported outcomes? Let us know by leaving a comment on The Spine Blog.
Adam Pearson, MD, MS
Associate Web Editor
Saturday, June 4, 2016
Hospital acquired conditions (HACs) have received much attention as the Centers for Medicare and Medicaid Services (CMS) have linked hospital payments and penalties to rates of these complications. Currently, these include incompatible blood transfusions, retained surgical objects, pressure ulcers, hospital falls, poor glycemic control, air embolism, central line associated bloodstream infection (CLABSI), catheter associated urinary tract infection (CAUTI), deep venous thrombosis (DVT), and surgical site infection (SSI). No spine specific complications have been added to the list, but it seems likely that condition-specific HACs will be developed in the future. In an attempt to define which of the current HACs and spine-specific complications are avoidable and unavoidable, Dr. Khaleel and colleagues from Boston performed a survey of spine surgery fellowship directors to rate complications on a spectrum from completely avoidable (1) to unavoidable (10). They surveyed 76 fellowship directors, and 46 responded (61% response rate). There was near universal agreement that incompatible blood transfusion, retained surgical objects, and wrong level surgery were completely avoidable. No complication was considered completely unavoidable by all respondents, but SSI in a patient with risk factors (median response 8), C5 palsy following cervical laminectomy (8), iliac crest bone graft site pain (8), and recurrent herniation following discectomy (9) were rated as nearly unavoidable. Poor glycemic control (3), hospital falls (4), CLABSI (5), DVT (6), CAUTI (6), and dural tear (7) were rated as intermediate on the preventability spectrum.
The results of this survey are interesting as they indicate that spine surgery experts agree that certain "never events" are in fact completely preventable, while others fell towards the unavoidable end of the spectrum (i.e. SSI in a patient with risk factors). There was also agreement that CLABSI, CAUTI, and DVT were not completely preventable. Given that many of these events that can or will affect reimbursement and may not be preventable, it indicates the need for spine surgeons to be involved in the process of defining quality measures for our field. Many of the current HACs are not within the control of the spine surgeon, and studies such as these can be used to lobby policy makers to adopt payment schemes that reward aspects of care under surgeon control that have meaningful impacts on patient outcomes. If reimbursement models persist that penalize hospitals and surgeons for unpreventable "never events" (i.e. SSI), patients at high risk for these complications may be denied care that could markedly improve their quality of life. Hopefully spine surgeons will remain involved in this type of work and continue to advocate for policies that support our patients and our profession.
Please read Dr. Khaleel's article on this topic in the June 1 issue. Does this paper change how you view the preventability of spine surgery complications? Let us know by leaving a comment on The Spine Blog.
Adam Pearson, MD, MS
Associate Web Editor
Friday, May 27, 2016
C5 palsy (C5P) is a poorly understood and relatively common complication after cervical spine surgery. It has been suggested that posterior procedures and wider laminectomy troughs are risk factors for developing C5P, though other studies have questioned those beliefs.1-4 Dr. Lee and his colleagues from Korea wanted to compare rates and outcomes of C5P following laminectomy and fusion (LF) vs. laminoplasty (LP). They retrospectively reviewed outcomes for 90 LF patients and 100 LP patients. They found a 29% rate of C5P following LF compared to 4% following LP. The severity of C5P was also more severe for the LF group, with a manual motor testing score of 2.1 for the LF C5P group and 3.5 for the LP patients with C5P. The LF group with C5P also took longer to recover (21 weeks vs. 11 weeks), and more patients in the LF group had incomplete recovery (15% vs. 0%). The authors evaluated risk factors for C5P among the LF patients and found that those who developed C5P were more likely to have myeloradicular symptoms compared to pure myelopathy (88% in the C5P group vs. 42% in the non-C5P group), more likely to have pre-operative motor weakness (65% C5P vs. 38% non-C5P), and more likely to have severe C4-C5 foraminal stenosis (77% C5P vs. 56% non-C5P). They concluded that C5P was more common after LF than LP and that the palsy was more severe after LF.
While C5P has been a recognized entity for decades, its etiology remains elusive. This study confirms the findings of previous studies that have suggested a higher rate of C5P following LF compared to LP.1,3 However, the rate of C5P in the LF cohort in this study is substantially higher than what has typically been reported in the literature. Nassr et al. reported a rate of 9.5% following LF in their large series, and Basaran et al. found a rate of 10.5% in their meta-analysis.1,3 It is unclear why the rate in this study was nearly three times higher, and it could be related to variation in underlying pathology, different surgical technique, or a more sensitive diagnostic process to identify C5P. It has been suggested that a wider laminectomy trough predisposes the patient to C5P, and the authors of the current study do not comment on the width of their decompression.4 While retrospective cohort studies have noted a higher rate of C5P in the LF population compared to the LP population, no randomized trial has been performed to evaluate this. By definition, the LF and LP populations are different and were selected to undergo different procedures. As such, it is unclear if the cause of the higher rate of C5P following LF is due to differences in the underlying pathology or is attributable to surgical technique. Until a Level 1 study is performed on this topic, we probably will not know the answer. Given that many patients for whom LF is indicated are not good candidates for LP (i.e. those with significant neck pain, neutral or slightly kyphotic posture, segmental instability, radiculopathy, etc.), the inclusion criteria for such a study would likely be so strict that the results would not generalizable to most of the myelopathy population. Given that C5P is a relatively benign complication in the long-term and that there is not clear evidence that surgical technique (i.e. LF vs. LP) contributes to its development, the rate of C5P should probably not be a main driver of surgical decision-making. The patient should undergo the procedure that the surgeon thinks will lead to the best outcome in his or her hands.
Please read Dr. Lee's article on this topic. Does this change how you view LF vs. LP? Let us know by leaving a comment on The Spine Blog.
Adam Pearson, MD, MS
Associate Web Editor
1. Basaran R, Kaner T. C5 nerve root palsy following decompression of cervical spine with anterior versus posterior types of procedures in patients with cervical myelopathy. Eur Spine J 2016.
2. Klement MR, Kleeman LT, Blizzard DJ, Gallizzi MA, Eure M, Brown CR. C5 palsy after cervical laminectomy and fusion: does width of laminectomy matter? Spine J 2016;16:462-7.
3. Nassr A, Eck JC, Ponnappan RK, Zanoun RR, Donaldson WF, 3rd, Kang JD. The incidence of C5 palsy after multilevel cervical decompression procedures: a review of 750 consecutive cases. Spine (Phila Pa 1976) 2012;37:174-8.
4. Radcliff KE, Limthongkul W, Kepler CK, et al. Cervical laminectomy width and spinal cord drift are risk factors for postoperative C5 palsy. J Spinal Disord Tech 2014;27:86-92.