Friday, April 29, 2016
The role of vertebroplasty (VP) remains poorly defined. Some physicians have suggested that it should be used more frequently for acute, painful fractures,1 while others think it should rarely if ever be used based on the results of two randomized trials that showed no benefit compared to a sham procedure2,3 Those trials have been criticized for including primarily patients who had fractures over six weeks old and some with relatively mild symptoms. Many prior studies without a sham control have shown a benefit of VP, and anecdotal evidence suggests that VP helps some patients. On the other hand, the vast majority of compression fractures will go onto heal uneventfully with minimal or no treatment over the course of 6-10 weeks. Given that most evidence on this topic has looked at outcomes for patients with subacute or chronic fractures, Dr. Yang and colleagues from Shanghai performed a non-blinded RCT comparing VP to conservative management, which included at least two weeks of bedrest followed by mobilization with bracing. All patients had acute compression fractures, with the average symptom duration prior to enrollment being 5.5 days. They randomized 135 patients, with 107 remaining compliant with their treatment assignment and completing 1 year follow-up. The VP group had better outcomes on VAS back pain, Oswestry Disability Index (ODI), and a measure specific to osteoporotic compression fractures (QUALEFFO). In the year following enrollment, 9% of VP patients and 8% of conservatively treated patients sustained at least 1 more compression fracture. The authors reported no technical complications related to VP but did note a higher rate of complications related to immobilization (i.e. pneumonia, DVT, UTI, depression, etc.) in the conservative treatment group (35% vs. 16%).
Similar to other non-blinded, randomized trials comparing VP to conservative treatment, this study showed a significant advantage for VP.1 Physicians now have to reconcile conflicting data between the non-blinded RCTs favoring VP and the blinded, sham-controlled RCTs that showed minimal or no benefit to VP. Aside from the lack of blinding, this study differed in two important ways from the sham-controlled trials. All patients in the current trial had acute fractures, with vertebroplasty being performed at an average of 3 days following diagnosis and 8 days after the onset of symptoms. Additionally, the control group in the current study was very different, with patients being prescribed a minimum of 2 weeks of bedrest, with the average patient spending 33 days at bedrest. In the United States, only a small proportion of patients require any bedrest at all, and patients are generally encouraged to mobilize as soon as possible. Two weeks of bedrest is certainly not the standard of care, and prescribing such a treatment would be considered bordering on malpractice. Given that prolonged bedrest is not a recommended treatment and is known to be associated with many complications and deconditioning, the advantages observed for VP in this trial may be due less to the benefits of VP and more related to the negative effects of prolonged bedrest. Additionally, while VP may provide faster resolution of pain than the natural history, by 6-10 weeks most fractures should be healed, and it seems unlikely that VP should result in much advantage beyond 3 months. At longer term follow-up, the advantage of VP is likely due to either the placebo effect or the negative results of prolonged bedrest. The authors seem to conclude that VP should be used more frequently for patients with acute osteoporotic compression fractures, but that conclusion is not supported by the evidence. What all of the VP studies have demonstrated is that we do not know which patients benefit most from the intervention. Patients who are bedbound by acute fractures and those who are severely limited by subacute fractures seem like two groups who are likely to benefit from VP, but these subgroups have not been well-studied. Until further studies better define the patients who stand to benefit from VP, physicians will continue to rely on intuition rather evidence while helping patients make this treatment decision.
Please read Dr. Yang's article on this topic in the April 15 issue. Does it change how you view the role of vertebroplasty? Let us know by leaving a comment on The Spine Blog.
Adam Pearson, MD, MS
Associate Web Editor
1. Farrokhi MR, Alibai E, Maghami Z. Randomized controlled trial of percutaneous vertebroplasty versus optimal medical management for the relief of pain and disability in acute osteoporotic vertebral compression fractures. J Neurosurg Spine 2011;14:561-9.
2. Buchbinder R, Osborne RH, Ebeling PR, et al. A randomized trial of vertebroplasty for painful osteoporotic vertebral fractures. N Engl J Med 2009;361:557-68.
3. Kallmes DF, Comstock BA, Heagerty PJ, et al. A randomized trial of vertebroplasty for osteoporotic spinal fractures. N Engl J Med 2009;361:569-79.
Friday, April 22, 2016
The role of fusion in degenerative spondylolisthesis (DS) was hotly debated in the 1990s, when two studies were performed suggesting that patients undergoing fusion in addition to laminectomy had better outcomes and less progression of listhesis than patients undergoing laminectomy alone.1,2 Based largely on the results of these studies, laminectomy and fusion became the standard treatment for DS, and over 95% of DS patients undergoing surgery in the United States now undergo a fusion.3 More recently, the role of fusion has been questioned, especially as less invasive decompressive techniques have been developed.4 As a result, two recent RCTs were performed, with the results published in the New England Journal of Medicine last week. Forsth et al. reported the results of the Swedish Spinal Stenosis Study (SSSS), which included 247 patients, 135 of whom had DS.5 They randomized patients to either decompression alone (D) or decompression and fusion (DF). Ghogawala et al. performed the Spinal Laminectomy vs. Instrumented Pedicle screw (SLIP) study that similarly randomized 66 DS patients to D or DF. The primary outcome in the SSSS was the Oswestry Disability Index (ODI) at two year follow-up. There were no significant differences on this outcome (16 point improvement in DF group, 20 point improvement in D group), or any other patient reported outcome, through 5 years of follow-up. In the SLIP trial, the SF-36 Physical Component Summary (PCS) at 2 years was the primary outcome measure, and the DF group improved by about 6 points more than the D group (a clinically and statistically significant difference). Additionally, the 4 year reoperation rate was 34% in the D group (all repeat surgeries were at the index level) and 14% in the DF group (all repeat surgery at adjacent levels). In contrast, reoperation rates in the SSSS were the same for the two groups (22% D vs. 21% DF over 6.5 years). The SLIP trial also evaluated ODI scores and reported an improvement of 26 points in the DF group vs. 18 in the D group at 2 years (p=0.06). The authors of the two studies reached essentially opposite conclusions, with the SSSS concluding that there was no advantage of fusion, while the SLIP trial determined that fusion both improved clinical outcome measures and reduced reoperation rate.
These studies present a conundrum for the practicing spine surgeon, since they both appeared to use similar methodologies to study the same question, yet reached substantially different conclusions. The patients in the two studies seem quite similar at baseline and had similar operations. The main driver in the difference in ODI improvement at 2 years was that the DF group in the SLIP trial improved substantially more than the DF group in the SSSS (26 vs. 20 points). The outcomes for the D group in the two studies was similar (18 points SLIP vs. 20 points SSSS). The degree of ODI improvement at 2 years in the SLIP trial DF group was similar to that observed in the DS surgical group in the Spine Patient Outcomes Research Trial (SPORT, 24 points), with the vast majority of these patients undergoing DF.6 It is unclear what was driving these moderate differences in outcomes in the DF groups between the American and Swedish studies, though there may be unmeasured differences in the patient population or unreported differences in surgical techniques. There may also be cultural or language differences that affect how patients report their outcomes. For example, among DS patients undergoing DF, 89% of the SPORT patients reported satisfaction with their outcomes compared to 64% in the SSSS. The differences in reoperation rates also raise questions. While the reoperation rates in the SSSS were similar for the two treatment groups, the reoperation rate was over twice as high for the D group compared to the DF group in the SLIP trial. While there was no D group in the SPORT DS study, the 4 year reoperation rate among the DF fusion patients was 15%, nearly identical to the DF patients in the SLIP trial (14%). Reoperation rate is a difficult outcome measure to interpret, as it reflects the subjective preferences of the surgeons and patients involved in the treatment decision. It would have been nice to have patient reported outcome data on the patients undergoing reoperation both before and after the revision surgery. Even after thorough analysis of these papers, it may not be possible to completely explain the differences in outcomes. A superficial explanation may simply be that American and Swedish patients and surgeons are different, and the results in one nation may not generalize to the other. These studies do reinforce the concept that DS represents a wide spectrum of disease that affects a heterogeneous patient population. We currently tend to treat DS with a one size fits all approach, with the majority of patients in the United States undergoing a laminectomy and instrumented posterolateral fusion (with or without interbody support). These studies reinforce that some DS patients do not derive much benefit from a fusion, and different surgical techniques may be best suited to different patient populations. Unfortunately, there is no scientific data to guide surgical technique selection in DS based on individual patient characteristics. Until there is, many surgeons will likely continue to perform decompression and fusion in an effort to avoid a difficult revision surgery at the index level. At the same time, surgeons may be more comfortable performing a laminectomy alone in older patients with medical comorbidities with stable appearing slips.
While these studies were not published in Spine, they are of great interest to the Spine readership. Do they change how you approach selecting a surgical technique in DS? Let us know by leaving a comment on The Spine Blog.
Adam Pearson, MD, MS
Associate Web Editor
1. Bridwell KH, Sedgewick TA, O'Brien MF, Lenke LG, Baldus C. The role of fusion and instrumentation in the treatment of degenerative spondylolisthesis with spinal stenosis. J Spinal Disord 1993;6:461-72.
2. 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.
3. 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.
4. Joaquim AF, Milano JB, Ghizoni E, Patel AA. Is There a Role for Decompression Alone for Treating Symptomatic Degenerative Lumbar Spondylolisthesis?: A Systematic Review. J Spinal Disord Tech 2015.
5. Forsth P, Olafsson G, Carlsson T, et al. A Randomized, Controlled Trial of Fusion Surgery for Lumbar Spinal Stenosis. N Engl J Med 2016;374:1413-23.
6. Weinstein JN, Lurie JD, Tosteson TD, et al. Surgical versus nonsurgical treatment for lumbar degenerative spondylolisthesis. The New England journal of medicine 2007;356:2257-70.
Saturday, April 16, 2016
Type II dens fractures are the most common cervical spine fractures in the elderly population and are associated with high rates of morbidity and mortality with both surgical and non-operative treatment. Surgery results in a higher radiographic union rate, but it is unclear if this leads to better clinical outcomes.1 Recent observational studies have suggested that surgery may decrease mortality, though all of these studies were at risk of selection bias, and surgeons may have simply been operating on healthier patients.2-4 In order to more formally evaluate the treatment decision process for elderly patients with Type II dens fractures, Dr. Barlow and colleagues from Dartmouth (myself included) performed a cost-effectiveness analysis. Given that mortality rates differ across age strata, the authors stratified the analysis for patients aged 65-74, 75-84, and over 84 years. Mortality, complication, utility, and cost data were generally derived from the literature, though some utility data were estimated. Not surprisingly, surgery was more expensive than non-operative treatment. Prior studies suggested that surgery conveyed a survival analysis for patients under age 85, which resulted in a substantial quality adjusted life year gain for this patient group. The cost-effectiveness analysis indicated that surgery was highly cost-effective for patients under 85, though it was more costly and less effective for patients 85 and up. Sensitivity analysis revealed the analysis was quite stable across a reasonable range of variables other than mortality rate, to which the model was quite sensitive.
This paper helps frame the decision-making process and demonstrated that surgery for Type II dens fractures in the elderly population is highly cost-effective if it conveys a mortality advantage. The key unknown that the study revealed is our current inability to identify patients for whom surgery increases survival. Very healthy patients are unlikely to die from complications related to non-operative treatment, and it is unclear how much benefit they derive by decreasing the non-union rate with surgery.1 On the other end of the spectrum, sick, elderly patients with a high comorbidity burden are unlikely to survive surgery and should be treated non-operatively. Between these two extremes, there likely exists a group of patients with moderately limited physiological reserve who can tolerate surgical treatment but might succumb to the complications associated with non-operative treatment such as immobility, dysphagia, delirium, and aspiration. For this group, surgery likely conveys a mortality advantage. The results of the current study needs to be considered in light of the limitations inherent with such a study design. Cost-effectiveness analyses are only as good as the studies that provide the data on which they are based. In this case, there are no level 1 studies on the topic, so the strength of the underlying evidence is moderate at best. Clinicians should not base treatment decisions on the arbitrary age cut-off of 85 and understand that age is a proxy for comorbidity burden. A recent Medicare database study that controlled for comorbidity burden suggested that all age groups, including those over age 85, still derived a survival benefit from surgery.4 At this point, the spine community is in need of a Level 1 study to answer this question more definitively, though the logistical barriers to performing such a study are significant. Until such a study is done, clinicians must use their judgment about which patients are likely to benefit from surgery and employ a shared decision making process so that this vulnerable population makes decisions in-line with their personal preferences.
Please read Dr. Barlow's article on this topic in the April 1 issue. Does this change how you view the treatment of Type II dens fractures in the elderly? Let us know by leaving a comment on The Spine Blog.
Adam Pearson, MD, MS
Associate Web Editor
1. Smith JS, Kepler CK, Kopjar B, et al. Effect of type II odontoid fracture nonunion on outcome among elderly patients treated without surgery: based on the AOSpine North America geriatric odontoid fracture study. Spine (Phila Pa 1976) 2013;38:2240-6.
2. Schoenfeld AJ, Bono CM, Reichmann WM, et al. Type II Odontoid Fractures of the Cervical Spine: Do Treatment Type and Medical Comorbidities Affect Mortality in Elderly Patients? Spine (Phila Pa 1976) 2011;36:879-85.
3. Vaccaro AR, Kepler CK, Kopjar B, et al. Functional and quality-of-life outcomes in geriatric patients with type-II dens fracture. The Journal of bone and joint surgery American volume 2013;95:729-35.
4. Pearson AM, Martin BI, Lindsey M, Mirza SK. C2 Vertebral Fractures in the Medicare Population: Incidence, Outcomes, and Costs. J Bone Joint Surg Am 2016;98:449-56.
Saturday, April 9, 2016
While BMP-2 likely increases fusion rates in ACDF, it has clearly been shown to increase airway problems and dysphagia.1 Many spine surgeons now feel it is contra-indicated in anterior cervical surgery except in cases that carry an exceptionally high risk of pseudarthrosis, and the FDA has added a black box warning about risks of airway complications. Surgeons who believe that BMP-2 could provide some advantages in the cervical spine have sought out ways to mitigate these complications, including using lower doses of BMP-2 and local or systemic steroids to reduce inflammation. Dr. Edwards and his colleagues from Baltimore wanted to rigorously evaluate the potential benefit of local steroids, so they performed a double-blinded, randomized, placebo-controlled trial comparing a retroesophageal sponge soaked with 40 mg of depomedrol to a sponge soaked with saline in patients undergoing ACDF using 0.4 mg of BMP-2 per level. Their main outcome measure was the 4-point modified dysphagia scale (0=no dysphagia, 4= severe dysphagia) measured on post-operative days 1, 4, 7, 14 and 28. Twenty-seven patients received steroid, while 23 received the control treatment. At baseline, the groups were similar except for the steroid group including fewer multilevel fusions than the control group (44% vs. 70%). The steroid group had significantly lower dysphagia scores on every day except post-operative day number 1, with the greatest difference observed on day 4 (0.54 vs. 1.74). On the other days, the score difference between the two groups was less than 1. The steroid group also had a higher percentage of patients reporting no dysphagia at every time point except day 1 (56% vs. 26% on day 28). More patients in the control group received systemic steroids post-operatively for dysphagia, though these differences were not significant. One patient in the control group was readmitted for airway problems and dysphagia. The study had initially planned to enroll 60 patients, but it was stopped early after enrolling 54 patients (4 were excluded after enrollment) due to the significant benefit of steroid.
The authors should be congratulated for successfully performing a Level 1 study demonstrating the efficacy of local steroid application following ACDF using low dose BMP-2. While the benefits of steroid were statistically significant, it is unknown if the difference is clinically important. No formal studies have been performed to assess the minimal clinically important difference on the modified dysphagia scale, though the authors estimated it at 1 while performing their power calculations. The difference in scores only exceeded 1 on post-operative day 4, so it is unclear if the benefit of steroids was clinically meaningful later in the study. The randomization also failed to equalize the proportion of patients undergoing multilevel fusion, and this confounding factor could have played a role in the greater degree of dysphagia in the control group, more of whom underwent multilevel ACDF. Nonetheless, the study did demonstrate the efficacy of local depomedrol to reduce dysphagia to some degree. Unfortunately, the study was not designed to look at safety, and there were far too few patients to evaluate potential complications associated with local steroids such as infection or nonunion. The real question raised by this study is the role of BMP-2 in routine anterior cervical surgery. The fusion rates for single and two-level ACDF using allograft and an anterior plate are quite high, and it is unclear if the potential benefit of BMP-2 justifies the potential complications and costs associated with its use. It may have a role in three or four level ACDF or in cases with compromised biology (i.e. established nonunion). Before surgeons consider going back to using BMP-2 in routine anterior cervical surgery, future studies need to better define the situations in which it is indicated, the proper dose, and the role of local or systemic steroids. These studies need to include a control group of patients who do not receive BMP-2, which was notably absent in this investigation.
Please read Dr. Edwards's paper in the April 1 issue. Does this change how you view the role of BMP-2 and steroids in anterior cervical surgery? Let us know by leaving a comment on The Spine Blog.
Adam Pearson, MD, MS
Associate Web Editor
1. Smucker JD, Rhee JM, Singh K, Yoon ST, Heller JG. Increased swelling complications associated with off-label usage of rhBMP-2 in the anterior cervical spine. Spine (Phila Pa 1976) 2006;31:2813-9.
Friday, April 1, 2016
Advocates of minimally invasive transforaminal lumbar interbody fusion (MIS TLIF) have suggested that the procedure results in less soft-tissue damage, lower blood loss, and shorter hospital stays with similar or better long-term outcomes compared to open TLIF.1 Others have reported on steep learning curves accompanied by high complication rates with few clinically measurable benefits.2,3 The quality of evidence supporting either position is quite weak.4 In order to add further data to the literature, Dr. Guan and his colleagues from the University of Utah used the data they had collected at their institution through the National Neurosurgery Quality and Outcomes Database to compare outcomes between patients undergoing MIS and open TLIF. Patients were not randomized, with one surgeon performing all of the MIS TLIFs (n=44) and multiple other surgeons performing the open procedures (n=54). About 85% of patients underwent single level fusions, with the remainder undergoing two level arthrodesis. The major baseline difference between the groups was that the MIS TLIF patients were 10 years younger than the open TLIF group (44 vs. 54 years). Another notable finding was that the mean baseline Oswestry Disability Index (ODI) score was under 25 in both groups, indicating relatively low levels of disability at baseline. Patient reported outcomes were not significantly different between the groups at 3 months or 1 year, though approximately 50% of patients were lost to follow-up by 12 months. Blood loss was lower in the MIS TLIF group (120 mL vs. 306 mL), though no patients required a transfusion. Operating time was significantly higher in the MIS group (5.5 hours vs. under 4 hours). Surprisingly, length of stay was over a day longer in the MIS TLIF group (5 days vs. 3.8 days), and significantly more MIS TLIF patients were discharged with home health care or to a skilled nursing facility (36% vs. 13%). No data on complications, re-operations, readmissions or fusion rate were reported.
This study adds some relatively high quality, prospectively collected data to the MIS TLIF literature. However, this is an observational study comparing patients with significant baseline differences and being treated by different surgeons, who likely have different indications for surgery, different skill levels, and different post-operative treatment protocols. As such, it is impossible to conclude if the observed differences were due to the different surgical techniques, different patient populations, or different surgeons. The patient reported outcomes at 3 and 12 months were similar, suggesting that surgical technique may not affect outcomes at these time points. The data support the generally accepted belief that MIS procedures lead to less blood loss but take more time to perform. The longer length of stay and higher rate of non-routine discharge among the MIS group is contrary to what some MIS advocates have reported, though this may be a function more of differences between surgeons than differences between techniques. The lack of data on complications, reoperations, and pseudarthrosis is a major limitation, as prior literature has suggested that these could be higher in MIS procedures.2,5 The main question that remains about MIS procedures is whether the potential benefits from less soft-tissue damage and lower blood loss justify a procedure that is technically more difficult to perform and may be associated with a less effective decompression, higher technical complication rate, and lower fusion rate in the hands of average surgeons. This paper adds to a growing number of observational studies suggesting that MIS techniques do not lead to better patient reported outcomes, and the major advantage is less blood loss. However, blood transfusion following one level TLIF is relatively uncommon, so the benefit of lower blood loss may be minimal. No level one data exist on this topic, and an RCT is the only way to answer most of the questions about MIS techniques. Given the heterogeneity of MIS procedures and indications for their use, defining the most appropriate techniques and patient groups to study in an RCT is challenging. Until such a study is done, surgeons will continue to base their decisions about surgical technique on opinion rather than evidence.
Please read Dr. Guan's article on this topic in the April 1 issue. Does this change how you view the pros and cons of MIS TLIF? Let us know by leaving a comment on The Spine Blog.
Adam Pearson, MD, MS
Associate Web Editor
1. Djurasovic M, Rouben DP, Glassman SD, Casnellie MT, Carreon LY. Clinical Outcomes of Minimally Invasive Versus Open TLIF: A Propensity-Matched Cohort Study. Am J Orthop (Belle Mead NJ) 2016;45:E77-82.
2. Jin-Tao Q, Yu T, Mei W, et al. Comparison of MIS vs. open PLIF/TLIF with regard to clinical improvement, fusion rate, and incidence of major complication: a meta-analysis. Eur Spine J 2015;24:1058-65.
3. Sidhu GS, Henkelman E, Vaccaro AR, et al. Minimally invasive versus open posterior lumbar interbody fusion: a systematic review. Clin Orthop Relat Res 2014;472:1792-9.
4. Goldstein CL, Macwan K, Sundararajan K, Rampersaud YR. Comparative outcomes of minimally invasive surgery for posterior lumbar fusion: a systematic review. Clin Orthop Relat Res 2014;472:1727-37.
5. 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.