Fusion of the cervical spine is among the most common interventions for degenerative cervical pathology, and both anterior and posterior operative approaches are in the surgical repertoire of spine surgeons.1,2 The objectives of decompression of symptomatic compressive pathology and fusion for degenerative or iatrogenic instability must be met by the decided intervention, and certainly such procedures have been demonstrated as being cost-effective for quality of life restoration when applied in the context of cervical spondylotic myelopathy (CSM).3,4 Although randomized studies defining the relative benefits and perioperative morbidity of anterior and posterior approaches for diffuse or multifocal cervical spondylosis are lacking, there are data that suggest that perioperative complication rates and societal cost-effectiveness are both well served by anterior approaches when clinically appropriate.5–8 Considerations about surgical approach involve the pathology and clinical presentation, relative location of the stenosis, cervical spine sagittal alignment, and other patient-specific factors.9–11
If an anterior approach is selected, there is further controversy as to whether optimal decompression and long-term stabilization is best accomplished with multilevel discectomies, corpectomies, or a combination of the 2. Concerns about multilevel anterior cervical procedures relate to the extensive nature of the operation, the long vertical segment over which potentially limited sagittal correction may be achieved, and the potential to achieve rigid bony fusion effectively.12–14 The evidence comparing these various strategies is limited, both by the challenges of examining this topic in a rigorous randomized fashion alongside the specific variability in surgical techniques, graft and instrumentation choices, and patient selection process. The potential advantages of discectomy-based approaches include potentially lesser intraoperative bleeding than may occur with corpectomy, greater opportunity to correct sagittal alignment, and the ability to achieve segmental fixation. An example of this is illustrated in Figure 1. The potential advantages of corpectomy-based approaches include more complete decompression that is paramount when there is significant canal stenosis directly behind the vertebral body, fewer bone-to-graft interfaces, and potentially greater fusion potential (as a strut graft in a trough may allow for side-to-side healing). An example of this is illustrated in Figure 2. Nevertheless, the necessity of gaining bony fusion and the positive impact of achieving that with corpectomy has hitherto remained unproven.
The goal of this systematic review is to compare clinical, radiographical, and perioperative, outcomes after various anterior cervical procedures used to approach diffuse or multifocal CSM (Table 1). The primary outcomes were clinical (neurological assessments and neck pain scores), radiographical (fusion rates and sagittal alignment), and perioperative safety (complications) of the different anterior procedures. Comparisons were performed between multiple discectomies versus single and multiple corpectomy, multiple discectomies versus discectomy-corpectomy hybrid, and discectomy-corpectomy hybrid procedures versus multiple corpectomies. Furthermore, given the complexity of multiple anterior reconstructive procedures, as a secondary objective we also summarize the most common ancillary stabilization techniques used among the included studies to help guide readers with the different anterior decompressive procedures.
In summary, this work addresses the following primary key questions:
- What is the comparative effectiveness (CE) of the following anterior decompression procedures for multilevel CSM?
- Multiple discectomies versus single or multiple corpectomy
- Multiple discectomies versus discectomy-corpectomy hybrid
- Multiple corpectomies versus discectomy-corpectomy hybrid
- What is the comparative safety of the following anterior decompression procedures for multilevel CSM?
- Multiple discectomies versus single or multiple corpectomy
- Multiple discectomies versus discectomy-corpectomy hybrid
- Multiple corpectomies versus discectomy-corpectomy hybrid
We further addressed the following secondary key question:
- What are the most common ancillary stabilization techniques used among patients undergoing multiple discectomy, single or multiple corpectomy, or discectomy-corpectomy hybrid procedures for multilevel spondylosis?
When confronted with this armamentarium of possible treatments, the surgeon would be well served to have evidence to drive decision making and patient education. Each approach may offer variable benefits with different likelihood of complications, thus appropriate treatment will likely continue to be tailored to individual patient clinical presentation and pathoanatomy. Nevertheless, with this work we sought to compare the effectiveness (outcomes) and safety (complications) of the various anterior approaches to multilevel CSM.
MATERIALS AND METHODS
Electronic Literature Search
We conducted a systematic search in MEDLINE and the Cochrane Collaboration Library for literature published through September 2012, limiting the search results to human studies published in the English language. Reference lists of key articles were also systematically checked to identify additional eligible articles. The first key question focused on effectiveness of anterior surgical approaches on improving clinical and radiographical outcomes, and we identified studies comparing multiple discectomies versus single or multiple corpectomy, multiple discectomies versus discectomy-corpectomy hybrid, and multiple corpectomies versus discectomy-corpectomy hybrid. Outcomes of interest included clinical functional outcomes, neurological outcomes, pain scores, radiographical fusion rates, and sagittal alignment. The second key question focused on safety of anterior surgical approaches on various perioperative complications, and we identified the same comparative studies from key question one to compare incidences of commonly reported complications. A secondary objective focused on summarizing ancillary methods used in conjunction with these anterior surgical procedures to provide for stabilization. We sought to determine the most commonly used ancillary stabilization devices used in the studies that formed the basis of this systematic review. Exclusion criteria included patients with degenerative disc disease or degenerative joint disease without CSM, single-level CSM, spinal tumor, concomitant infection, ossification of the posterior longitudinal ligament, and ankylozing spondylitis. We excluded studies that did not report results separately by treatment group because comparisons could not be made. Furthermore, studies that differentially selected treatments based on important baseline characteristics (e.g., age, number of diseased levels) were excluded because they create comparisons with significant selection bias influencing outcome. Case series, case reports, data not reported separately for each comparison group, or studies that consisted of an N less than 10 for either comparison group were excluded. Other studies excluded were animal, cadaveric, and biomechanical studies. A summary of this selection process is demonstrated in Table 1.
We extracted the following data from the included articles: study design, patient demographics, diagnosis and operated levels, follow-up duration and the follow-up rates for each treatment group, and treatment intervention descriptions. From these studies, we report summary side-by-side data for Japanese Orthopaedic Association (JOA) scores, visual analogue scale (VAS) scores for neck pain, neck disability index (NDI) scores, sagittal alignment, cervical lordosis, and complication rates.
Study Quality and Overall Strength of Body of Literature
Level-of-evidence ratings were assigned to each article independently by 2 reviewers (J.T.H., D.C.N.) using criteria set by The Journal of Bone & Joint Surgery, American Volume15 for therapeutic studies, and modified to delineate criteria associated with methodological quality described elsewhere.16 (See Supplemental Digital Content available at http://links.lww.com/BRS/A824 for individual study ratings.)
The overall body of evidence with respect to each key question was determined on the basis of precepts outlined by the GRADE (Grades of Recommendation Assessment, Development, and Evaluation) Working Group17 and recommendations made by the Agency for Healthcare Research and Quality.18 Risk of bias was evaluated during the individual study evaluation described earlier in this section. This system, which derives a strength-of-evidence grade of “high,” “moderate,” “low,” or “insufficient” for each outcome or key question, is described in further detail in the Methodology article in this focus issue.9 A detailed description of how we arrived at the strength of evidence for each key question can be found in the Supplemental Digital Content available at http://links.lww.com/BRS/A824.
We performed all analyses on an individual study level. For continuous outcome measures, we reported, or if necessary calculated, the mean change scores and standard deviations. We imputed the change score standard deviations using a formula recommended by the Cochrane Handbook for Systematic Review of Interventions.19 Mean baseline, mean follow-up, mean change scores, and their respective standard deviations allowed calculation of a standardized mean difference (SMD) that permitted comparison of change in scores between anterior surgical procedures in the different studies. SMDs and their 95% confidence intervals are displayed using forest plots, with a random effects model assumed to address study heterogeneity. Cohen20 has reported an effect size of 0.2 to 0.3 as a “small” effect, around 0.5 a “medium” effect, and 0.8 to infinity, a “large” effect. We used unpaired t tests to determine if the outcome score changes were variable between different approaches. Significant results are those in which the confidence interval did not cross the neutral line in the forest plot. Because of significant clinical and statistical heterogeneity, we did not combine studies into a meta-analysis; rather, presented studies side by side in summary tables and figures to make a qualitative assessment of treatment effectiveness and safety. Ancillary stabilization devices were reviewed on a per-study basis for each anterior procedure of interest, with increasing invasiveness from no device, to orthosis, to halo vest, to posterior stabilization. To calculate the percentage use of each stabilization device by surgical procedure, we considered the denominator as the number of patients having undergone the specific anterior procedure of interest (i.e., discectomy, corpectomy, discectomy- corpectomy hybrid procedure). For the numerator, we counted the ancillary device prescribed for the specific treatment as stated by the author. If multiple devices were used for a treatment type, we only counted the one used in the majority of cases in that study. This occurred only one time and the procedure most commonly used was prescribed in over 90% of the subjects. We performed all statistical analyses using Stata 9.1 (StataCorp, College Station, TX).21
Clinical Recommendations and Consensus Statements
Clinical recommendations or consensus statements were made through a modified Delphi approach by applying the GRADE/Agency for Healthcare Research and Quality criteria that imparts a deliberate separation between the strength of the evidence (i.e., high, moderate, low, or insufficient) from the strength of the recommendation. When appropriate, recommendations or statements “for” or “against” were given “strong” or “weak” designations based on the quality of the evidence, the balance of benefits/harms, values, and patient preferences. In some instances, costs may have been considered. A thorough description of this process can be found in the Methodology article in this focus issue by Skelly et al.22
The search strategy yielded 49 relevant citations, of which 32 were excluded on the basis of title and/or abstract. Seventeen were selected for full-text review, from which 7 were further excluded on the basis of outcomes not reported separately by treatment group, populations including diseases other than CSM, one study arm containing a majority of single-level CSM and one comparison containing fewer than 10 patients. The flow of this process is summarized in Figure 3. Ten studies 23–32 were selected for inclusion and are summarized in this work. Among these, 7 compared multiple discectomies to single or multiple corpectomy 23,24,26–28,30,31 (see Supplemental Digital Content available at http://links.lww.com/BRS/A824 for detailed demographic and results data). Three studies23,27,28 compared multiple discectomies to discectomy-corpectomy hybrid, and 6 studies compared multiple corpectomies to discectomy-corpectomy hybrid.23,25,27–29,32
We summarized the CE and safety among the following study groups: multiple discectomies versus single or multiple corpectomy; multiple discectomies versus discectomy-corpectomy hybrid; multiple corpectomies versus discectomy-corpectomy hybrid. Clinical effectiveness outcomes consistent across these studies included JOA scores, VAS neck pain scores, and NDI scores. The results by study on these outcomes are reported in summary tables and figures. Sagittal alignment was reported by several studies with such heterogeneous methodology that precluded a meaningful summary; neck range of motion was rarely reported and therefore not useful for summary; only one study reported the SF-36 quality-of-life outcome. Perioperative complications and sequelae that were consistently reported for summary included pseudoarthrosis, C5 palsy, infection, and dysphagia.
Multiple Discectomies Compared With Single or Multiple Corpectomy
Mean age was similar across both comparison groups.23,24,26–28,30,31 Multiple discectomies had a mean age range of 46.1 to 58.7 years, whereas the corpectomies had a mean age range of 47.8 to 58.0 years. Both groups were more frequently male, ranging from 50.5% to 76% and 51% to 73%, respectively. All studies contained patients with 3 operated levels. Three-level surgical procedures were most common followed by 2-level and 4-level procedures. One study24 contained one arm with a small percentage of 1-level surgery and therefore was included with the multilevel procedures. Follow-up times differed between studies. Five studies23,26–28,30 had follow-up times between 24 and 42 months and 2 studies24,31 between 57 and 94 months. The main limitation of these studies was that most studies did not report follow-up rates. Only one study24 reported a follow-up rate which was 75.4%.
Multiple Discectomies Compared With Discectomy-Corpectomy Hybrid
Mean age was similar across both comparison groups.23,27,28 Multiple discectomies had a mean age range of 46.1 to 53.5 years, whereas the discectomy-corpectomy hybrid had a mean age range of 46.9 to 54.4 years. Again, both groups were more frequently male, ranging from 55% to 57% and 60% to 66%, respectively. All studies consisted of only 3-level surgical procedures. Follow-up times differed between studies ranging from 25.6 to 42.0 months. The main limitation of these studies was follow-up rates not being reported.
Multiple Corpectomies Compared With Discectomy-Corpectomy Hybrid
Mean age was similar across both comparison groups.23,25,27–29,32 Corpectomies had a mean age range of 47.8 to 61.9 years, whereas the discectomy-corpectomy hybrid had a mean age range of 46.9 to 59.7 years. Again, both groups were more frequently male, ranging from 54% to 74% and 60% to 68%, respectively. Three-level surgical procedures were most common followed by 2-level and 4-level procedures. Follow-up times differed between studies. Three studies27,29,32 were in the range from 17.3 to 26.4 months and 3 studies23,25,28 were in the range from 31.5 to 37.3 months. The main limitation of these studies was that most studies did not report follow-up rates. Only 2 studies reported follow-up rates which were 89.4%32 and 95.0%.25
CE of Discectomy Versus Corpectomy Procedures for Multilevel CSM
When evaluating the change in NDI scores comparing discectomy to corpectomy among studies that reported both pre- and postoperative scores, all 3 studies26–28 reported greater improvement after discectomy than after corpectomy. Among these studies, the difference in change scores was statistically significant in 2 of them which resulted in SMDs of 2.126 and 1.428 both of which represent a “large effect” (Figure 4).
When evaluating the change in JOA scores comparing discectomy to corpectomy among studies that reported both pre- and postoperative scores, 4 of the 6 studies26,28,30,31 reported greater improvement after discectomy than after corpectomy, (Figure 4B). However, the difference in change scores was statistically significant in only one of these resulting in a SMD of 0.71,28 which represents a “medium effect.” One study27 reported a statistically significant improvement in change scores favoring the corpectomy group, which resulted in a SMD of 0.87 (“large effect”) (Figure 4B). One study reported the same mean change scores comparing the 2 surgical procedures.23
When evaluating the change in VAS scores comparing discectomy to corpectomy among studies that reported both pre- and postoperative scores, 1 of the 2 studies30 reported greater improvement after discectomy than after corpectomy (Figure 4C). The difference in change score was statistically significant resulting in a SMD of 2.4,30 which represents a “large effect.” In the other study, the difference in change scores was nearly identical31 favoring neither surgery (Figure 4C).
When evaluating the change in sagittal alignment comparing discectomy to corpectomy among studies that reported both pre- and postoperative scores, all 5 studies23,26,27,30,31 reported greater change after discectomy than after corpectomy. Among these studies, the difference in change scores was statistically significant in 4 of them, which resulted in SMDs of 2.4,26 1.9,27 1.3,30 and 2.5,23 all of which represent a “large effect” (Figure 5). The fifth study favored discectomy but the difference in change was not statistically significant.31
CE of Discectomy Versus Discectomy-Corpectomy Hybrid Procedures for Multilevel CSM
When evaluating the change in NDI scores comparing discectomy to discectomy-corpectomy hybrid among studies that reported both pre- and postoperative scores, both studies27,28 reported greater improvement after discectomy than after the hybrid procedure. Among these studies, the difference in change scores resulted in SMDs of 0.8627 and 0.25,28 representing a “large effect” and “small effect,” respectively (Figure 6A). The differences in change scores in the Liu 2012 study were statistically significant.
When evaluating the change in JOA scores comparing discectomy to discectomy-corpectomy hybrid among studies that reported both pre- and postoperative scores, 2 studies27,28 reported greater improvement after discectomy than after the hybrid procedure (Figure 6B). Among these studies, the difference in change scores was statistically significant and resulted in SMDs of 0.7527 and 1.4,28 representing a “medium effect” and “large effect,” respectively (Figure 6B). The third study reported the same mean change score between the 2 surgical treatments.
When evaluating the change in sagittal alignment comparing discectomy to hybrid surgery among studies that reported both pre- and postoperative scores, both studies23,27 reported greater change after discectomy than after hybrid surgery. The difference in change scores was statistically significant in both of them, which resulted in SMDs of 0.68,27 and 1.423 which represent a “medium effect” and “large effect,” respectively (Figure 7).
CE of Corpectomy Versus Discectomy-Corpectomy Hybrid Procedures for Multilevel CSM
When evaluating the change in NDI scores comparing corpectomy to discectomy-corpectomy hybrid among studies that reported pre- and postoperative scores, 2 studies28,29 reported greater improvement after the hybrid procedure than after corpectomy alone. Among these studies, the difference in change scores was statistically significant in both of them, which resulted in SMDs of 1.228 and 1.6,29 both of which represent a “large effect” (Figure 8A). One study reported greater improvement after corpectomy than after the hybrid surgery. The difference in change scores was statistically significant, which resulted in a SMD of 0.68, which represents a “medium effect.”
When evaluating the change in JOA scores comparing corpectomy to discectomy-corpectomy hybrid among studies that reported both pre- and postoperative scores, all 4 studies27–29,32 reported greater improvement after corpectomy than after the hybrid procedure (Figure 8). Among these studies, the difference in change scores in 2 studies27,28 were statistically significant and resulted in SMDs of 1.827 and 1.1,28 representing a “large effect” (Figure 8B). The differences in change scores in the other 2 studies29,32 were not statistically significant. The latter 2 studies were smaller, and the variable effects may reflect their small size underpowered to detect difference or the differential preoperative JOA scores for which different extents of improvement were possible.
When evaluating the change in sagittal alignment comparing corpectomy to hybrid surgery among studies that reported both pre- and postoperative scores, 2 of the 3 studies23,27 reported greater change after hybrid surgery than after corpectomy surgery. The difference in change scores was statistically significant in both of them which resulted in SMDs of 1.327 and 1.523 both of which represent a “large effect” (Figure 9). The third study demonstrated a “small effect” favoring corpectomy, but the result was not statistically significant.29
Comparative Safety of Anterior Procedures for Multilevel CSM
The following complications among the studies making 1 of the comparisons listed in the CE section for multilevel CSM were extracted from the 11 included studies (Table 2): pseudoarthrosis, C5 palsy, infection, and dysphagia. Three studies reported pseudoarthrosis rates.23,24,31 Among these, pseudoarthrosis rates were higher in the discectomy group than the corpectomy groups with rates of 33.6% vs. 6.8%24 (P = 0.0001), 2.3% vs. 0%23 (P = 1.0), and 12% vs. 6.7%31 (P = 0.59). Only one study reported this outcome after the hybrid procedure and reported no events.23
Four studies reported C5 palsy rates.23,27,28 In the study by Guo et al,23 no events were reported in any of the 3 surgical groups. In the study by Lin et al,26 the C5 palsy rate was 3.5% and 4.8% in the discectomy and corpectomy groups, respectively (P = 0.75). In the study by Liu et al,27 the discectomy and hybrid rates were 2.9% and 2.8%, respectively. The rate in the corpectomy group was 10.3%. These differences were approaching statistical significance (P = 0.08). In the study by Liu et al,28 the discectomy and hybrid rates were 3.9% and 8.3%, respectively. The rate in the corpectomy group was 11.5%. These differences between the discectomy group and the corpectomy group was statistically significant (P = 0.045). The other comparisons were not statistically significant.
Infection rates were very low in all groups. Among the 4 studies that reported this complication, there were no events in the discectomy groups26–28,31 (Table 2). Three studies reported a 0% rate25,26,31 and a rate of 2.6%27 and 1.2%28 in the corpectomy groups. Three studies reported a 1.4%,27 1%,28 and 0%25 in the hybrid group.
In the 2 studies comparing discectomy to corpectomy, dysphagia rates were 7% vs. 8% (P = 0.85) and 12% vs. 20%, respectively (P = 0.49. In 1 study that included all 3 comparisons,27 dysphagia rates were 11.6%, 10.3%, and 9.7% in the discectomy, corpectomy, and hybrid groups, respectively. In the other, dysphagia rates were 5.8%, 2.3%, and 5.2%, respectively. None of the comparisons in either study were statistically significant.
Other perioperative data were inconsistently reported among the various studies that formed the basis of this systematic review. Indeed, features such as blood loss, OR time, duration of in-hospital stay, early postoperative emergencies from graft malposition, neck or epidural hematomas, and further revision surgery are all important. They could not form a part of this analysis, but are included in the Supplemental Digital Content Table 3 available at http://links.lww.com/BRS/A824 summarized individually for each source article.
Ancillary Stabilization Device
The vast majority of patients undergoing multilevel anterior cervical procedures used some form of external immobilization afterward (Figure 10). Cervical collar or two postcervical orthosis was most frequently used, together accounting for 96% of multilevel discectomies and 96% of hybrid procedures, and 86% of multiple corpectomies. The difference with the latter was accounted for by greater utilization of more aggressive external stabilization in the form of a halo vest (12% of multiple corpectomies, none of the multiple discectomies or hybrid procedures). All authors reporting use of an ancillary device, except for one, applied the same stabilization device to all treatment procedures. The exception was Song et al31 who selectively applied the Philadelphia collar to discectomy patients and the halo vest to patients who underwent corpectomy.
Table 3 summarizes the overall strength of the body of literature surrounding the choice among various anterior decompressive and fusion procedures for CSM. Moderate means we have moderate confidence that the evidence reflects the true effect, and further research may change our confidence in the estimate of effect and may change the estimate. Low means we have low confidence that the evidence reflects the true effect, and further research is likely to change the confidence in the estimate of effect and likely to change the estimate, or that the evidence is either unavailable or does not permit a conclusion. Insufficient means the evidence is either unavailable or does not permit a conclusion.
The strength of evidence favoring discectomy compared with corpectomy is moderate evidence for NDI, insufficient evidence for JOA, low evidence for VAS, and moderate evidence for sagittal alignment. The strength of evidence favoring discectomy compared with discectomy-corpectomy hybrid is low evidence for NDI, moderate evidence for JOA, and moderate evidence for sagittal alignment. There is conflicting evidence about superiority of corpectomy or discectomy-corpectomy hybrid, with low evidence favoring discectomy-corpectomy hybrid for NDI, moderate evidence favoring corpectomy for JOA, and moderate evidence favoring discectomy-corpectomy hybrid for sagittal alignment. Regarding the incidence of perioperative complications, the strength of evidence for the superiority of any of the surgical approaches for pseudoarthrosis/nonunion, dysphagia, C5 palsy, and infection is “insufficient,” “low,” “low,” and “low,” respectively.
The optimal management of CSM due to diffuse or multifocal disease remains hitherto undefined. Historically, key factors that affect the choice between anterior and posterior surgical approaches include identity and location of the stenotic pathology, sagittal alignment of the cervical spine, and the number of diseased levels.10,33 Certainly, complication rates have been found to depend on selected surgical approach, both in all cervical procedures6,7,34,35 and specific for more extensive disease.5 In the latter group, Shamji et al5 reviewed the Nationwide Inpatient Sample database defining 8548 patients undergoing multilevel cervical fusion procedures (greater than 3 motion levels) between 2003 and 2005, finding in a multivariate logistic regression model that anterior surgery was associated with fewer respiratory complications, infections, symptomatic hematomas, and transfusions, as well less resource utilization for hospitalization costs and assisted-living discharge facilities when compared with propensity-matched patients treated with a posterior surgical approach. Similar findings were reported by Fehlings et al7 as a part of a rigorous, prospective observational study exploring the surgical treatment of CSM, with 302 enrolled patients exhibiting fewer overall complications and primarily wound infections among the anterior-only surgery group. The great value of this work followed into the future will be in demonstrating not only the perioperative experience of these patients while in hospital, but also the longer-term clinical stability or improvement in this group of patients. In a recent systematic review of treatment efficacy in both CSM and ossified posterior longitudinal ligament populations, Liu et al8 report short-term superiority in JOA scores (less than 5 years of follow-up) for anterior over posterior surgery, with equivalence observed in the long-term.
When selecting among the various anterior options, the surgeon's toolbox is replete with an armamentarium of options, ranging from multiple conventional discectomies in which segmental decompression is performed, corpectomies whereby retrovertebral disease can be directly addressed, and hybrid procedures. Certainly, the individual pathoanatomy may contribute to the decision as in the case of a substantial burden of retrovertebral disease motivating a corpectomy. Indeed, the incidence of construct failure may well be different among study groups, and being that these studies are retrospective, the patient selection will no doubt have been motivated by the patient pathoanatomy—those patients with retrovertebral disease more likely to have undergone multiple corpectomies or discectomy-corpectomy hybrid procedures. These various reconstructive options have altered biomechanical properties that may lead to disparate incidences of short-term and long-term procedural revision. Such concerns could not be addressed in this review because of inconsistent data reporting and highly variable follow-up from the source studies. However, in those cases where the surgeon has equipoise about the ideal decompressive procedure, the theoretical benefits of perioperative safety, long-term clinical improvement, and discordant rates of fusion and sagittal correction, are clearly important and can help contribute to the decision-making process. Indeed, these have been previously examined in small studies, but small sample sizes has precluded meaningful decisions being made about what provides for best clinical outcome and least perioperative complication. The results of this study help address the void of information to guide procedure selection in those circumstances. It is important to note that with ossified posterior longitudinal ligament being an exclusionary criteria in this review, such results may not be generalizable to that population of patients.
All 3 anterior options examined in this work provided patients with significant improvements in clinical outcomes of NDI, JOA, VAS, as well as improvements in radiographical parameters of sagittal alignment. All are reasonable approaches and echo the recent meta-analysis by Liu et al8 that anterior surgery for CSM can be efficacious in the appropriately selected patient.
The key objective of this work was to explore whether there was clear superiority of one of the anterior surgical options over the others, to help guide the surgeon in selecting among procedures when faced with the patient for whom the pathoanatomy permits such choice. Table 3 summarizes the strength of evidence comparing the effectiveness at achieving improvements in clinical outcomes, with the primary data listed by study and stratified by comparisons of interest in the Supplemental Digital Content available at http://links.lww.com/BRS/A824. Multiple discectomy is favored compared with corpectomy procedures, with moderate strength of evidence supporting that stance. Consistent benefits were observed for NDI and VAS score, with equivalence observed for JOA scores. Furthermore, all studies reported improved correction of sagittal alignment after the multiple discectomies approach. Multiple discectomy is favored over discectomy-corpectomy hybrid procedures, with moderate strength of evidence supporting that stance. Consistent benefits were observed for NDI and JOA scores. Furthermore, all studies reported improved correction of sagittal alignment after the multiple discectomies approach. There is conflicting information for selecting between multiple corpectomy and discectomy-corpectomy hybrid procedures. Benefits in JOA score favor the corpectomy procedure, whereas improvements in NDI score and correction of sagittal alignment favor the hybrid procedure. Neither procedure is thus considered to be superior.
Although it may be attractive to state that the more favorable improvements in NDI score among multiple discectomy procedures compared with those involving at least one corpectomy reflect less severe disease, the similar preoperative neurological status among the various groups would not support that claim. Rather, this group of patients experienced similar neurological improvement but superior sagittal correction and improvement in neck pain, all of which contribute to the NDI parameter. Indeed, the benefits of sagittal correction and improved cervical spine biomechanics will also confer long-term benefits in reduced incidence of adjacent segment disease.36–38
There was substantial heterogeneity among studies in baseline patient demographics as well as the consistency of perioperative complication reporting, however all 3 surgical approaches are associated with low complication rates in both short term and long term. We chose to tabulate this data and present it stratified by procedural choice, shown in Table 2, with all procedures appearing equivalent for nonunion, dysphagia, and infection. Nevertheless, to help guide surgeons with patient counseling, we can also discuss pooled estimates of perioperative morbidity. With regard to nonunion, pooled estimates for discectomy, corpectomy, and hybrid procedures are 11%, 6%, and 4%, respectively, although the dominant contributor to nonunion in the multiple discectomy group is the work by Hilibrand et al24 with a 33% nonunion rate, long retrospective, and consequently without augmentation of extensive anterior procedures with plate instrumentation. Modern surgical techniques have certainly improved those nonunion rates, and exclusion of that one study puts the nonunion rate for multiple discectomies to 1.3%, more in line with the other more recent studies in that group. With regard to infection, pooled estimates for discectomy, corpectomy, and hybrid procedures are 0%, 1%, and 1%, respectively. With regard to dysphagia, pooled estimates for discectomy, corpectomy, and hybrid procedures are 8%, 7%, and 7%, respectively. With low evidence but large effect, multiple discectomies is favored as an approach over both corpectomy and discectomy-corpectomy hybrid for lower incidence of C5 palsy. Pooled estimates for discectomy, corpectomy, and hybrid procedures are 3%, 8%, and 5%, respectively. Such estimates of complications associated with the various multilevel cervical anterior reconstructve options is important for the surgeon to select the ideal surgical approach and to counsel patients prior to performing the procedure.
A further limitation of this work is the heterogeneity in use of reconstructive maneuvers that ranged among studies from synthetic cages to allograft to cage packed with autograft to structural autograft, and included both anterior cervical plating and no such implant as stabilization maneuvers. The efficacy of anterior cervical procedures at achieving adequate decompression and neurological recovery should remain unaffected by that choice of devices, although these differences may be reflected in longer-term revision rates and need for secondary operations.
Evidence-Based Clinical Recommendations.
Recommendation 1. When pathoanatomically appropriate with minimal retrovertebral disease, we recommend the selection of multiple discectomy over corpectomy or discectomy-corpectomy hybrid procedures.
Overall Strength of Evidence. Low
Strength of Recommendation. Strong
Recommendation 2: When retrovertebral disease is significant, we recommend, when possible, that discectomy-corpectomy hybrid procedures be performed over multiple corpectomies.
Overall Strength of Evidence. Moderate
Strength of Recommendation. Strong
Summary Statements. There is no evidence to guide choice of ancillary external immobilization techniques following multilevel anterior decompression and fusion procedures for CSM.
- When the pathoanatomy permits, multiple discectomies is favored compared with corpectomy and discectomy-corpectomy hybrid procedures for clinical outcomes and for achieving superior sagittal correction.
- There is conflicting evidence for the superiority of corpectomy or discectomy-corpectomy hybrid procedures with regard to clinical outcomes and sagittal alignment. Neither of the 2 is recommended as superior.
- The safety of the various anterior procedures seems similar for nonunion, dysphagia, and infection. Multiple discectomies, when appropriate, may have a lesser incidence of C5 palsy than corpectomy or discectomy-corpectomy hybrid procedures.
Author contributions are as follows: M.F.S: Study design, data analysis and interpretation, manuscript preparation, and manuscript revision; E.M.M: Study design, data analysis and interpretation, and manuscript revision; V.T: Study design, data analysis and interpretation, manuscript preparation, and manuscript revision; D.C.N: Study design, data analysis and interpretation, manuscript preparation, and manuscript revision; J.T.H: Study design, data analysis and interpretation, manuscript preparation, and manuscript revision; M.G.F: Study design, data analysis and interpretation, and manuscript revision.
Supplemental digital content is available for this article. Direct URL citations appearing in the printed text are provided in the HTML and PDF version of this article on the journal's web site (www.spinejournal.com).
1. Cowan JA Jr, Dimick JB, Wainess R, et al. Changes in the utilization of spinal fusion in the United States. Neurosurgery 2006;59:15–20; discussion 15–20.
2. Patil PG, Turner DA, Pietrobon R. National trends in surgical procedures for degenerative cervical spine disease: 1990–2000. Neurosurgery 2005;57:753–8; discussion 8.
3. Fehlings MG, Jha NK, Hewson SM, et al. Is surgery for cervical spondylotic myelopathy cost-effective? A cost-utility analysis based on data from the AOSpine North America prospective CSM study. J Neurosurg Spine 2012;17:89–93.
4. Fehlings MG, Arvin B. Surgical management of cervical degenerative disease: the evidence related to indications, impact, and outcome. J Neurosurg Spine 2009;11:97–100.
5. Shamji MF, Cook C, Pietrobon R, et al. Impact of surgical approach on complications and resource utilization of cervical spine fusion: a nationwide perspective to the surgical treatment of diffuse cervical spondylosis. Spine J 2009;9:31–8.
6. Shamji MF, Cook C, Tackett S, et al. Impact of preoperative neurological status on perioperative morbidity associated with anterior and posterior cervical fusion. J Neurosurg Spine 2008;9:10–6.
7. Fehlings MG, Smith JS, Kopjar B, et al. Perioperative and delayed complications associated with the surgical treatment of cervical spondylotic myelopathy based on 302 patients from the AOSpine North America Cervical Spondylotic Myelopathy Study. J Neurosurg Spine 2012;16:425–32.
8. Liu T, Xu W, Cheng T, et al. Anterior versus
posterior surgery for multilevel cervical myelopathy, which one is better? A systematic review. Eur Spine J 2011;20:224–35.
9. Hirabayashi K, Bohlman HH. Multilevel cervical spondylosis. Laminoplasty versus
anterior decompression. Spine 1995;20:1732–4.
10. Naderi S, Benzel EC, Baldwin NG. Cervical spondylotic myelopathy: surgical decision making. Neurosurg Focus 1996;1:e1.
11. Fehlings MG, Gray R. Importance of sagittal balance in determining the outcome of anterior versus
posterior surgery for cervical spondylotic myelopathy. J Neurosurg Spine 2009;11:518–9; discussion 9–20.
12. Bolesta MJ, Rechtine GR 2nd, Chrin AM. Three- and four-level anterior cervical discectomy and fusion with plate fixation: a prospective study. Spine 2000;25:2040–4; discussion 5–6.
13. Emery SE, Fisher JR, Bohlman HH. Three-level anterior cervical discectomy and fusion: radiographic and clinical results. Spine 1997;22:2622–4; discussion 5.
14. Papadopoulos EC, Huang RC, Girardi FP, et al. Three-level anterior cervical discectomy and fusion with plate fixation: radiographic and clinical results. Spine 2006;31:897–902.
15. Wright JG, Swiontkowski MF, Heckman JD. Introducing levels of evidence to the journal. J Bone Jt Surg Am 2003;85-A:1–3.
16. Norvell DC, Dettori JR, Fehlings MG, et al. Methodology for the systematic reviews on an evidence-based approach for the management of chronic low back pain. Spine 2011;36:S10–8.
17. Atkins D, Best D, Briss PA, et al. Grading quality of evidence and strength of recommendations. BMJ 2004;328:1490.
18. West S, King V, Carey TS, et al. Systems to rate the strength of scientific evidence. Evid Rep Technol Assess (Summ) 2002;47:1–11.
19. Higgins JP, Green S eds. Cochrane Handbook for Systematic Reviews of Interventions. The Cochrane Collaboration 2011; version 5.1.0.
20. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Lawrence Erlbaum Associates; 1988.
21. Stata Statistical Software. Version 9.1. College Station, TX: Stata Corporation LP; 2005.
22. Skelly AC, Hashimoto RE, Norvell DC, et al. Cervical Spondylotic Myelopathy: Methodological Approaches to Evaluate the Literature and Establish Best Evidence. Spine 2013;38:S9–18.
23. Guo Q, Bi X, Ni B, et al. Outcomes of three anterior decompression and fusion techniques in the treatment of three-level cervical spondylosis. Eur Spine J 2011;20:1539–44.
24. Hilibrand AS, Fye MA, Emery SE, et al. Increased rate of arthrodesis with strut grafting after multilevel anterior cervical decompression. Spine 2002;27:146–51.
25. Lian XF, Xu JG, Zeng BF, et al. Noncontiguous anterior decompression and fusion for multilevel cervical spondylotic myelopathy: a prospective randomized control clinical study. Eur Spine J 2010;19:713–9.
26. Lin Q, Zhou X, Wang X, et al. A comparison of anterior cervical discectomy and corpectomy in patients with multilevel cervical spondylotic myelopathy. Eur Spine J 2012;21:474–81.
27. Liu Y, Hou Y, Yang L, et al. Comparison of three reconstructive techniques in the surgical management of multilevel cervical spondylotic myelopathy. Spine 2012;3:3.
28. Liu Y, Qi M, Chen H, et al. Comparative analysis of complications of different reconstructive techniques following anterior decompression for multilevel cervical spondylotic myelopathy. Eur Spine J 2012;29:29.
29. Liu Y, Yu KY, Hu JH. Hybrid decompression technique and two-level corpectomy are effective treatments for three-level cervical spondylotic myelopathy. J Zhejiang Univ Sci B 2009;10:696–701.
30. Oh MC, Zhang HY, Park JY, et al. Two-level anterior cervical discectomy versus
one-level corpectomy in cervical spondylotic myelopathy. Spine 2009;34:692–6.
31. Song KJ, Lee KB, Song JH. Efficacy of multilevel anterior cervical discectomy and fusion versus
corpectomy and fusion for multilevel cervical spondylotic myelopathy: a minimum 5-year follow-up study. Eur Spine J 2012;21:1551–7.
32. Wei-bing X, Wun-Jer S, Gang L, et al. Reconstructive techniques study after anterior decompression of multilevel cervical spondylotic myelopathy. J Spinal Disord Tech 2009;22:511–5.
33. Rao RD, Currier BL, Albert TJ, et al. Degenerative cervical spondylosis: clinical syndromes, pathogenesis, and management. J Bone Jt Surg 2007;89:1360–78.
34. Romano PS, Campa DR, Rainwater JA. Elective cervical discectomy in California: postoperative in-hospital complications and their risk factors. Spine 1997;22:2677–92.
35. Wang MC, Chan L, Maiman DJ, et al. Complications and mortality associated with cervical spine surgery for degenerative disease in the United States. Spine 2007;32:342–7.
36. Hansen MA, Kim HJ, Van Alstyne EM, et al. Does postsurgical cervical deformity affect the risk of cervical adjacent segment pathology? A systematic review. Spine 2012;37:S75–84.
37. Lafage V, Schwab FJ, Skalli W, et al. Standing balance and sagittal plane spinal deformity: analysis of spinopelvic and gravity line parameters. Spine 2008;33:1572–8.
38. Shimizu K, Nakamura M, Nishikawa Y, et al. Spinal kyphosis causes demyelination and neuronal loss in the spinal cord: a new model of kyphotic deformity using juvenile Japanese small game fowls. Spine 2005;30:2388–92.