Background: Ossification of the posterior longitudinal ligament (OPLL) is defined as ectopic bone formation within the posterior longitudinal ligament. Although various OPLL features (including the extent, shape, and thickness of the OPLL as well as the presence of dural ossification) have been defined in the literature, we are not aware of any systematic reviews that have summarized the associations between these features and clinical outcomes following surgery. The objective of the present study was to conduct a systematic review of the literature to determine whether OPLL characteristics are predictive of outcome in patients undergoing surgery for the treatment of cervical myelopathy.
Methods: An extensive search was performed using 4 electronic databases: MEDLINE, MEDLINE In-Process, Embase, and the Cochrane Central Register of Controlled Trials. Our search terms were OPLL and cervical. We identified studies in English or Japanese that evaluated the association between cervical OPLL features and surgical outcome. The overall body of evidence was assessed with use of a scoring system developed by the Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group with recommendations from the Agency for Healthcare Research and Quality (AHQR). The present systematic literature review is formatted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.
Results: The search yielded a total of 2,318 citations. A total of 28 prognostic cohort studies were deemed relevant following a rigorous review process. Among them, only 7 retrospective studies involved a multivariate analysis that controlled for potential confounding variables. Sample sizes ranged from 47 to 133 patients. The main outcome was the postoperative Japanese Orthopaedic Association (JOA) score and/or recovery rate in 6 studies and the Nurick grade in 1. Of these, 2 were rated as Level-II evidence and 5 were rated as Level-III evidence. On the basis of our results, there was low evidence that patients with a hill-shaped ossification have a worse postoperative JOA score following laminoplasty than those with a plateau-shaped lesion; low evidence that the space available for the spinal cord cannot predict postoperative JOA scores; moderate evidence that there is no association between the occupying ratio and improvement on the Nurick scale; and insufficient evidence to determine the association between JOA outcomes and the type of OPLL, the presence of dural ossification, and the occupying ratio.
Conclusions: Patients with hill-shaped OPLL have a worse postoperative JOA score than those with plateau-shaped ossification after laminoplasty. Because of limited evidence, it is unclear whether the occupying ratio, the type of OPLL, and the presence of dural ossification are predictive of surgical outcomes following either anterior or posterior decompression. A limited number of studies have used a multivariate analysis to evaluate the association between clinical outcomes and OPLL features. Additional studies representing high-quality evidence are needed.
Level of Evidence: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.
Ossification of the posterior longitudinal ligament (OPLL) is a cause of degenerative cervical myelopathy1 and is characterized by ectopic bone formation within the posterior longitudinal ligament2,3. OPLL can result in foraminal and spinal canal narrowing, can cause compression of the spinal cord and/or nerve roots, and can manifest as pain and/or neurological dysfunction3,4. Surgical decompression is recommended in cases characterized by progressive and/or severe myelopathy5.
Various classification systems have been developed to describe the morphology, distribution, and thickness of ossification as well as the configuration of the compressed spinal cord as seen on computed tomography (CT) scans, radiographs, or magnetic resonance imaging (MRI) scans. The Investigation Committee for Ossification of the Spinal Ligaments (part of the Japanese Ministry of Health, Labour and Welfare [JMHLW]) established a commonly used 4-type classification system for cervical ossification: segmental, continuous, mixed, and circumscribed6. OPLL also may be categorized as either plateau-shaped or hill-shaped on the basis of its morphology on a lateral radiograph7. Plateau-shaped ossification is defined by a relatively narrow spinal canal without any localized massive ossification, whereas hill-shaped ossification appears as a massive, localized, beak-shaped lesion. In 2008, Fujiyoshi et al. introduced the K-line, which is drawn from the midpoints of the spinal canal at C2 and C7 on neutral lateral radiographs (Fig. 1)8. A patient is classified as K-line (−) if the ossification extends beyond the K-line and as K-line (+) if it does not. In addition, OPLL may be categorized as boomerang, teardrop, or triangular on the basis of the configuration of the compressed spinal cord on a T2-weighted axial MRI scan9. In the boomerang type, the spinal cord has a convex posterior surface and a concave anterior surface with a smooth round corner. In the teardrop type, it has a convex posterior surface and concave anterior surface with only a smooth round corner on 1 side. In the triangular type, it has an angular lateral surface and a flat anterior surface. Finally, OPLL may be classified according to whether there is concomitant dura mater ossification10,11.
Several other OPLL features have been described in the literature, including the occupying ratio, the number of ossified levels, and the space available for the spinal cord (SAC). These factors help to assess the degree and severity of cord compression and the extent of ossification.
Given the heterogeneity of OPLL, patients may present with different neurological symptoms, risks of disease progression, and complication profiles and may require different treatment strategies. Furthermore, surgical outcomes may vary depending on the type, extent, and occupying ratio of the ossification. However, to our knowledge, there have been no systematic reviews of the literature that have summarized the association between OPLL features and surgical outcomes. The purpose of the present study was to conduct a systematic review to determine whether OPLL characteristics are predictive of functional or quality-of-life outcomes in patients managed surgically for cervical myelopathy.
Materials and Methods
We sought studies involving adult patients (≥18 years of age) who had been (1) diagnosed with cervical myelopathy resulting from OPLL, (2) managed surgically, and (3) followed postoperatively. Studies involving patients with other forms of degenerative cervical myelopathy (e.g., spondylosis, hypertrophy of the ligamentum flavum or disc degeneration without OPLL), tumor, trauma, or infection were excluded. Studies involving patients undergoing revision surgery were also excluded (Table I).
We targeted studies that evaluated the association between OPLL characteristics and surgical outcome. OPLL factors of interest included occupying ratio, thickness, morphology (e.g., hill-shaped, plateau-shaped, or mushroom-shaped), type (e.g., segmental, continuous, mixed, or localized), and the extent of ossification. Studies were excluded if they only evaluated the predictive value of clinical factors (e.g., age, duration of symptoms, sex, or baseline severity score), non-OPLL-related imaging findings (e.g., signal changes on T1-weighted or T2-weighted MRI scans, cervical alignment, or range of motion), or surgical factors (e.g., approach, technique, number of operative segments, or operative duration) (Table I).
Outcomes of interest included functional impairment (Japanese Orthopaedic Association [JOA] score, modified Japanese Orthopaedic Association [mJOA] score, Nurick classification, and 30-m timed walking test), disability (Neck Disability Index), and health-related quality of life (Short-Form-36 [SF-36] score)12-15. Studies were excluded if the primary outcome was return to work, patient satisfaction, progression of OPLL, surgical complications, heterotopic ossification, radiographic assessment, or improvement of neurological signs and symptoms (Table I).
Four electronic databases were searched for studies published until October 5, 2015: MEDLINE, MEDLINE In-Process, Embase, and the Cochrane Central Register of Controlled Trials.
The following search terms were used to search all databases: OPLL and cervical. Only studies on humans were considered for inclusion.
All titles and abstracts were independently evaluated by 2 of the authors (L.T. and M.T.K.). The abstracts were categorized as relevant, possibly relevant, or irrelevant on the basis of our predefined inclusion criteria by 5 of the authors (H.N., L.T., S.K., N.N., and A.N.). Any study that included patients with OPLL was further investigated to determine whether the authors examined the association between OPLL features and surgical outcomes. Case reports were excluded, as were case series or cohort studies that included <15 patients. Systematic reviews, opinions, editorials, and commentaries were also excluded.
Data Extraction and Synthesis
We extracted the following data from each included article: study design (prospective or retrospective), sample size, patient characteristics, type of surgery, predictive factors evaluated, outcome assessment tools, and relevant statistics, including odds ratios (ORs), confidence intervals, and p values.
Risk of Bias of Individual Studies
Each article was independently rated as Level I, II, III, or IV by 2 reviewers according to the criteria for prognostic studies as summarized by Skelly et al.16. These criteria incorporate components from rating schemes proposed by the Oxford Centre for Evidence-Based Medicine and The Journal of Bone & Joint Surgery16 as well as the recommendations by the Agency for Healthcare Research and Quality (AHRQ) on methodological quality and risk of bias. This rating system evaluates study design (prospective or retrospective), whether patients were at a similar time point in their disease and treatment, the rate of follow-up (>80%), and whether the analysis adjusted for confounders.
Risk of Bias Across Studies
The strength of the overall body of evidence was graded independently by 2 reviewers with use of a scoring system developed by the Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group with recommendations from the AHRQ17-19.
The initial strength of the overall body of evidence was classified as high if the majority of studies were Level I or II and as low if the majority of studies were Level III or IV20. Reasons for downgrading the strength of evidence included (1) inconsistent results across studies, (2) indirect evidence, (3) imprecise estimates (i.e., wide confidence intervals), or (4) subgroup analyses that were not specified a priori. If there was no substantial risk of bias, the body of evidence could be upgraded 1 or 2 levels if there was a large magnitude of effect, if there was a dose-response gradient, or if all plausible confounding was explained. The final strength of the body of evidence expresses the confidence in the effect estimate and the impact that further research may have on this confidence. An overall strength of high indicates high confidence that the estimate reflects the true effect; further research is very unlikely to change the confidence in the estimate. An overall strength of moderate reflects moderate confidence that the estimate reflects the true effect; further research may change the confidence in the estimate of effect as well as the estimate itself. A grade of low indicates low confidence that the estimate reflects the true effect; further research is likely to change the confidence in the estimate as well as the estimate itself. Finally, a grade of insufficient means that there is no available evidence or that no conclusions can be made based on the existing evidence.
The present review was formatted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.
The search resulted in a total of 2,318 articles from MEDLINE (n = 940), MEDLINE In-Process (n = 90), Embase (n = 1,270), and the Cochrane Central Register of Controlled Trials (n = 18). Of those, 702 duplicates were removed and 845 studies did not meet our inclusion criteria after an initial review of the abstracts. An additional 544 studies were excluded following full-text evaluation because (1) they included patients who did not have OPLL, (2) they did not evaluate important OPLL predictors of outcome, (3) they did not evaluate outcomes using a validated assessment tool, and/or (4) they were systematic reviews or case reports. Furthermore, 199 studies could not be located through the libraries at the University of Toronto, University Health Network, or University of Western Ontario, mostly because they were not in English or Japanese. Seventy-five studies in Japanese were obtained from the library at Nagoya University Graduate School of Medicine; however, none of those studies met our inclusion criteria. In total, 28 studies were relevant and are summarized in the present review.
We found 28 prognostic cohort studies that evaluated the association between OPLL features and surgical outcomes7-9,11,21-44. Of these, only 7 retrospective studies involved a multivariate analysis that adjusted for confounders (n = 47 to 133)7,21-26. The mean ages ranged from 54.7 to 58.0 years across those studies. The main outcome was the postoperative JOA score and/or recovery rate ([postoperative JOA score – preoperative JOA score] × 100/[17 – preoperative JOA score]) in 6 studies and the Nurick grade in 1 (Table II). The average duration of follow-up ranged from 16 months to 10.2 years. The predictive values of several OPLL factors were examined, including the occupying ratio; the type, shape, and distribution of ossification; the configuration of the spinal cord; the SAC; the K-line classification (+/−); dural ossification; and the number of affected levels (Table III). Figure 1 illustrates how to evaluate the occupying ratio, SAC, and K-line status on axial and lateral images. Figure 2 displays different shapes of ossification.
In addition, 21 articles described OPLL predictors of outcome but did not include a multivariate analysis8,9,11,27-44. The primary outcome measures used in those studies included ranges of JOA recovery rates (n = 7), the JOA or mJOA recovery rate (n = 14), the postoperative JOA score (n = 8), and improvement in the Nurick grade (n = 2).
Risk of Bias
We critically appraised the 7 studies that involved a multivariate analysis. Of those, 2 were rated as Level II and 5 were rated as Level III. The 2 Level-II studies met all criteria required to be Level I but were retrospective in design. The Level-III studies had a moderately high risk of bias as all were retrospective cohort studies that violated ≥1 of the criteria required for a good-quality (Level-II) cohort study.
Results of Studies Involving Multivariate Analysis
Table IV highlights the results from the 7 studies that involved a multivariate analysis.
Postoperative JOA Score
Five studies evaluated the association between preoperative OPLL factors and postoperative JOA score7,21,23,25,26. In the study by Iwasaki et al., the most significant predictor of both maximum and final JOA score was the morphology of the ossification7. Specifically, patients with a hill-shaped lesion had a lower postoperative JOA score than those with a plateau-shaped lesion (11.7 compared with 14.1; p < 0.01). A single study by Uchida et al. indicated that an occupying ratio of <40% was correlated with improved postoperative JOA scores (Pearson correlation coefficient, 0.881)26. In contrast, 3 studies indicated that the occupying ratio was not an important predictor of outcome7,21,23. Furthermore, the postoperative JOA score was not associated with the presence of dural ossification or the SAC7,21,23.
Ogawa et al. evaluated differences in postoperative JOA scores, upper-extremity and trunk scores, and lower-extremity and bladder scores between patients with segmental and nonsegmental OPLL25. At the time of the latest follow-up (>7 years), patients with segmental OPLL had significantly lower median JOA, upper-extremity, and trunk scores (p < 0.05) as well as lower change scores (p < 0.05) than patients with nonsegmental OPLL. In contrast, there was no difference in median or change in lower-extremity and bladder scores between groups.
A single study assessed whether the shape of the ossification lesion, the type of OPLL, and the occupying ratio are important predictors of the JOA recovery rate24. There was no significant difference in recovery rate between patients with hill-shaped and plateau-shaped OPLL (mean and standard deviation, 74.05% ± 41.06% compared with 70.62% ± 34.52%; p = 0.751). Patients with segmental OPLL (51.79% ± 44.21%) had a worse recovery rate than those with continuous (87.12% ± 30.13%), mixed (80.01% ± 24.75%), or localized (69.07% ± 43.55%) ossification, although this relationship did not reach significance in univariate analysis (p = 0.154). Occupying ratio was the only significant predictor of the recovery rate following multiple linear regression analysis (p = 0.041).
A single study evaluated important predictors of outcome using the Nurick scale22. A “good” outcome was defined as an increase of at least 1 grade, whereas a “poor” outcome was either no change or a decline in function. There were no differences in the type of OPLL (segmental, mixed, continuous, or localized) or in the frequency of the double-layer sign between patients who achieved a good outcome and those who did not. Occupying ratio was the only OPLL factor evaluated in multiple logistic regression analysis; this factor was not predictive of outcome (OR, 0.973; 95% confidence interval, 0.85 to 1.07; p = 0.5356).
Results of Studies Not Involving Multivariate Analysis
Table V summarizes the results of studies that evaluated important OPLL predictors of outcome but did not include a multivariate analysis.
Seven studies examined the association between preoperative OPLL factors and the postoperative JOA score8,9,27-29,42,44. Outcome was not predicted by the type of OPLL, the occupying ratio, the thickness of ossification, or the SAC27,28. In the study by Iwasaki et al., patients with hill-shaped lesions had significantly lower JOA scores following laminoplasty than those with plateau-shaped lesions (11.7 compared with 14.1; p < 0.01)44. These findings could not be replicated in patients undergoing anterior decompression with fusion. Fujiyoshi et al. evaluated the predictive value of the K-line classification, a parameter that describes both the size of the OPLL lesion and the alignment of the cervical spine8. The results indicated that patients with K-line (+) OPLL had significantly higher postoperative JOA scores than those with K-line (−) ossification (13.8 ± 0.6 compared with 8.9 ± 0.8; p < 0.01). In contrast, Nagashima et al. reported similar postoperative scores between a K-line (+) group and a K-line (−) group29. However, in their sample, K-line (+) patients had a significantly lower preoperative score than K-line (−) patients (p = 0.04) and therefore exhibited larger improvements in JOA scores. The configuration of the compressed spinal cord was also an important predictor of postoperative JOA scores: triangular-shaped compression resulted in worse outcomes (11.6) than teardrop (14.2) or boomerang (14.2) configurations9. Finally, dural ossification was associated with lower postoperative JOA scores42.
Six studies that evaluated important predictors of postoperative JOA scores also evaluated the recovery rate8,9,27,29,42,44. The majority of results did not depend on whether the postoperative JOA score or the recovery rate was used as the outcome measure. However, in the study by Chen et al., although dural ossification was associated with postoperative JOA scores, it was not predictive of the recovery rate42. Seven additional studies evaluated the relationship between OPLL characteristics and the JOA recovery rate11,30-35. In studies by Baba et al., the occupying ratio was associated with the JOA recovery rate in patients undergoing laminoplasty but not in those managed with an anterior decompression with fusion30-32. Furthermore, Choi et al. reported that in patients who were managed posteriorly (laminoplasty or laminectomy with fusion), both the occupying ratio (p = 0.017) and the SAC (p = 0.028) were significant predictors of the recovery rate33. In contrast, the JOA recovery rate was not associated with the type of OPLL or classifications of dural ossification11,27,33,42.
In several studies, samples were divided on the basis of JOA recovery rates as follows: (a) ≥50% or <50%; (b) ≥40% or <40%; (c) ≥75%, 75% to ≥50%, 50% to ≥25%, or <25%; and (d) >75%, 75% to >50%, 50% to >0%, or ≤0%11,37-39,41-43.There were no differences across these groups with respect to the frequency or type of dural ossification, OPLL thickness, type of OPLL, the shape of the lesion, the extent of ossification, the SAC, or the K-line status (+/−). Furthermore, in the majority of studies, the occupying ratio was not significantly predictive of the JOA recovery rate. However, Fujimori et al. reported a significantly higher occupying ratio in patients with a poor recovery rate (<50%) (OR = 71.8% ± 6.2%) than in patients with an excellent recovery rate (≥50%) (OR = 63.2% ± 2.5%, p = 0.03) following anterior decompression and fusion41. This relationship was not significant in patients treated with laminoplasty.
Hossam et al. reported that the type of OPLL was not predictive of the mJOA recovery rate or improvement in the Nurick grade36. The study by Kommu et al. indicated that there was no difference in the number of ossified levels between patients who improved by ≥1 Nurick grade and those who were either stable or deteriorated following surgery40.
Table VI summarizes how the quality of evidence was rated. Low evidence suggested that patients with a hill-shaped lesion have a worse postoperative JOA score after laminoplasty than those with a plateau-shaped lesion. Low evidence suggested that the SAC cannot predict postoperative JOA scores. Moderate evidence suggested that there is no association between the occupying ratio and improvement on the Nurick scale. There was insufficient evidence to determine the relationship between JOA outcomes and the type of OPLL according to the JMHLW classification system, the presence of dural ossification, and the occupying ratio after anterior or posterior decompression of the spinal cord.
To our knowledge, the present study is the first systematic review to summarize the association between the features of cervical OPLL and surgical outcomes. Our results indicate that patients with hill-shaped lesions have worse postoperative JOA scores following laminoplasty than those with plateau-shaped lesions. However, there is insufficient evidence to evaluate the association between surgical outcomes and the type of OPLL according to the JMHLW classification system, the presence of dural ossification, and the occupying ratio.
Several systematic reviews have been conducted to evaluate important clinical and imaging predictors of outcomes in patients undergoing surgery for the treatment of degenerative cervical myelopathy20,45. On the basis of those reviews, patients have worse surgical outcomes if they have more severe degenerative pathology and myelopathy, reduced physiological reserves, poorer general health status, and imaging evidence of irreversible histological damage. In the present review, we hypothesized that surgical outcomes would be associated with OPLL features that cause more severe cord compression, require more challenging surgical procedures, and are associated with a higher risk of complications or disease progression.
The severity and degree of spinal cord compression may vary among subtypes of OPLL. Mixed and continuous types of OPLL have been associated with a higher occupying ratio and result in a greater percentage of spinal canal stenosis, a smaller SAC, and an increased compression ratio compared with the values for patients with segmental or localized ossification46-48. In contrast, the study by Fujiyoshi et al. showed no significant difference in the SAC or occupying ratio among patients with mixed, segmental, or continuous OPLL49. Furthermore, Koyanagi et al. concluded that patients with segmental ossification had significantly narrower diameters at C3-C7 than patients with continuous or mixed OPLL50. On the basis of these findings, it is evident that the subtype of OPLL is not predictive of the degree of cord compression and that there may be substantial variability with respect to severity in patients with the same type of OPLL. As a result of this heterogeneity, the type of OPLL according to the JMHLW classification system6 is likely not predictive of surgical outcomes. This hypothesis was confirmed by our systematic review.
On the other hand, the morphology of OPLL may be better associated with the severity of cord compression and may have an improved predictive value. In the study by Iwasaki et al., patients with hill-shaped OPLL had a higher mean occupying ratio (50.8% compared with 43.0%) and a lower SAC (6.9 compared with 7.7 mm) than patients with plateau-shaped OPLL, although these relationships did not reach significance7. Hill-shaped OPLL was predictive of both the postoperative JOA score and the recovery rate, likely because it results in more severe compression and sharper angulation of the spinal cord than plateau-shaped ossification.
The occupying ratio, the SAC, and the extent of ossification also may be representative of the degree of spinal cord compression. On the basis of the current body of literature, there is insufficient evidence to conclude whether the occupying ratio or the number of ossified levels are important predictors of the postoperative JOA or the recovery rate. There currently exists only low-level evidence that the SAC is not predictive of surgical outcomes; however, our confidence that the estimate reflects the true effect is low and is likely to change with further research.
These results may vary depending on the type of surgical approach. For example, a high occupying ratio frequently has been reported to be a significant predictor of poor surgical outcomes following laminoplasty7,41,51,52. Iwasaki et al. evaluated surgical outcomes for patients with a high (≥60%) and low (<60%) occupying ratio and reported worse postoperative JOA scores and recovery rates in the high-ratio group7. Following multivariate analysis, however, the occupying ratio was not a significant predictor of surgical outcomes. Other analyses have used either a continuous variable or a different cutoff for occupying ratio (including ≥50% versus <50%21 and ≥40% versus <40%26); some demonstrated a significant relationship with outcome31-33, whereas others failed to detect an association27,28,30,35,37,38,43. Several studies have also compared the outcomes of anterior and posterior surgery in patients with varying occupying ratios; the results of those studies demonstrated that anterior surgery resulted in superior surgical outcomes in patients with higher occupying ratios41,51. However, in some cases, a higher occupying ratio resulted in poor outcomes following anterior surgery because of severe spinal cord damage. Further investigation is required to assess whether the occupying ratio is a significant predictor of outcome and to determine the ideal surgical strategy for patients with higher occupying ratios.
Dural ossification may be an important predictor of outcome, especially following anterior surgery, as it is technically challenging to separate the posterior longitudinal ligament from the ossified dura. As a result, dural ossification may be associated with an increased risk of cerebrospinal fluid leakage and other secondary complications such as meningitis, delayed wound-healing, airway obstruction, and pseudomeningocele. Furthermore, patients with dural ossification also may have a higher occupying ratio and a greater number of ossified levels than those without dural ossification42. Despite these findings, the presence or type of dural ossification were not predictive of surgical outcomes. This lack of association may be due to the misdiagnosis of dural ossification preoperatively as it is difficult to identify using lateral radiographs or MRI scans10. Furthermore, although the presence of a double-layer sign on a CT scan helps to diagnose dural ossification, the sensitivity of this modality is still low (55%). In future prospective studies, the presence of dural ossification should first be confirmed intraoperatively and then correlated with outcomes.
Strengths and Limitations of the Present Review
The strengths of the present review are that (1) it is the first systematic review, to our knowledge, to evaluate the association between surgical outcome and OPLL features; (2) it was formatted according to the PRISMA statement; (3) we reviewed a total of 2,318 citations with keywords OPLL and cervical, including those in Japanese; and (4) we rated the strength of the body of evidence with use of the GRADE system. The limitations of this review were that (1) we could not extensively review papers in languages other than English and Japanese and based our exclusions on the abstracts of 59 (2.55%) of the studies; (2) the critical appraisal process requires scientific judgment and therefore there is potential for bias; and (3) articles with dichotomized predictors might have used different cutoff values (e.g., for the occupying ratio).
The limitations of the current body of evidence are that (1) few studies have involved multivariate analysis and controlled for relevant confounders; (2) outcome measures have differed across studies, preventing data synthesis and meta-analysis; and (3) the duration of follow-up has varied across studies and, in a minority of cases, may have been too short to evaluate the association between OPLL features and long-term outcomes. Although only 7 studies were included in our assessment of the overall body of evidence, we summarized results from other analyses that evaluated OPLL predictors of outcome. The present systematic review has highlighted an important knowledge gap in the literature and will serve as a basis for future studies.
In conclusion, a hill-shaped lesion on a sagittal image may be an important predictor of a poor surgical outcome following laminoplasty. The body of high-quality evidence that assesses the most important OPLL predictors of outcome, however, is still limited. Further prospective research is required to elucidate the relationships between surgical outcomes and various OPLL factors.
Investigation performed at the University of Toronto, Toronto Western Hospital, Toronto, Ontario, Canada
Disclosure: There was no external source of funding for this study. The Disclosure of Potential Conflicts of Interest forms are provided with the online version of this article.
- Copyright © 2017 by The Journal of Bone and Joint Surgery, Incorporated