Secondary Logo

Journal Logo

Predictive Factors Affecting Outcome After Cervical Laminoplasty

Yoon, S. Tim, MD, PhD*; Raich, Annie, MPH; Hashimoto, Robin E., PhD; Riew, K. Daniel, MD; Shaffrey, Christopher I., MD§; Rhee, John M., MD*; Tetreault, Lindsay A., HBSc; Skelly, Andrea C., PhD; Fehlings, Michael G., MD, PhD, FRCSC, FACS

doi: 10.1097/BRS.0b013e3182a7eb55
Surgical Treatment of Cervical Spondylotic Myelopathy
Free
SDC

Study Design. Systematic review.

Objective. To determine whether various preoperative factors affect patient outcome after cervical laminoplasty for cervical spondylotic myelopathy (CSM) and/or ossification of posterior longitudinal ligament (OPLL).

Summary of Background Data. Cervical laminoplasty is a procedure designed to decompress the spinal cord by enlarging the spinal canal while preserving the lamina. Prior research has identified a variety of potential predictive factors that might affect outcomes after this procedure.

Methods. A systematic search of multiple major medical reference databases was conducted to identify studies explicitly designed to evaluate the effect of preoperative factors on patient outcome after cervical laminoplasty for CSM or OPLL. Studies specifically designed to evaluate potential predictive factors and their associations with outcome were included. Only cohort studies that used multivariate analysis, enrolled at least 20 patients, and adjusted for age as a potential confounding variable were included. JOA (Japanese Orthopaedic Association), modified JOA, and JOACMEQ-L (JOA Cervical Myelopathy Evaluation Questionnaire lower extremity function section) scores were the main outcome measures. Clinical recommendations and consensus statements were made through a modified Delphi approach by applying the GRADE (Grading of Recommendation Assessment, Development and Evaluation)/AHRQ (Agency for Healthcare Research and Quality) criteria.

Results. The search strategy yielded 433 citations, of which 1 prospective and 11 retrospective cohort studies met our inclusion criteria. Overall, the strength of evidence from the 12 studies is low or insufficient for most of the predictive factors. Increased age was not associated with poorer JOA outcomes for patients with CSM, but there is insufficient evidence to make a conclusion for patients with OPLL. Increased severity of disease and a longer duration of symptoms might be associated with JOA outcomes for patients with CSM. Hill-shaped lesions might be associated with poorer JOA outcomes for patients with OPLL. There is insufficient evidence to permit conclusions regarding other predictive factors.

Conclusion. Overall, the strength of evidence for all of the predictive factors was insufficient or low. Given that cervical myelopathy due to CSM tends to be progressive and that increased severity of myelopathy and duration of symptoms might be associated with poorer outcomes after cervical laminoplasty for CSM, it is preferable to perform laminoplasty in patients with CSM earlier rather than waiting for symptoms to get worse. Further research is needed to more clearly identify predictive factors that affect outcomes after cervical laminoplasty because there were relatively few studies identified that used multivariate analyses to control for confounding factors and many of these studies did not provide a detailed description of the multivariate analyses or the magnitude of effect estimates.

Evidence-Based Clinical Recommendations.

Recommendation 1. For patients with CSM, increased age is not a strong predictor of clinical neurological outcomes after laminoplasty; therefore, age by itself should not preclude cervical laminoplasty for CSM.

Overall Strength of Evidence. Low

Strength of Recommendation. Strong

Recommendation 2. For patients with CSM, increased severity of disease and a longer duration of symptoms might be associated with poorer clinical neurological outcomes after laminoplasty; therefore, we recommend that patients be informed about this.

Overall Strength of Evidence. Low

Strength of Recommendation. Strong

Summary Statements. For patients with OPLL, hill-shaped lesions might be associated with poorer clinical neurological outcomes after laminoplasty; therefore, surgeons might consider potential benefits and risks of alternative or additional surgery.

This systematic review was designed to identify predictive factors of cervical laminoplasty outcomes using only studies that were specifically designed to evaluate potential predictive factors associated with outcomes. For patients with cervical spondylotic myelopathy, increased age was not a strong predictor of clinical neurological outcomes; however, increased severity of disease and a longer duration of symptoms might be associated with worse JOA (Japanese Orthopaedic Association) outcomes. For patients with ossification of posterior longitudinal ligament, hill-shaped lesions might be associated with poorer JOA outcomes.

*Emory Spine Center, Atlanta, GA

Spectrum Research, Inc., Tacoma, WA

Washington University Orthopedics, St. Louis, MO

§Department of Neurological Surgery, University of Virginia, Charlottesville, VA; and

University Health Network, Toronto Western Hospital, University of Toronto, Toronto, Ontario, Canada; and

Division of Neurosurgery, Toronto Western Hospital, University of Toronto, Toronto, Ontario, Canada.

Address correspondence and reprint requests to S. Tim Yoon, MD, PhD, Emory Spine Center, 59 Executive Park S. STE 3000, Atlanta, GA 30033; E-mail: styoon@emory.edu

Acknowledgment date: March 15, 2013. First revision date: May 30, 2013. Acceptance date: July 12, 2013.

The manuscript submitted does not contain information about medical device(s)/drug(s).

Supported by AOSpine North America, Inc. Analytic support for this work was provided by Spectrum Research, Inc., with funding from the AOSpine North America.

Relevant financial activities outside the submitted work: fees for participation in review activities such as data monitoring boards, statistical analysis, end point committees, and the like, payment for writing or reviewing the manuscript, support for travel to meetings for the study or other purposes, board membership, consultancy, expert testimony, grants/grants pending, payment for lectures including service on speakers bureaus, patents, royalties, payment for development of educational presentations, stock/stock options, travel/accommodations/meeting expenses unrelated to activities listed, and payment for development of educational presentations.

Cervical laminoplasty is a procedure designed to decompress the spinal cord by enlarging the spinal canal while preserving the lamina. Since the invention of this procedure in Japan, its use has spread throughout the world and has gained increasing usage among spine surgeons. Commensurate with the increase in experience with this technique, the literature on laminoplasty has become extensive. This literature includes refinements of indications for laminoplasty, analysis of complications of laminoplasty, and outcomes of laminoplasty in both the short and long terms for the treatment of cervical spondylotic myelopathy (CSM) and ossification of the posterior longitudinal ligament (OPLL). Previous research has identified a variety of potential prognostic indicators that may affect outcomes after cervical laminoplasty, including age of patient, presence of diabetes, severity of preoperative myelopathy, severity of preoperative axial pain, length of preoperative symptoms, presence and/or degree of OPLL, shape of OPLL lesion, degree of stenosis, number of levels with compression, type of preoperative stenosis, alignment of the cervical spine, the presence of T2 and/or T1 signal change on magnetic resonance imaging (MRI), and preoperative sagittal alignment. This report systematically evaluates whether these preoperative factors affect patient outcome after cervical laminoplasty for CSM and/or OPLL. Outcomes of interest include JOA/mJOA ([modified] Japanese Orthopaedic Association) scores, Nurick scores, NDI (Neck Disability Index) scores, pain, and/or reoperation.

Back to Top | Article Outline

MATERIALS AND METHODS

Electronic Literature Search

A systematic search of PubMed/MEDLINE, the Cochrane Collaboration Library, and Google Scholar for literature published through October 22, 2012, was conducted. Only studies in humans, written in English, and containing abstracts were considered for inclusion. The focus was on identification of studies explicitly designed to evaluate the effect of preoperative factors on patient outcome after cervical laminoplasty for cervical myelopathy (CSM and OPLL). Details of the inclusion and exclusion criteria are given in Table 1. Terms specific to CSM are as follows: cervical spondylotic myelopathy OR cervical myelopathy OR (cervical AND myelopathy) OR Ossification of Posterior Longitudinal Ligament[MeSH] OR Ossification of Posterior Longitudinal Ligament. They were combined with terms specifying the cervical spine (Cervical OR Cervical Vertebrae[MeSH]) and those related to treatment or surgery (Laminoplasty OR laminoplast* OR laminaplast*). Studies specifically designed to evaluate potential predictive factors and their associations with outcomes were sought. The search was limited to cohort studies that used multivariate analysis and that enrolled at least 20 patients. Furthermore, to evaluate the highest quality studies, we included only those studies that adjusted for preoperative JOA score and age as potential confounding variables.

TABLE 1

TABLE 1

Case reports, meeting abstracts/proceedings, white papers, and editorials were excluded. Table 1 provides additional information on inclusion/exclusion criteria. We also hand-searched key references to identify any additional studies that were not found by our systematic literature search.

Back to Top | Article Outline

Data Extraction

The following demographic information was abstracted from each study: population, condition, details of the intervention, and follow-up information. Outcomes abstracted focused on those outcomes listed in our inclusion table (Table 1), namely, myelopathy scores (i.e., JOA/mJOA or Nurick scores), patient-reported pain outcomes (including NDI scores), and reoperation rates. Details of the multivariate analysis, including dependent and independent variables, odds ratios (ORs), 95% confidence intervals (CIs), and P values, were abstracted. All predictive factors in each study's multivariate analysis are included in the results tables. The interpretation of results includes adjustment (holding factors constant) for all of the predictive factors evaluated for each particular study.

Back to Top | Article Outline

Study Quality and Overall Strength of Body of Literature

Class of evidence (CoE) ratings were assigned to each included article independently by 2 reviewers (A.R., R.H.), using criteria set by The Journal of Bone and Joint Surgery1 for prognostic studies and modified to delineate criteria associated with methodological quality and risk of bias based on recommendations made by the Agency for Healthcare Research and Quality.2,3 The appraisal system used in this article accounts for features of methodological quality and important sources of bias by combining epidemiological principles with characteristics of study design to determine the CoE and are consistent with those used in previous focus issues.4 See the Supplemental Digital Content material (available at http://links.lww.com/BRS/A825) for details on individual study ratings.

After individual article evaluation, the strength of the overall body of evidence with respect to each outcome was determined on the basis of precepts outlined by the Grades of Recommendation Assessment, Development and Evaluation (GRADE) Working Group5,6 and recommendations made by the Agency for Healthcare Research and Quality (AHRQ).2,3 Qualitative analysis is performed considering AHRQ-required and additional domains.7 The initial strength of the overall body of evidence was considered “high” if the majority of the studies were class I or II and “low” if the majority of the studies were class III or IV. Criteria for downgrading published evidence 1 or 2 levels included the following: (1) inconsistency of results, (2) indirectness of evidence, (3) imprecision of the effect estimates (e.g., wide CIs), or (4) non–a priori statement of subgroup analyses. Alternatively, the body of evidence could be upgraded 1 or 2 levels on the basis of the following factors: (1) large magnitude of effect or (2) dose-response gradient. The final overall strength of the body of literature expresses our confidence in the estimate of effect and the impact that further research may have on the results. An overall strength of “high” means we have high confidence that the evidence reflects the true effect. Further research is very unlikely to change our confidence in the estimate of effect. The overall strength of “moderate” means we have moderate confidence that the evidence reflects the true effect. Further research may change our confidence in both the estimate of effect and the estimate. A grade of “low” means we have low confidence that the evidence reflects the true effect. Further research is likely to change the confidence in the estimate of effect and likely to change the estimate. Finally, a grade of “insufficient” means that evidence either is unavailable or does not permit a conclusion. The Supplemental Digital Content (available at http://links.lww.com/BRS/A825) contains the details of how we arrived at the strength of evidence for each key question.

Back to Top | Article Outline

Data Analysis

When the data were available, we reported ORs, 95% CIs, and P values. ORs greater than 1.0 suggest increased risk of a poor recovery (poor improvement of JOA or mJOA score) for a given predictor; values less than 1 suggest decreased risk. For the studies that reported ORs for predictors of a good recovery, the OR was recalculated for a poor recovery by taking the inverse of the OR and revising the description of the results to reflect prediction of a poor recovery for consistency. For studies that did not report effect measures, we reported P values if available or the significance of the predictive factor as described by the study authors. Some studies reporting a JOA outcome score performed a linear regression analysis.8–12

Back to Top | Article Outline

Outcome Measures Reported

The most common outcome measures used included the following:

  • JOA score: evaluation of the neurological function of patients with cervical myelopathy; range of score 0 to 17, with a lower score indicating a poor outcome.
  • JOACMEQ-L (Japanese Orthopedic Association Cervical Myelopathy Evaluation Questionnaire, lower extremity function section of JOACMEQ): used to evaluate cervical compression myelopathy; range of score 0 to 100, with lower score indicating a poor outcome.
  • JOA recovery rate = 100 × (postoperative JOA score − preoperative JOA score)/(17 − preoperative JOA score).
  • Motor function score of lower extremity portion of JOA: range of score 0 to 4, with a lower score indicating a poor outcome.
Back to Top | Article Outline

Clinical Recommendations and Consensus Statements

Clinical recommendations or consensus statements were made through a modified Delphi approach by applying the GRADE/AHRQ 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, and values and patient preferences. In some instances, costs may have been considered. A more thorough description of this process can be found in the focus issue Methods article.

Back to Top | Article Outline

RESULTS

Study Selection

The search strategy yielded 433 potentially relevant citations; of these, 373 were excluded on the basis of title and/or abstract and 60 were selected for full-text review (Figure 1). An additional 48 were excluded on the basis of full-text review for the following reasons: the study did not control for potential confounding factors using multivariate analysis (38 studies), study participants did not receive laminoplasty treatment (7 studies), results were not reported clearly or sufficiently enough to ascertain whether a multivariate analysis was performed (2 studies),13,14 or outcomes of interest were not reported (1 study). Details of the excluded articles can be found in the Supplemental Digital Content material (available at http://links.lww.com/BRS/A825).

Figure 1

Figure 1

Twelve studies were selected for inclusion and are summarized in this report (see Table 2 and Supplemental Digital Content Table 1, available at http://links.lww.com/BRS/A825, for detailed information). Five studies included a population of at least 80% patients with CSM,11,15–18 4 studies included only patients with OPLL,8,9,19,20 2 studies included both patients with CSM and patients with OPLL,10,12,21 and 1 study included patients with CSM and patients with OPLL but evaluated them separately.12 Mean patient ages ranged from 43.8 to 77 years, and males comprised 30% to 79% of the study populations. Mean follow-up times ranged from 1 to 12.2 years. All studies used laminoplasty, and although the specific approaches varied (see Supplemental Digital Content Table 1, available at http://links.lww.com/BRS/A825, for details), the studies can be aggregated into 2 French-door studies and 9 open-door studies. All included studies evaluated whether the predictive factors of interest influence outcome as measured by the JOA or mJOA score and controlled for any potentially confounding effects that age and preoperative severity of myelopathy may have on outcomes. None of the studies identified for inclusion evaluated the effect that these potential factors had on other outcomes of interest, including Nurick scores, NDI scores, pain, or reoperation. After methodological review, 1 prospective cohort study15 and 4 retrospective cohort studies9–11,16 were graded CoE II, indicating a moderately low risk of bias, and 7 retrospective studies8,12,17–21 received a CoE III grade, indicating that they were at moderately high risk of bias. Further details, including additional predictive factors that were reported but not discussed later, are available in Supplemental Digital Content Tables 2–4 (available at http://links.lww.com/BRS/A825).

TABLE 2-a

TABLE 2-a

TABLE 2-b

TABLE 2-b

TABLE 2-c

TABLE 2-c

Back to Top | Article Outline

Sociodemographic Characteristics and Comorbidities

Age

All 12 studies investigated the role of age in laminoplasty outcomes in univariate and/or multivariate analyses (Table 3). Of the studies in which the majority of patients had CSM, 4 of 6 studies found that age was not significantly associated with laminoplasty outcomes in univariate analysis and thus did not further evaluate its effects using multivariate analyses.11,16–18 One study found that age of at least 60 years correlated with decreased JOA scores on multivariate analysis; however, the statistical significance of this outcome was not clear and the partial correlation coefficient value was not high enough to meet the study's criteria for entering age into a multiple regression analysis.12 In a multivariate analysis, 1 study of patients with CSM found that older patients were slightly more likely to have a poor recovery (<50% recovery rate) than younger patients (OR: 1.05; 95% CI: 1.00–1.11; P = 0.04), although the effect was marginally significant.15 In contrast, the same study found that age was not a predictor of postoperative JOACMEQ-L scores on univariate analysis.15

TABLE 3-a

TABLE 3-a

TABLE 3-b

TABLE 3-b

In studies that included only patients with OPLL, there were conflicting results for the effect of age on poor recovery. Two of the 5 studies reported that older age was a statistically significant predictor of a lower JOA recovery rate on multivariate analysis. Fujimura et al19 found that older patients were significantly more likely to be at risk for a lower recovery rate (P < 0.0001) and a significant decrease in recovery rate between the 1- and 5-year follow-ups (P < 0.0001) than younger patients. Wang et al20 reported that older age increased the risk of a lower JOA recovery rate on multiple linear regression analysis; however, the P value was not reported. In addition, one study found that age of at least 60 years was associated with lower JOA scores according to multivariate analysis; however, the statistical significance of this outcome was not reported.12 In contrast, 2 studies found that age was not significantly associated with lower postoperative JOA scores: P = 0.064 (Iwasaki et al9) and P = 0.1866 (Iwasaki et al8).

These 2 studies that included both patients with CSM and patients with OPLL found that age was significantly associated with outcome. One study found that older age significantly increased the risk of a lower recovery rate and lower postoperative JOA score (P < 0.0002 and P = 0.0001, respectively) on multivariate analysis.10 A second study reported that older age was a significant predictor of a JOA recovery rate of less than 50% on univariate analyses but did not evaluate age using multivariate analysis.21

Back to Top | Article Outline

Diabetes

In a study of both patients with CSM and patients with OPLL undergoing laminoplasty, Kim et al21 found that diabetic patients had almost 3 times the risk of a poor neurological recovery (<50% recovery rate) than patients without diabetes (OR: 2.92; 95% CI: 1.32–6.12; P = 0.01). The same study also found an increased risk for a poor recovery in diabetic patients of older age (OR: 2.21; 95% CI: 1.15–4.23; P = 0.04) or who smoked (OR: 4.01; 95% CI: 1.89–8.32; P = 0.02) (see Supplemental Digital Content Table 4, available at http://links.lww.com/BRS/A825).

Back to Top | Article Outline

CSM and Characteristics

Severity of Preoperative Myelopathy (Preoperative JOA Score)

Ten studies investigated the role of the preoperative JOA score in laminoplasty outcomes in univariate and/or multivariate analyses (Table 4). Of the 6 studies comprising only patients with CSM, the results were inconsistent. Three studies evaluated the preoperative JOA score in multivariate analyses and found that this factor was significantly associated with outcome.11,12,16 Naruse et al16 found that patients with a low preoperative JOA score were significantly more likely to have a less than 50% recovery rate than patients with a high preoperative JOA score (OR: 1.65; 95% CI: 1.20–2.25; P = 0.0019). Tanaka et al11 reported that a lower preoperative JOA score significantly increased the risk of a lower postoperative JOA score (P < 0.0001). The same study also found that a lower preoperative lower extremity JOA score significantly increased the risk of a lower postoperative lower extremity JOA score (P < 0.0001). Uchida et al12 found that a preoperative JOA score of less than 7 points in patients with CSM was significantly associated with a decreased postoperative JOA score (P = 0.0329) in multivariate analysis. One study found that preoperative JOA score was significantly associated with poor outcome by univariate analysis (OR: 0.76; 95% CI: 0.64–0.91; P < 0.01), but the factor was not included in the stepwise multivariate analysis.17 Two studies found that the preoperative JOA score was not significantly associated with laminoplasty outcomes in univariate analysis and did not further evaluate its effects using multivariate analyses.15,17,18

TABLE 4

TABLE 4

Of the 4 studies evaluating this predictive factor in which the majority of patients had OPLL, all found that a lower preoperative JOA score was significantly associated with poorer outcomes in multivariate analysis.8,9,12,20 Two studies reported that a lower preoperative JOA score increased the risk of a lower postoperative JOA score: P = 0.0001 (Iwasaki et al8), P = 0.0015 at the final follow-up, and P = 0.0003 at the follow-up with the maximum JOA score.9 A third study reported that a lower preoperative JOA score was a predictor of a lower recovery rate (P value not reported).20 Finally, the fourth study found that a preoperative JOA score of less than 7 points was significantly associated with a lower postoperative JOA score (P = 0.0375).12

In a study that included both patients with CSM and patients with OPLL, Morio et al10 reported that a lower preoperative JOA score significantly increased the risk of a lower postoperative JOA score (P < 0.0001).

Back to Top | Article Outline

Duration of Preoperative Symptoms/Disease/Myelopathy

Six studies evaluated whether duration of symptoms affected laminoplasty outcomes, using univariate and/or multivariate analyses (Table 5). Of the 3 studies of patients with CSM, all 3 found that patients with a longer duration of symptoms were at a higher risk for a recovery rate of less than 50% or a lower postoperative JOA score.11,15,18 Nakashima et al15 found that patients with a longer duration of symptoms were slightly more likely to have a recovery rate of less than 50% than patients with a shorter duration of symptoms (OR: 1.19; 95% CI: 1.03–1.37; P = 0.016). In contrast, this study also found that the duration of symptoms was not significantly associated with an ineffective result using the JOACMEQ-L score on univariate analysis and thus did not further evaluate its effects in a multivariate analysis. Tanaka et al11 reported that a longer duration of symptoms and lower extremity disability significantly increased the risk of a lower postoperative JOA score and a JOA motor function score of lower extremity (P < 0.0001 for both factors). Similarly, Wada et al18 found that a longer duration of symptoms significantly increased the risk of a lower recovery rate (P < 0.05).

TABLE 5

TABLE 5

Of the 2 studies in which the majority of patients had OPLL, both reported that an increased duration of preoperative symptoms in laminoplasty outcomes was significantly associated with poorer outcomes, using univariate and/or multivariate analyses.19,20 Fujimura et al19 reported that a longer duration of myelopathy significantly increased the risk of a lower recovery rate in multivariate analysis (P < 0.0001 at 1-, 3-, and 5-yr follow-up). Wang et al20 reported that a longer duration of disease is the best combination of predictors of a lower recovery rate in a multivariate analysis (P value not reported).

In a study that included both patients with CSM and patients with OPLL, Morio et al10 found that a longer duration of symptoms significantly increased the risk of a lower recovery rate and a postoperative JOA score in a multivariate analysis (P = 0.0002 and P < 0.0001, respectively).

Back to Top | Article Outline

Shape of OPLL Lesion

One study in which the majority of patients had OPLL evaluated the role of OPLL lesion shape in laminoplasty outcomes in univariate and/or multivariate analyses.9 Iwasaki et al9 found that hill-shaped lesions significantly increased the risk of a lower recovery rate (P < 0.0001 at the final follow-up and P < 0.0001 at the time of maximum recovery) and a lower JOA score (P < 0.0001 at the final follow-up and P < 0.0001 at the time of maximum recovery) than plateau-shaped lesions based on the OPLL classification system of Iwasaki.22 The author defined hill-shaped ossifications as demonstrating a massive beak-shaped ossification localized to certain levels (see Supplemental Digital Content Table 3, available at http://links.lww.com/BRS/A825).

Back to Top | Article Outline

Degree or Severity of Stenosis

Three studies evaluated whether severity of stenosis affected laminoplasty outcomes (Table 6). Overall, results from the 2 studies in which the majority of patients had CSM suggested that increased preoperative stenosis is predictive of poorer outcomes in univariate and/or multivariate analyses. Wada et al18 found that stenosis (as measured by 2 methods) significantly increased the risk of a negative postoperative recovery rate (P < 0.05 for both measures of stenosis). This study defined stenosis as a decrease in anteroposterior canal diameter at maximum compression (measured on radiograph) or the transverse area of the spinal cord at maximum compression (measured on computed tomographic [CT] myelography). Uchida et al12 also used 2 measures of stenosis to investigate its role as a potential predictor of the postoperative JOA score. This study found that less than 50% flattening of the spinal cord on MRI in patients with spondylosis significantly increased the risk of a lower JOA score (P = 0.0116); however, spinal canal stenosis defined as a canal diameter of less than 12 mm in the sagittal plane on radiograph was not significantly associated with a lower JOA score in patients with spondylosis (P > 0.05).

TABLE 6

TABLE 6

In studies that included only patients with OPLL, there were conflicting results in the 2 studies that evaluated the role of the degree of stenosis on laminoplasty outcomes in univariate and/or multivariate analyses.9,12 Iwasaki et al8 reported that the space available for the spinal cord was not significantly associated with the postoperative JOA score (P > 0.05). Uchida et al12 reported that a narrowing rate of less than 40% at the most affected segment on the preoperative CT scan did not increase the risk of a lower JOA score in patients with OPLL in multivariate analysis (P > 0.05), but, in contrast, reported that less than 30% flattening on the preoperative MRI significantly increased the probability of a poor long-term outcome in multivariate analysis (P = 0.0434).

No studies comprising both CSM and OPLL studies evaluated stenosis as a predictive factor for laminoplasty outcomes.

Back to Top | Article Outline

Number of Levels With Compression

Only one study investigated the effect of the number of levels with compression on laminoplasty outcomes in univariate and/or multivariate analyses. Uchida et al12 found that involvement at 3 or more levels significantly increased the risk of a lower JOA score for both patients with spondylosis and patients with OPLL (P = 0.0293 and P = 0.0029, respectively) (see Supplemental Digital Content Tables 2 and 3, available at http://links.lww.com/BRS/A825).

Back to Top | Article Outline

Radiographical/Neurological Characteristics

Presence of Preoperative Kyphosis

One study of patients with CSM investigated preoperative cervical alignment as a predictor for laminoplasty outcomes in univariate and/or multivariate analyses. Suda et al17 found that patients with local kyphosis (angle >5°) had a slightly increased risk of a less than 50% recovery rate than patients without local kyphosis (OR: 1.21; 95% CI: 1.08–1.36; P < 0.01) (see Supplemental Digital Content Table 2, available at http://links.lww.com/BRS/A825).17

Back to Top | Article Outline

Presence of T2 and/or T1 Signal Change on MRI

T1 and T2 signal changes on MRI were variously defined by the 6 studies that evaluated them as a potential predictive factor, so caution must be exercised in the interpretation of the results (Table 7). The studies evaluated either only T2 signal changes or concurrent T1 and T2 signal changes.

TABLE 7

TABLE 7

Back to Top | Article Outline

Only T2 Changes

In studies that included only patients with CSM, 2 studies reported that a T2 signal intensity change had no effect on laminoplasty outcomes in univariate and/or multivariate analyses. Nakashima et al15 measured spinal cord signal intensity changes on T2-weighted MRI in the sagittal and axial planes and reported percent change of signal intensity. This study found that T2 signal intensity changes in patients with CSM did not significantly increase the risk of an ineffective result measured by the JOACMEQ-L score (P = 0.98; refer to Table 3 for the definition of an ineffective result).15 This same study also found that a T2 signal intensity change was not significantly associated with a recovery rate of less than 50% in a univariate analysis and thus did not further evaluate its effects using multivariate analyses. Wada et al18 also found that a T2 signal intensity change was not significantly associated with a negative recovery rate in a univariate analysis (P > 0.05) and did not further evaluate its effects in a multivariate analysis.

One study comprising patients with OPLL evaluated the role of T2 signal intensity change on laminoplasty outcomes in univariate and/or multivariate analyses. Wang et al20 used sagittal T2-weighted MRI and reported the signal intensity ratio between areas of compressed spinal cord and normal spinal cord at the C7–T1 disc level and found that an increased signal intensity ratio with positive pyramidal signs was associated with a lower recovery rate (P value not reported).

Back to Top | Article Outline

Concurrent T1 and T2 Changes

In a study that included only patients with CSM, Suda et al17 measured preoperative signal intensity changes on T1- and T2-weighted MRI in the sagittal and axial planes and reported the presence or absence of these changes. This study found that patients with a signal intensity change on MRI had a 4 times higher risk of a recovery rate of less than 50% than patients with CSM with no signal intensity change (OR: 4.10; 95% CI: 1.51–11.12; P < 0.01).17 However, the wide CI for the OR suggests great variability in the effect estimate and casts doubt as to the certainty of the estimate. In considering both preoperative kyphosis and signal intensity changes, Suda et al17 also reported that the maximum local kyphosis angle leading to a good outcome was within 13° without a signal intensity change and within 5° with a signal intensity change.

These two studies including both patients with CSM and patients with OPLL found that concurrent T1 and T2 signal intensity changes were significantly associated with poorer laminoplasty outcomes in univariate and/or multivariate analyses. Kim et al21 defined a signal intensity change as being present when there was an increased signal on T2 with an accompanying decreased signal on T1 and found that patients with a signal intensity change were almost 4 times more likely to have a recovery rate of less than 50% than patients without a signal intensity change in multivariate analysis (OR = 3.53; 95% CI: 1.67–5.95; P = 0.01). Morio et al10 defined 3 patterns of signal intensity changes on T1-weighted/T2-weighted MRI, normal/normal, normal/high, and low/high, and found that the low-intensity signal changes on T1-weighted sequences significantly increased the risk of a lower recovery rate (P = 0.0002) and a lower postoperative JOA score (P < 0.0001) in multivariate analyses.

Back to Top | Article Outline

Evidence Summary

Results are summarized in Table 8, and evidence summaries may be found in Table 9. The overall strength of the evidence for increased age, severity of disease, and a longer duration of symptoms as risk factors in patients with CSM is “low,” that is, there is a low confidence that the evidence reflects the true effect, and further research is both likely to change the confidence in the estimate of effect and likely to change the estimate. The overall strength of evidence for severity/degree of stenosis, number of levels with compression, local kyphosis, and signal intensity change/ratio as risk factors in patients with CSM is “insufficient,” that is, the available evidence does not permit a conclusion.

TABLE 8

TABLE 8

TABLE 9-a

TABLE 9-a

TABLE 9-b

TABLE 9-b

TABLE 9-c

TABLE 9-c

The overall strength of the evidence for OPLL lesion type as a risk factor in patients with OPLL is “low,” that is, there is a low confidence that the evidence reflects the true effect, and further research is both likely to change the confidence in the estimate of effect and likely to change the estimate. The overall strength of evidence for increased age, severity of disease, a longer duration of symptoms, severity/degree of stenosis, number of levels with compression, and signal intensity change/ratio as risk factors in patients with OPLL is “insufficient,” that is, the available evidence does not permit a conclusion.

The overall strength of the evidence for increased age, severity of disease, and a longer duration of symptoms as risk factors in a mixed population of patients with CSM and OPLL is “low,” that is, there is a low confidence that the evidence reflects the true effect, and further research is both likely to change the confidence in the estimate of effect and likely to change the estimate. The overall strength of evidence for signal intensity change/ratio and diabetes and/or smoking or older age as risk factors in a mixed population of both patients with CSM and patients with OPLL is “insufficient,” that is, the available evidence does not permit a conclusion.

Further details are available in Table 9.

Back to Top | Article Outline

DISCUSSION

Overall, reporting of the multivariate analyses for the studies was poor; many studies did not state explicitly the variables included in the multivariate regression models or how the final regression model was selected. This, combined with the failure of authors to describe the magnitude of effect estimates adequately, makes it difficult both to assess the overall strength of evidence across studies and to draw meaningful conclusions for some predictive factors. In this report, the OR is liberally interpreted to be an estimate of relative risk; however, because the prevalence of a poor outcome was not reported in most studies reporting ORs, the OR may overestimate the relative risk and caution should be exercised in interpreting the risk of a poor outcome. Wide CIs suggest greater variability in effect estimates and call into question the certainty of the effect estimate.

The varying definitions of outcome across studies and failure in some studies to define thresholds for improved or worse scores precluded the pooling of data and made meaningful synthesis challenging. None of the authors provided a definition of a clinically meaningful improvement for any outcome. One author group15 reported performing a logistic regression yet reported an HR (assumed to be a hazard ratio). We reported the OR, which is the appropriate measure from a logistic regression.

It is not surprising that preoperative (baseline) JOA scores are associated with the JOA outcome measures. The formula for calculating the JOA recovery rate includes the preoperative JOA score, which adjusts for, or takes into account, the preoperative baseline JOA score. Some of the factors were considered indirect measures because it is not clear to what extent they may have a direct influence on the outcome(s) of interest and how they may influence patient counseling or clinical decision making. Such factors include lesion type/configuration, changes in magnetic resonance signal, and increased age. Factors such as symptom intensity and duration may be subjective.

With regard to age as a risk factor, 5 of the studies assessed it as a potential factor, but because a univariate association was not found, it was not included in final multivariate models. In multivariate analyses, 1 study found that age and diabetes together were significant predictors of a poor recovery, 3 studies found that age alone was a significant predictor of a poor recovery, and 3 studies found that age did not increase the risk of a poor recovery. Study sample size and/or degree of variation in age in some studies may have influenced the ability to detect a statistical association.

We have included illustrative cases of a successful laminoplasty (Figure 2A–C) and a failed laminoplasty (Figure 3A–C).

Figure 2

Figure 2

Figure 3

Figure 3

Evidence-Based Clinical Recommendations.

Recommendation 1. For CSM patients, increased age is not a strong predictor of clinical neurological outcomes following laminoplasty, therefore age by itself should not preclude cervical laminoplasty for CSM.

Overall Strength of Evidence. Low

Strength of Recommendation. Strong

Recommendation 2. For CSM patients, increased severity of disease and longer duration of symptoms might be associated with poorer clinical neurological outcomes following laminoplasty, therefore we recommend that patients be informed about this.

Overall Strength of Evidence. Low

Strength of Recommendation. Strong

Summary Statements. For OPLL patients, Hill-shaped lesions might be associated with poorer clinical neurological outcomes following laminoplasty therefore surgeons might consider potential benefits and risks of alternative or additional surgery.

Back to Top | Article Outline

Key Points

  • A systematic search of the literature identified 12 studies with at least 20 patients; 3 studies that used multivariate analysis to analyze predictive factors for outcomes after cervical laminoplasty for the treatment of CSM or OPLL.
  • For patients with CSM, increased age was not associated with a poorer clinical outcome, but for patients with OPLL, there is insufficient evidence to arrive at a conclusion.
  • For patients with CSM, increased severity of disease and a longer duration of symptoms might be associated with worse clinical outcome.
  • For patients with OPLL, hill-shaped lesions might be associated with poorer clinical outcomes.
Back to Top | Article Outline

Acknowledgments

The authors thank Nancy Holmes and Ms. Chi Lam for their administrative assistance.

Author contributions are as follows: T.Y.: study concept, manuscript preparation, manuscript revision, data analysis and interpretation, and contribution of illustrative cases; A.R.: manuscript preparation, manuscript revision, data analysis, and interpretation; R.H.: manuscript preparation, manuscript revision, data analysis, and interpretation; D.R.: study concept, manuscript preparation, manuscript revision, and data analysis and interpretation; C.I.S.: study concept, manuscript revision, and data interpretation; J.R.: study concept, manuscript revision, and data interpretation; L.A.T.: study concept, manuscript revision, and data interpretation; A.S.: data analysis and interpretation, manuscript preparation, and manuscript revision; M.G.F.: study concept, manuscript revision, and data interpretation.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF version of this article on the journal's Web site (www.spinejournal.com).

Back to Top | Article Outline

References

1. Wright JG, Swiontkowski MF, Heckman JD. Introducing levels of evidence to the journal. J Bone Joint Surg Am 2003;85-A:1–3.
2. Agency for Healthcare Research and Quality. Methods Guide for Effectiveness and Comparative Effectiveness Reviews. Rockville, MD: Agency for Healthcare Research and Quality; 2012. AHRQ Publication No. 10(12)-EHC063-EF.
3. West S, King V, Carey TS, et al. Systems to Rate the Strength of Scientific Evidence. Evidence Report/Technology Assessment No. 47. Agency for Healthcare Research and Quality, Rockville, MD; 2002. Prepared by the Research Triangle Institute–University of North Carolina Evidence-based Practice Center, Contract No. 290-97-0011.
4. Norvell DC, Dettori JR, Skelly AC, et al. Methodology for the systematic reviews on an adjacent segment pathology. Spine (Phila Pa 1976) 2012;37:S10–7.
5. Atkins D, Best D, Briss PA, et al. Grading quality of evidence and strength of recommendations. BMJ 2004;328:1490.
6. Balshem H, Helfand M, Schunemann HJ, et al. GRADE guidelines, part 3: rating the quality of evidence. J Clin Epidemiol 2011;64:401–6.
7. Owens DK, Lohr KN, Atkins D, et al. AHRQ series paper 5: grading the strength of a body of evidence when comparing medical interventions—Agency for Healthcare Research and Quality and the effective health-care program. J Clin Epidemiol 2010;63:513–23.
8. Iwasaki M, Kawaguchi Y, Kimura T, et al. Long-term results of expansive laminoplasty for ossification of the posterior longitudinal ligament of the cervical spine: more than 10 years follow up. J Neurosurg 2002;96:180–9.
9. Iwasaki M, Okuda S, Miyauchi A, et al. Surgical strategy for cervical myelopathy due to ossification of the posterior longitudinal ligament, part 1: clinical results and limitations of laminoplasty. Spine (Phila Pa 1976) 2007;32:647–53.
10. Morio Y, Teshima R, Nagashima H, et al. Correlation between operative outcomes of cervical compression myelopathy and MRI of the spinal cord. Spine (Phila Pa 1976) 2001;26:1238–45.
11. Tanaka J, Seki N, Tokimura F, et al. Operative results of canal-expansive laminoplasty for cervical spondylotic myelopathy in elderly patients. Spine (Phila Pa 1976) 1999;24:2308–12.
12. Uchida K, Nakajima H, Sato R, et al. Multivariate analysis of the neurological outcome of surgery for cervical compressive myelopathy. J Orthop Sci 2005;10:564–73.
13. Chung SS, Lee CS, Chung KH. Factors affecting the surgical results of expansive laminoplasty for cervical spondylotic myelopathy. Int Orthop 2002;26:334–8.
14. Ogawa Y, Chiba K, Matsumoto M, et al. Long-term results after expansive open-door laminoplasty for the segmental-type of ossification of the posterior longitudinal ligament of the cervical spine: a comparison with nonsegmental-type lesions. J Neurosurg Spine 2005;3:198–204.
15. Nakashima H, Yukawa Y, Ito K, et al. Prediction of lower limb functional recovery after laminoplasty for cervical myelopathy: focusing on the 10-s step test. Eur Spine J 2012;21:1389–95.
16. Naruse T, Yanase M, Takahashi H, et al. Prediction of clinical results of laminoplasty for cervical myelopathy focusing on spinal cord motion in intraoperative ultrasonography and postoperative magnetic resonance imaging. Spine (Phila Pa 1976) 2009;34:2634–41.
17. Suda K, Abumi K, Ito M, et al. Local kyphosis reduces surgical outcomes of expansive open-door laminoplasty for cervical spondylotic myelopathy. Spine (Phila Pa 1976) 2003;28:1258–62.
18. Wada E, Yonenobu K, Suzuki S, et al. Can intramedullary signal change on magnetic resonance imaging predict surgical outcome in cervical spondylotic myelopathy? Spine (Phila Pa 1976) 1999;24:455–61; discussion 62.
19. Fujimura Y, Nishi Y, Chiba K, et al. Multiple regression analysis of the factors influencing the results of expansive open-door laminoplasty for cervical myelopathy due to ossification of the posterior longitudinal ligament. Arch Orthop Trauma Surg 1998;117:471–4.
20. Wang LF, Zhang YZ, Shen Y, et al. Using the T2-weighted magnetic resonance imaging signal intensity ratio and clinical manifestations to assess the prognosis of patients with cervical ossification of the posterior longitudinal ligament. J Neurosurg Spine 2010;13:319–23.
21. Kim HJ, Moon SH, Kim HS, et al. Diabetes and smoking as prognostic factors after cervical laminoplasty. J Bone Joint Surg Br 2008;90:1468–72.
22. Iwasaki M. Ossification of the posterior longitudinal ligament. In: Rothman RH, Simeone FA, Herkowitz HN, eds. The Spine. Philadelphia: Saunders Elsevier; 2006:896–912.
Keywords:

cervical laminoplasty; JOA; mJOA; cervical spondylotic myelopathy; ossification of posterior longitudinal ligament; CSM; OPLL

Supplemental Digital Content

Back to Top | Article Outline
© 2013 by Lippincott Williams & Wilkins