In adults older than 55 years of age, CSM is the most

Transcription

1 J Neurosurg Spine 14: , 2011 Functional and clinical outcomes following surgical treatment in patients with cervical spondylotic myelopathy: a prospective study of 81 cases Clinical article Julio C. Furlan, M.D., M.B.A., M.Sc., Ph.D., 1 Sukhvinder Kalsi-Ryan, B.Sc.P.T., M.Sc., Ph.D., 2,3 Ahilan Kailaya-Vasan, M.B.B.S., B.Sc., 4 Eric M. Massicotte, M.D., M.Sc., F.R.C.S.C., 3,5 and Michael G. Fehlings, M.D., Ph.D., F.R.C.S.C. 1,3,5 1 Division of Genetics and Development, Toronto Western Research Institute, and 3 Spinal Program, Krembil Neuroscience Centre, Toronto Western Hospital, University Health Network; 2 Graduate Department of Rehabilitation Science, and 5 Department of Surgery, Division of Neurosurgery, University of Toronto, Ontario, Canada; and 4 University College London, United Kingdom Object. Cervical spondylotic myelopathy (CSM) is the most common cause of spinal dysfunction in the elderly. Operative management is beneficial for most patients with moderate/severe myelopathy. This study examines the potential confounding effects of age, sex, duration of symptoms, and comorbidities on the functional outcomes and postoperative complications in patients who underwent cervical decompressive surgery. Methods. We included consecutive patients who underwent surgery from December 2005 to October Functional outcomes were assessed using the Nurick grading system and the modified Japanese Orthopaedic Association and Berg Balance scales. Comorbidity indices included the Charlson Comorbidity Index and the number of ICD-9 codes. Results. There were 57 men and 24 women with a mean age of 57 years (range years). The mean duration of symptoms was 25.2 months (range months). There was a significant functional recovery from baseline to 6 months after surgery (p < 0.01). Postoperative complications occurred in 18.5% of cases. Although the occurrence of complications was not significantly associated with sex (p = 0.188), number of ICD-9 codes (p = 0.113), duration of symptoms (p = 0.309), surgical approach (p = 0.248), or number of spine levels treated (p = 0.454), logistic regression analysis showed that patients who developed complications were significantly older than patients who had no complications (p = 0.018). Only older age (p < 0.002) and greater number of ICD-9 codes (p < 0.01) were significantly associated with poorer functional recovery after surgical treatment. However, none of the studied factors were significantly associated with clinically relevant functional recovery after surgical treatment for CSM (p > 0.05). Conclusions. Our results indicate that surgery for CSM is associated with significant functional recovery, which appears to reach a plateau at 6 months after surgery. Age is a potential predictor of complications after decompressive surgery for CSM. Whereas older patients with a greater number of preexisting medical comorbidities had less favorable functional outcomes after surgery for CSM in the multivariate regression analysis, none of the studied factors were associated with clinically relevant functional recovery after surgery in the logistic regression analysis. Therefore, age-matched protocols based on preexisting medical comorbidities may reduce the risk for postoperative complications and improve functional outcomes after surgical treatment for CSM. (DOI: / SPINE091029) Key Words cervical spondylotic myelopathy decompressive surgery cervical spine Nurick grade Berg Balance Scale modified Japanese Orthopaedic Association score In adults older than 55 years of age, CSM is the most common cause of spinal cord dysfunction that causes progressive disability and adversely affects quality of life. 13 Cervical spondylosis is a degenerative process involving the discs and joints of the cervical spine, which Abbreviations used in this paper: BBS = Berg Balance Scale; CCI = Charlson Comorbidity Index; CSM = cervical spondylotic myelopathy; MCID = minimum clinically important difference; mjoa = modified Japanese Orthopaedic Association. can eventually result in spinal cord compression. 1,13 The classic studies on the natural history of CSM report a stepwise pattern of disease progression with exacerbation of symptoms and subsequent long periods of static or worsening function or, seldom, improvement. 18 Development of moderate to severe signs and symptoms usually indicates a poor prognosis, and, in those cases in which the clinical status and imaging findings corroborate each other, surgical intervention may play a key role in the management of patients with CSM to 348 J Neurosurg: Spine / Volume 14 / March 2011

2 Outcomes of surgery for cervical spondylotic myelopathy counter an unfavorable natural history. 13 Cervical spondylotic myelopathy has become one of the most common diseases treated by spine surgeons. In a population-based study, hospital admissions of patients with CSM increased 2-fold, from 3.73 to 7.88 per 100,000 inhabitants, in the US between 1993 and 2002, whereas the number of patients with CSM who underwent spine fusion increased 7-fold, from 0.6 to 4.1 per 100,000 US population. 17 Despite the greater increase in the number of surgical interventions relative to the number of hospital admissions for CSM, there were no significant differences between the earlier 5-year period ( ) and the later 5-year period ( ) with respect to postoperative complications (10.9% vs 10.3%, respectively) and mortality (0.6% vs 0.55%, respectively). 17 Also, Lad et al. 17 reported a significant decrease in the number of CSM patients without preexisting medical comorbidity from the earlier period (48.8%) to the later period (33.5%). Given this background, a prospective study was undertaken to examine the potential effects of age, sex, duration of preoperative symptoms, and preexisting medical comorbidities on functional outcomes and postoperative complications in patients who underwent cervical decompressive surgery for CSM. Methods The University Health Network Research Ethics Board approved the protocol for this study. This prospective case series included all consecutive patients with CSM who underwent cervical decompressive surgery from December 2005 to October 2007 at the Toronto Western Hospital. Baseline Data and Potential Covariates The baseline data included age at the time of surgery, sex, period of time since symptom onset, date of surgical intervention, and preexisting medical comorbidities at hospital admission. Preexisting medical comorbidities were quantified using the number of ICD-9 codes and the CCI. 8 Other potential covariates for postoperative complications included surgical approach and the number of spinal levels treated. Development of postoperative complications was also included as a potential predictor of functional recovery after surgical treatment for CSM. Outcome Measures Primary outcome measures included results of 3 disease-specific instruments for functional assessments before surgery and both 6 months and 1 year after decompressive surgery: Nurick grade, mjoa score, and BBS. The frequency of postoperative complications was included as a secondary outcome measure. The Nurick grading system was developed to assess walking ability of patients with CSM, who could be classified into 6 grades from 0 ( signs or symptoms of root involvement but without evidence of spinal cord disease ) to 5 ( chairbound or bedridden ). Nurick grading was reported as a suitable outcome measure in prior studies focused on its reliability and construct validity in the CSM population. 16,25,26 Although the MCID of the Nurick grading, to our best knowledge, has never been documented, J Neurosurg: Spine / Volume 14 / March 2011 the construct of this 6-grade scale suggests that a 1-grade reduction would represent a clinically relevant difference for patients ability to walk. Of note, MCID refers to the observed change measured by an outcome measure in a particular population deemed to have had an important neurological improvement or deterioration. The Japanese Orthopaedic Association criteria, which were reportedly reliable and valid, were modified by Benzel et al. 3,15,25 The mjoa score varies from 0 to 18 and reflects 4 body functions, including arm function, leg function, micturition, and hand sensibility. 3 Based on expert opinion, a difference of at least 2 points between the postoperative and preoperative mjoa scores was considered clinically important. 2 The BBS was developed to measure static and dynamic standing balance among older people with impairment. 4 A 14-item test, it uses ordinal scoring from 0 to 4 for each item, to give a total score of Previous studies reported excellent reliability of the BBS, with intraclass correlation coefficients ranging from 0.81 to ,10,27,28 Various prior studies also indicated that the BBS has validity and responsiveness in the assessment of body balance. 22,24,28 Although the estimated minimum detectable change for the BBS is 5 points, Conradsson et al. indicated that the MCID for the BBS is 8 points. 10,27 Statistical Analysis The effects of potential confounders were analyzed using 3 strategies. First, univariate analyses were carried out using a 2-sided Fisher exact test for categorical variables and 2-sided Mann-Whitney U-test or ANOVA with Bonferroni post hoc test for continuous variables. Second, the best models for prediction of functional status at 1 year following decompressive spine surgery were selected from the analysis of all the potential confounders, including preoperative functional status, age, sex, preoperative number of ICD-9 codes, duration of preoperative symptoms, and postoperative complications. A stepwise multivariate linear regression was used to determine those models that included only significant variables and the highest R 2 regression. 30 Finally, stepwise logistic regression analyses were performed to test the association of potential confounders with clinically relevant improvement in the functional outcome measures. In those regression analyses, the best models for prediction of functional status at the minimum level of the MCID for Nurick grading (1 grade), mjoa score (2 points), and BBS (8 points) at 1 year after decompressive spinal surgery were selected from the analysis of all potential confounders. Only patients with the potential to have a functional improvement greater than the MCID were included in the regression analyses. All data analysis was carried out using SAS program version 9.1 (SAS Institute, Inc.). Values were represented in terms of means ± SEMs. Results During the study period, 24 women and 57 men with ages ranging from 32 to 88 years underwent surgical spinal cord decompression for treatment of CSM (Table 1). 349

3 J. C. Furlan et al. The majority of those patients had no relevant preexisting medical comorbidities, but they showed different degrees of functional deficits due to CSM, with symptoms for over 2 years on average. The mean follow-up time was 10.1 months (range 6 12 months). A 6-month follow-up assessment was performed in 87.65% of the cases and a 12-month follow-up evaluation in 80.25%. Clinical Outcome After Surgery Postoperative complications occurred in 15 (18.52%) of 81 patients. The majority of the postoperative complications (12 of 15) resolved without residual effects; 3 patients had residual sequelae of the complications within the first year after surgery (Table 2). Only 4 patients (4.94%) showed neurological deterioration during follow-up after decompressive spinal surgery. Based on the Nurick grading, neurological deterioration of at least 1 grade within the first year after surgical treatment was observed in 3 patients (3.7%) all of whom developed symptomatic tandem lumbar stenosis with neurogenic claudication. None of the patients died during the first year after surgery. In the univariate analyses, those patients who had postoperative complications were significantly older (p = 0.006) and had more ICD-9 codes (p = 0.033) than the patients who showed an unremarkable postoperative recovery (Table 3). There was a trend for a greater CCI among patients with postoperative complications compared with patients without postoperative complications, but this did not reach significance (p = 0.092). Patient sex, the duration of symptoms, and the number of spine levels treated were not significantly associated with postoperative complications. The anterior approach for decompressive spinal surgery was associated with a significantly lower frequency of postoperative complications than the posterior approach or the combined anterior and posterior approach (p = 0.018). However, the stepwise logistic regression analysis only identified age as a significant confounder for postoperative complications (OR 1.09 [95% CI ]; p = 0.018; -2logL = ; c = 0.733). Postoperative complications were not significantly associated with patient sex (p = 0.188), number of ICD-9 codes (p = 0.113), duration of symptoms (p = 0.309), surgical approach (p = 0.248), or number of spine levels treated (p = 0.454). Functional Outcomes After Surgery The mean Nurick grades at 6 months and 1 year after surgery were significantly lower than the mean preoperative Nurick grade (p < ; Fig. 1A). However, there were no significant differences between the mean Nurick grades at 6 months and 1 year after surgery (Fig. 1A). The mean mjoa scores at 6 months and 1 year after surgery were significantly higher than the mean preoperative mjoa score (p < ; Fig. 1B). Nonetheless, the mean postoperative mjoa score at 6 months did not significantly differ from the mean mjoa score 1 year after surgery (Fig. 1B). Also, the mean BBS at 6 months and at 1 year after surgery significantly increased in comparison with the mean preoperative BBS (p = ), whereas there was no significant difference between the BBS at 6 months and 1 year following surgery (Fig. 1C). TABLE 1: Characteristics of 81 patients with CSM* Characteristic Value age at op (yrs) mean ± 1.36 median 55 range sex (M/F ratio) 2.38 CCI no. of ICD-9 codes on admission time from symptom onset to op (mos) mean duration ± 2.7 range Nurick grading Grade 1 1 Grade 2 22 Grade 3 48 Grade 4 9 Grade 5 1 mjoa score mean ± 0.34 range 3 18 BBS score mean 44.5 ± 1.94 range 0 56 surgical approach anterior approach 56 posterior approach 23 combined anterior & posterior approach 2 no. of spine levels treated surgically mean 2.65 ± 0.24 median 2 range 1 6 * Values represent numbers of patients unless otherwise indicated. Means are provided with SEMs. The functional assessments using Nurick grade, mjoa score, and BBS score before and after surgery were significantly associated with patient age and number of ICD-9 codes (Tables 4 6). In contrast, the functional assessments using Nurick grade, mjoa score, and BBS score before and after surgery were not significantly associated with duration of symptoms. Although the CCI was significantly associated with preoperative and postoperative functional status at 6 months in most cases, there was no significant correlation between the CCI and functional assessment at 1 year after surgery regardless of the func- 350 J Neurosurg: Spine / Volume 14 / March 2011

4 Outcomes of surgery for cervical spondylotic myelopathy TABLE 2: Complications after decompressive spinal surgery in 15 patients Complications & Surgical Approach Long-Term Consequences Management* neurological complications after anterior approach new-onset (postop) radiculopathy continuing conservative epidural hematoma resolved, no residual effects conservative C-5 radiculopathy resolved, residual effects surgical intraop bleeding resolved, no residual effects surgical after posterior approach CSF leak resolved, no residual effects conservative nonneurological complications after anterior approach chest pain/angina continuing concomitant dysphagia resolved, no residual effects conservative after posterior approach acute on chronic renal failure resolved, no residual effects conservative atrial fibrillation resolved, no residual effects conservative cardiopulmonary event resolved, no residual effects conservative facial swelling resolved, no residual effects conservative superficial wound infection resolved, no residual effects conservative superficial wound infection resolved, no residual effects conservative unresponsive episode resolved, no residual effects conservative after anterior & posterior superficial wound infection & dysphagia resolved, no residual effects conservative * Conservative management means nonsurgical treatment. tional outcome measure. Complications at 6 months and 1 year after surgery were significantly associated with lower mjoa scores, but postoperative complications were not significantly associated with functional assessments as assessed using Nurick grading and the BBS. Our stepwise multivariate regression analyses indicate that the best models of prediction of postoperative functional status adjusted for preoperative functional status consistently include patient age as a significant confounder (Table 7). In addition to the adverse effects of older age on functional status, a greater number of ICD-9 codes was also significantly correlated with lower mjoa score at 1 year after surgery. Using stepwise logistic regression analysis, function- TABLE 3: Univariate analysis of potential confounders for complications after decompressive spinal surgery for CSM in our series of 81 patients* Potential Confounders Patients w/ Postop Complications (15 patients) Patients w/o Postop Complications (66 patients) p Value mean age at op (yrs) ± ± sex F 3 (12.5) 21 M 12 (21.05) 45 mean duration of preop symptoms (mos) ± ± mean CCI 0.33 ± ± mean no. of ICD-9 codes preop 0.8 ± ± surgical approach anterior 6 (10.71) 50 posterior 8 (34.78) 15 combined anterior & posterior 1 (50) 1 mean no. of levels treated 2.55 ± ± * Values represent numbers of patients (%) unless otherwise indicated. Means are provided with SEMs. J Neurosurg: Spine / Volume 14 / March

5 J. C. Furlan et al. TABLE 4: Univariate analysis of potential confounders for functional assessments using Nurick grading* Potential Confounders R 2 Preop 6 mos Postop 1 yr Postop p p p Value R 2 Value R 2 Value sex age at op duration of preop symptoms CCI preop no. of < ICD-9 codes postop complications NA NA * NA = not applicable. Fig. 1. Functional recovery at 6 months and 1 year following decompressive spine surgery for CSM as assessed by Nurick grade (A), mjoa score (B), and BBS score (C). * Significant difference based on an ANOVA with the Bonferroni post hoc test. al improvement of at least 1 degree in the Nurick grading from preoperative to postoperative assessment at 1 year following decompressive spine surgery was not significantly associated with age (p = 0.478), sex (p = 0.193), number of ICD-9 codes (p = 0.363), duration of symptoms (p = 0.555), or postoperative complications (p = 0.326). In the stepwise logistic regression analysis, mjoa score at 1 year after surgery as dichotomized according to its MCID was not significantly associated with age (p = 0.624), sex (p = 0.331), preoperative number of ICD- 9 codes (p = 0.12), duration of preoperative symptoms (p = 0.863), or postoperative complications (p = 0.235). Finally, the stepwise logistic regression analysis for the BBS score at the level of MCID indicated that the BBS score at 1 year after decompressive spinal surgery was not significantly associated with age (p = 0.919), sex (p = 0.278), number of ICD-9 codes (p = 0.241), duration of preoperative symptoms (p = 0.43), or postoperative complications (p = 0.278). Discussion The results of our univariate analyses suggest that older patients and those with a greater number of preexisting medical comorbidities who undergo posterior decompressive spinal surgery for CSM are more susceptible to postoperative complications. However, only patient age was significantly correlated with postoperative complications in the multiple logistic regression analysis. The postoperative functional status at 6 months and 1 year after surgery was significantly improved in comparison with the preoperative functional status regardless of the functional outcome measure. In the univariate analyses, older patients who had more ICD-9 codes before surgery had significantly poorer functional recovery as assessed by Nurick grade, mjoa score, and BBS score. In our stepwise multivariate regression analyses adjusted for preoperative functional status, older age was also significantly associated with poorer functional status at 1 year after decompressive spine surgery. However, none of the potential confounders were significantly associated with clinically relevant functional improvement as assessed by Nurick grade, mjoa score, or BBS score when MCID was taken into account. Complications After Decompressive Spinal Surgery Our postoperative complication rate was 18.52%, but only 4.94% of our patients who underwent decompressive spinal surgery for CSM had a neurological complication. Similarly, previous studies reported overall complication rates ranging from 5% to 47.5% 6,14,17,23,31,34 and neurological complication rates between 0.2% and 15%. 6,14,17,23,31,34 Furthermore, the proportion of patients with neurological deterioration after surgery in our study (3.7%) was consistent with the findings of prior investigations that documented neurological deterioration rates ranging from 1.08% to 29.8%. 12,14,23,33 Our univariate analyses identified older patients with greater numbers of preexisting medical comorbidities who undergo posterior decompressive spinal surgery for CSM as the group at risk for postoperative complications. Of note, patients with focal anterior pathology (generally 1- or 2-level spondylosis) were treated using anterior techniques. Patients with multilevel pathology were treated 352 J Neurosurg: Spine / Volume 14 / March 2011

6 Outcomes of surgery for cervical spondylotic myelopathy TABLE 5: Univariate analysis of potential confounders for functional assessments using mjoa scores Potential Confounders Preop 6 Mos Postop 1 Yr Postop Mean R 2 p Value Mean R 2 p Value Mean R 2 p Value sex M ± ± ± 0.41 F ± ± ± 0.34 age at op < duration of preop symptoms CCI preop no. of ICD-9 codes < postop complications yes NA NA NA 13 ± ± 1.01 no NA NA NA ± ± 0.3 with a posterior approach (in the absence of kyphotic deformity). Patient sex and duration of symptoms were not significantly associated with postoperative complications in our univariate analyses. However, only age was significantly correlated with postoperative complications in our stepwise logistic regression analysis. Prior studies on complications after lumbar spine surgeries also indicated that older patients are more susceptible to postoperative complications. 7,11,20,21 Using data obtained in 58,115 patients with CSM, Boakye et al. 6 have recently reported that complications after spine fusion were significantly more frequent among older patients with at least 3 preexisting medical comorbidities who underwent nonelective spine operation in a larger hospital. Given that all our patients were surgically treated in a tertiary spine center, some differences between our results and the study of Boakye et al. 6 could be attributed to discrepancies among hospitals regarding the quality of health care services and professional expertise. Also, the substantial differences between the studies in terms of sample sizes may explain the negative results regarding the effects of comorbidities on postoperative complication in our multiple logistic regression analysis. Given that our data analysis had a power of at least 80% and 5% significance, the potential interaction between older age and greater number of preexisting medical comorbidities should be taken into consideration when comparing our results and the findings reported by Boakye et al. Functional Outcomes After Decompressive Spine Surgery In our study functional recovery reached a plateau at 6 months after decompressive spine surgery in comparison with the 1-year follow-up assessments. This finding is similar to the findings of a prior prospective study that assessed functional recovery using the JOA score within the first year after surgical decompression of the spinal cord in patients with CSM. 9 These data suggest that the most relevant functional improvements occur within the first 6 months after surgical treatment in this group of patients. Although the results of our univariate analyses indicated that older patients with more preoperative ICD- 9 codes had significantly worse Nurick grades, mjoa scores, and BBS scores before and after spine surgery for CSM, only older age was significantly associated with the poorer functional status at 1 year after decompressive spine surgery in the multivariate regression analyses that also included sex, duration of preoperative symptoms, and postoperative complications as potential confound- TABLE 6: Univariate analysis of potential confounders for functional assessments using BBS scores Potential Confounders Preop 6 Mos Postop 1 Yr Postop Mean R 2 p Value Mean R 2 p Value Mean R 2 p Value sex M ± ± ± 1.23 F ± ± ± 1.21 age at op < < < duration of preop symptoms CCI preop no. of ICD < < codes postop complications yes NA NA NA ± ± 2.92 no NA NA NA ± ± 0.95 J Neurosurg: Spine / Volume 14 / March

7 J. C. Furlan et al. TABLE 7: Results of the stepwise multivariable regression analysis* Model R 2 p Value C (p) Dependent Variable Independent Variable Parameter Estimate F Value p Value Model Nurick grade preop Nurick grade age Model < mjoa score preop mjoa score age preop no. of ICD-9 codes Model < BBS score preop BBS score age * Including age, sex, preoperative number of ICD-9 codes, duration of preoperative symptoms, and postoperative complications as potential predictors for Nurick grade at 1 year after surgery adjusted for the preoperative Nurick grade (Model 1), mjoa score at 1 year adjusted for the preoperative mjoa score (Model 2), and BBS score at 1 year adjusted for the preoperative BBS score (Model 3). ers. However, none of those potential confounders were significantly associated with clinically relevant functional improvement as assessed by the Nurick grading, mjoa score, and BBS in our study. The factors associated with improvement of impairment and disability after surgical interventions for CSM are not completely known. Although a number of prior studies reported no significant differences between younger and older patients with CSM regarding postoperative functional status as assessed by mjoa and Nurick grading, 9,12,14,32 other investigators documented that older patients had poorer postoperative functional recovery in comparison with younger patients after spine surgery for CSM. 9,33 More recently, Lu et al. 19 showed that patients aged 70 years or older had significantly lower preoperative and postoperative mjoa scores than younger individuals, but there were no significant differences between the patient groups regarding their recovery rates. Whereas Ebersold et al. 12 reported that the duration of preoperative symptoms was associated with neurological deterioration after surgical treatment for CSM, Saunders et al. 23 and Yamazaki et al. 32 found no effect of duration of symptoms on postoperative functional recovery. In addition, postoperative functional recovery was reportedly unaffected by sex, 9,14 the number of levels decompressed, 12,14 or preoperative myelopathy severity 12,14,23 in prior studies. Study Limitations Although a prospective study holds advantages in overcoming some of the shortcomings of previous retrospective investigations in the field of CSM, there are still a few limitations in our study. One of the most relevant limitations is the lack of a gold standard and a reliable, validated outcome measure for assessment of impairment and disability in patients with CSM. The Nurick grading system and JOA scale and its modified versions have been widely used for assessment of patients with CSM, but there is a paucity of studies examining their psychometric properties. In addition, those outcome measures appear to include items related to body functions and structures as well as activity in the same instruments. 29 Despite the potential effects of patients age and perhaps other clinical factors on the outcome after treatment of CSM, the actual impact of any treatment requires an adequate analysis of the clinically relevant change. However, to our knowledge, the MCID for Nurick grade, mjoa score, and BBS score was not assessed in this particular patient population. Finally, the heterogeneity of this population indicates that stratifications in the data analyses would be important, but this requires a much larger sample size for appropriate analysis. Conclusions Our results indicate that functional recovery appears to reach a plateau at 6 months after cervical decompressive surgery in patients with CSM, regardless of the functional assessment instrument used. The number of ICD-9 codes appears to be more reliable than the CCI for assessment of preexisting medical comorbidities in the CSM population. Complications after decompressive spine surgery are infrequent and, when they do occur, are generally temporary. Older patients appear to be at a greater risk for postoperative complications. Although older patients and those with a greater number of preexisting medical comorbidities had less favorable functional outcomes after surgical treatment for CSM in the univariate analysis, none of the studied factors was associated with clinically relevant functional recovery after surgery. Although age and preexisting medical comorbidities do not appear to have a clinically relevant effect on postoperative outcomes, further investigations are required to confirm those results. Disclosure This work was supported by funds of the Lawson Fellow- Neurology from The Toronto General & Western Hospital Foundation (J.C.F.), the Krembil Chair in Neural Repair and Regeneration (M.G.F.), and AOSpineNA (M.G.F.). The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. Author contributions to the study and manuscript preparation include the following. Conception and design: Furlan, Kalsi- Ryan, Kailaya-Vasan, Massicotte. Acquisition of data: Kalsi-Ryan, Kailaya-Vasan, Massicotte, Fehlings. Analysis and interpretation of data: Furlan. Drafting the article: Furlan. Critically revising the article: all authors. Reviewed final version and approved it for 354 J Neurosurg: Spine / Volume 14 / March 2011

8 Outcomes of surgery for cervical spondylotic myelopathy submission: all authors. Statistical analysis: Furlan. Administrative/ technical/material support: Kalsi-Ryan, Kailaya-Vasan, Massicotte, Fehlings. Study supervision: Furlan, Kalsi-Ryan, Fehlings. References 1. Baron EM, Young WF: Cervical spondylotic myelopathy: a brief review of its pathophysiology, clinical course, and diagnosis. Neurosurgery 60 (1 Suppl 1):S35 S41, Bartels RH, Verbeek AL, Grotenhuis JA: Design of Lamifuse: a randomised, multi-centre controlled trial comparing laminectomy without or with dorsal fusion for cervical myeloradiculopathy. BMC Musculoskelet Disord 8:111, Benzel EC, Lancon J, Kesterson L, Hadden T: Cervical laminectomy and dentate ligament section for cervical spondylotic myelopathy. J Spinal Disord 4: , Berg K, Wood-Dauphinée S, Williams JI, Gayton D: Measuring balance in the elderly: preliminary development of an instrument. Physiother Can 41: , Berg KO, Wood-Dauphinee SL, Williams JI, Maki B: Measuring balance in the elderly: validation of an instrument. Can J Public Health 83 (Suppl 2):S7 S11, Boakye M, Patil CG, Santarelli J, Ho C, Tian W, Lad SP: Cervical spondylotic myelopathy: complications and outcomes after spinal fusion. Neurosurgery 62: , Carreon LY, Puno RM, Dimar JR II, Glassman SD, Johnson JR: Perioperative complications of posterior lumbar decompression and arthrodesis in older adults. J Bone Joint Surg Am 85-A: , Charlson ME, Pompei P, Ales KL, MacKenzie CR: A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 40: , Cheung WY, Arvinte D, Wong YW, Luk KD, Cheung KM: Neurological recovery after surgical decompression in patients with cervical spondylotic myelopathy a prospective study. Int Orthop 32: , Conradsson M, Lundin-Olsson L, Lindelöf N, Littbrand H, Malmqvist L, Gustafson Y, et al: Berg balance scale: intrarater test-retest reliability among older people dependent in activities of daily living and living in residential care facilities. Phys Ther 87: , Deyo RA, Cherkin DC, Loeser JD, Bigos SJ, Ciol MA: Morbidity and mortality in association with operations on the lumbar spine. The influence of age, diagnosis, and procedure. J Bone Joint Surg Am 74: , Ebersold MJ, Pare MC, Quast LM: Surgical treatment for cervical spondylitic myelopathy. J Neurosurg 82: , Emery SE: Cervical spondylotic myelopathy: diagnosis and treatment. J Am Acad Orthop Surg 9: , Fessler RG, Steck JC, Giovanini MA: Anterior cervical corpectomy for cervical spondylotic myelopathy. Neurosurgery 43: , Hukuda S, Mochizuki T, Ogata M, Shichikawa K, Shimomura Y: Operations for cervical spondylotic myelopathy. A comparison of the results of anterior and posterior procedures. J Bone Joint Surg Br 67: , King JT Jr, Roberts MS: Validity and reliability of the Short Form-36 in cervical spondylotic myelopathy. J Neurosurg 97 (2 Suppl): , Lad SP, Patil CG, Berta S, Santarelli JG, Ho C, Boakye M: National trends in spinal fusion for cervical spondylotic myelopathy. Surg Neurol 71:66 69, Lees F, Turner JW: Natural history and prognosis of cervical spondylosis. BMJ 2: , Lu J, Wu X, Li Y, Kong X: Surgical results of anterior corpectomy in the aged patients with cervical myelopathy. Eur Spine J 17: , Oldridge NB, Yuan Z, Stoll JE, Rimm AR: Lumbar spine surgery and mortality among Medicare beneficiaries, Am J Public Health 84: , Raffo CS, Lauerman WC: Predicting morbidity and mortality of lumbar spine arthrodesis in patients in their ninth decade. Spine 31:99 103, Riddle DL, Stratford PW: Interpreting validity indexes for diagnostic tests: an illustration using the Berg balance test. Phys Ther 79: , Saunders RL, Bernini PM, Shirreffs TG Jr, Reeves AG: Central corpectomy for cervical spondylotic myelopathy: a consecutive series with long-term follow-up evaluation. J Neurosurg 74: , Shumway-Cook A, Baldwin M, Polissar NL, Gruber W: Predicting the probability for falls in community-dwelling older adults. Phys Ther 77: , Singh A, Crockard HA: Comparison of seven different scales used to quantify severity of cervical spondylotic myelopathy and post-operative improvement. J Outcome Meas 5: , Singh A, Crockard HA, Platts A, Stevens J: Clinical and radiological correlates of severity and surgery-related outcome in cervical spondylosis. J Neurosurg 94 (2 Suppl): , Steffen T, Seney M: Test-retest reliability and minimal detectable change on balance and ambulation tests, the 36-item short-form health survey, and the unified Parkinson disease rating scale in people with parkinsonism. Phys Ther 88: , Stevenson TJ: Detecting change in patients with stroke using the Berg Balance Scale. Aust J Physiother 47:29 38, Stucki G: International Classification of Functioning, Disability, and Health (ICF): a promising framework and classification for rehabilitation medicine. Am J Phys Med Rehabil 84: , Tabachnick BG, Fidell LS: Using Multivariate Statistics, ed 4. Boston: Allyn & Bacon, Wang MC, Chan L, Maiman DJ, Kreuter W, Deyo RA: Complications and mortality associated with cervical spine surgery for degenerative disease in the United States. Spine 32: , Yamazaki T, Yanaka K, Sato H, Uemura K, Tsukada A, Nose T: Cervical spondylotic myelopathy: surgical results and factors affecting outcome with special reference to age differences. Neurosurgery 52: , Yonenobu K, Hosono N, Iwasaki M, Asano M, Ono K: Neurologic complications of surgery for cervical compression myelopathy. Spine 16: , Zeidman SM, Ducker TB, Raycroft J: Trends and complications in cervical spine surgery: J Spinal Disord 10: , 1997 Manuscript submitted December 28, Accepted October 24, This study was presented at the 8th Annual Meeting of the Canadian Spine Society, Sun Peaks, British Columbia, March Please include this information when citing this paper: published online January 14, 2011; DOI: / SPINE Address correspondence to: Julio C. Furlan, M.D., Ph.D., 399 Bathurst Street, McL , Toronto, Ontario, Canada M5T 2S8. jcfurlan@gmail.com. J Neurosurg: Spine / Volume 14 / March