Cervical myelopathy is generally considered to be a surgical disorder due to its progressive natural history. Once diagnosed, surgery is commonly recommended because such intervention has been shown to arrest progression and may improve neurological outcomes in many cases. In those patients who cannot or choose not to seek surgical intervention, however, little is known about the efficacy of active nonoperative treatment and whether it is advisable compared with observation alone. Nonsurgical care is also sometimes recommended for patients with milder forms of cervical myelopathy, especially when there are concurrent symptoms of neck pain or radiculopathy, which may be thought to benefit from nonoperative modalities such as physical therapy, medications, spinal injections, and orthoses. Another clinical concern centers on whether patients with known myelopathy or asymptomatic cervical stenosis should alter their lifestyles with the intent to prevent neurological progression or injury.
The purpose of this systematic review was to address the following key questions (KQs): (1) What is the evidence of the efficacy, effectiveness, and safety of nonoperative treatment of patients with cervical myelopathy? (2) Do the outcomes of nonoperative treatment vary according to severity of myelopathy? (3) Are minor injuries associated with neurological deterioration among patients with cervical myelopathy or asymptomatic cervical cord compression treated nonoperatively?
MATERIALS AND METHODS
Electronic Literature Search
For KQs 1 to 3, we conducted a systematic search in PubMed and the Cochrane Collaboration Library for literature published between January 1956 and November 20, 2012, on nonoperative management of cervical myelopathy. The search results were limited to human studies published in the English language. Reference lists of key articles were also systematically checked to identify additional eligible articles. For KQs 1 and 2, we included studies comparing nonsurgical treatment with surgical intervention and different nonsurgical treatment modalities with other nonsurgical modalities in patients with cervical spondylotic myelopathy (CSM) or cervical myelopathy due to ossification of the posterior longitudinal ligament (OPLL) (Table 1). For KQ 3, we included studies evaluating the effects of activities or mild injuries on neurological deterioration in patients with asymptomatic cervical cord compression or cervical myelopathy. We excluded studies of patients with myelopathy as a result of major trauma, tumor, infection, inflammatory arthritis, and neurodegenerative diseases; studies reporting only surgical outcomes; studies that included less than 10 patients; and animal, cadaver, and biomechanical studies. Two investigators (J.R.D., C.G.E.) reviewed the full texts of potential articles meeting the inclusion criteria to obtain the final collection of included studies (Figure 1).
For KQ 3, we sought to evaluate studies of a cohort of patients with asymptomatic cervical cord compression or symptomatic myelopathy, comparing the risk of neurological deterioration between those involved and those not involved in various activities that are perceived to be a risk for cervical trauma. We also included studies that identified groups by outcome (those with and without neurological deterioration) and evaluated the groups to determine the presence or absence of various activities or mild trauma.
From the included articles, the following data were extracted: study design, patient demographics, follow-up duration and the percentage of follow-up for each treatment group, initial neurological assessment and outcomes for the nonoperative and surgery treatment groups, and the type of treatment and outcomes reported. We attempted to identify studies with overlapping data and reported only the data from the most complete study (largest sample size) in order to prevent double counting.
Study Quality and Overall Strength of Body of Literature
Class-of-evidence ratings were assigned to each article independently by 2 reviewers (J.R.D., C.G.E.) using criteria set by The Journal of Bone & Joint Surgery1 for therapeutic and prognostic studies and modified to delineate criteria associated with methodological quality and risk of bias on the basis of 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 epidemiologic principles with characteristics of study design to determine the class of evidence and is consistent with those used in previous focus issues.4 (See Supplemental Digital Content material, available at http://links.lww.com/BRS/A830, for 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 Grading of Recommendation Assessment, Development and Evaluation Working Group5,6 and recommendations made by the Agency for Healthcare Research and Quality.2,3 Qualitative analysis is performed considering Agency for Healthcare Research and Quality–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 (1) inconsistency of results, (2) indirectness of evidence, (3) imprecision of the effect estimates (e.g., wide confidence intervals), 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 that the effect size lies close to the true effect and the extent to which it is thought to be stable based on the adequacy or deficiencies in the body of evidence. An overall strength of “high” means that we are very confident that the true effect lies close to that of the estimated effect. A “moderate” rating means that we are moderately confident in the effect estimate; the true effect is likely to be close to the estimated effect, but there is a possibility that it is substantially different. An overall strength of “low” means that our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate. Finally, a rating of “insufficient” means that we have very little confidence in the effect estimate: The true effect is likely to be substantially different than the estimated effect. In addition, this rating may be used if there is no evidence or it is not possible to estimate an effect.
For KQs 1 and 2, we reported the results as recorded in individual article (means and proportions). For KQ 3, we calculated risks and crude relative risks and their 95% confidence intervals of neurological deterioration when data were present, using the CSI command in Stata 9.1 (StataCorp, College Station, TX).8
Clinical Recommendations and Consensus Statements
Consensus statements were made through a modified Delphi approach by applying the Grading of Recommendation Assessment, Development and Evaluation/Agency for Healthcare Research and Quality criteria that impart a deliberate separation between the strength of the evidence (i.e., high, moderate, low, or insufficient) and the strength of the recommendation. When appropriate, recommendations or statements “for” or “against” were given “strong” or “weak” designations on the basis of 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 current focus issue Methods article.
For KQs 1 and 2, we identified 3 studies yielding 4 publications from our search strategy that met the inclusion criteria (Figure 1). For KQ 3, we identified 2 studies that met our inclusion criteria. Our initial search produced 54 possible studies for KQs 1 and 2 and 252 studies for KQ 3. After title and abstract review, we excluded 38 studies for KQs 1 and 2 and 242 studies for KQ 3, the majority of which did not include a comparison of conservative or nonoperative treatment to surgery or did not contain patients with myelopathy. After full text review of the KQ 1 and 2 studies, 12 were excluded for the following reasons: no comparisons (n = 5), no myelopathy (n = 3), historical controls (n = 2), review (n = 1), and less than 10 patients in a group (n = 1). After full text review of the KQ 3 studies, 8 were excluded for the following reasons: no myelopathy (n = 4), review (n = 2), and no outcomes followed over time (n = 2). A list of excluded articles can be obtained in the electronic Supplemental Digital Content material, available at http://links.lww.com/BRS/A830. The resulting 6 prospective and retrospective studies met our criteria for inclusion (Tables 2 and 3).
What Is the Evidence of Efficacy, Effectiveness, and Safety of Nonoperative Treatment of Patients With Cervical Myelopathy?
Nonoperative Treatment Versus Surgery
Among patients with milder degrees of CSM (defined here as modified Japanese Orthopaedic Association [mJOA] ≥12), 1 small randomized controlled trial (N = 68) compared nonoperative treatment (intermittent bed rest, use of collar, anti-inflammatory medication, and discouragement of high-risk activities) with surgery (anterior decompression, corpectomy, or laminoplasty) during a 10-year follow-up period.9,10 High-risk scenarios included heavy lifting, jumping and landing, and avoiding of risky environments such as physical overloading, cold temperatures, slippery surfaces, manipulation therapy, and significant amplitudes of cervical movement. The authors evaluated 4 outcomes: mJOA score, timed 10-m walk, and activities of daily living as scored by the clinician and by the patient (Table 4). The mean mJOA score remained relatively constant over time in both groups. There was no statistical difference between groups at any time period (Figure 2). The timed 10-m walk was significantly faster in the nonoperative group compared with the surgery group for each time point up through 3 years (Figure 3). There were no statistical differences between the nonoperative group and the surgery group with respect to activities of daily living as scored by the clinician or the patient 12 to 120 months after initiation of treatment. The article did not mention any specific cases of substantial neurological decline in either group during the follow-up period. However, during the 36-month follow-up period, 5 patients died in the surgery group, whereas there was no mention of deaths in the nonoperative group. The authors report that the deaths were physically unrelated to surgery. One criticism of this study is that the operative cohort did not improve postoperatively as much as patients reported in other published surgical series, which could be one potential reason for the lack of difference in outcomes versus the nonoperative cohort.
Two cohort studies compared nonoperative treatment with surgical treatment in patients with moderate to severe CSM. In a study by Sampath et al,11 the neurological rating was based on a 6-point scale depending on the presence of the following symptoms: pain in the upper extremities, arm numbness, arm weakness, headache, leg numbness, or difficulty walking (0 for no symptoms, 6 for all 6 symptoms). The baseline neurological rating for this study was 3.26 of 6 in the nonoperative group and 3.75 in the surgical group.11 In a study by Yoshimatsu et al,12 the baseline JOA score was 13 or greater in the nonoperative group and less than 13 in the surgical group. Both of these studies are rated as low quality because the treatment groups were decided by the severity of the initial diagnosis and/or patient preference rather than random allocation (confounding by indication), the severity of the patient population was not explicitly described, and neither study compared surgery with nonoperative care directly while controlling for potential confounders. These 2 studies are further described in the following 2 paragraphs.
The prospective cohort study by Sampath et al11 compared nonoperative with surgical treatment groups during a 1-year follow-up period (Table 5). They noted that surgically treated patients had improvement in overall pain and functional status (including work status rating, inability to work, social status rating, and opiate use), with nonsignificant improvement also observed in neurological symptoms compared with baseline (P > 0.05). Nonoperatively treated patients had a significant worsening of their ability to perform activities of daily living, with nonsignificant worsening of neurological symptoms compared with baseline. No comparisons were made between the surgical and nonoperative groups. However, the result that the surgical patients (who had more severe myelopathy at baseline) improved, whereas the nonoperative patients (who had less severe myelopathy at baseline) worsened, suggests that surgery may be more effective in treating cervical myelopathy.
In the retrospective study by Yoshimatsu et al,12 patients with CSM elected to undergo either immediate surgical treatment (32 patients) or nonoperative care (69 patients). Patients who chose to receive surgery had more severe CSM, with a mean JOA score of 9.1; those who received conservative treatment had milder forms of CSM, with a minimum JOA score of at least 13.12 Of those patients undergoing surgery, 29 received laminoplasty and 3 received anterior decompression and fusion. Of the 69 patients who elected nonsurgical care, 12 refused any treatment, 37 underwent rigorous nonoperative care, and 20 received nonrigorous care, which was not described. Those receiving immediate surgical intervention had worse disease as measured by the baseline JOA scale (mean JOA, 9.1) compared with those receiving nonoperative care (most with a JOA score of ≥13). In the early surgery group, 78% improved their JOA score at mean 29 months after surgery (Table 6). The proportion of patients in the nonoperative group who improved their JOA score over baseline was 23%. As with the study by Sampath et al,11 the finding that the surgically treated patients improved more than the nonoperative ones—despite the former having more severe myelopathy at baseline—suggests that surgery is more efficacious. However, baseline differences in treatment groups prevent well-defined comparisons.
Rigorous Versus Nonrigorous Nonoperative Treatment
The aforementioned study by Yoshimatsu et al12 also divided nonoperative patients (those with JOA score of ≥13) into rigorous versus nonrigorous treatment groups. As noted previously, those with more severe myelopathy (JOA score of <13) were treated surgically. The rigorous nonoperative treatment group received continuous cervical traction for 3 to 4 hours daily for 1 to 3 months. The cervical spine was immobilized with an orthosis in combination with drug therapy and exercise during nontraction hours. Outpatient orthosis, thermal therapy, and drug therapy continued for an unspecified duration after the traction regimen. Nonrigorous nonoperative care included 12 patients who refused treatment of any kind and 20 patients whose nonrigorous treatment was not defined. Among patients receiving rigorous nonoperative treatment, 38% of the patients improved compared with only 6% of the patients who did not undergo rigorous care. A total of 22 patients eventually underwent surgery, although the authors do not identify from which group these patients came. By comparison, 78% of the surgically treated patients improved, again suggesting that surgery leads to better neurological outcomes than nonoperative treatment.
Do the Outcomes of Nonoperative Treatment Vary According to Severity of Myelopathy?
We found no studies directly comparing outcomes of nonoperative treatment on the basis of the severity of myelopathy. However, as noted previously, studies suggest that those with more severe myelopathy have better outcomes with surgery than with nonoperative treatment, whereas those who have milder degrees of myelopathy may not benefit as much from surgery.
In addition, we could not identify evidence to define severity of disease in a consistent manner. In the aforementioned studies, a JOA score of 12 or greater is often defined as mild, with moderate and severe descriptors being used for progressive worsening function and/or neurological examination. The inconsistencies in the definition of disease severity also prevented an evidence-based assessment of KQ 2.
What Is the Evidence That Different Activities or Minor Injuries Alter the Risk of Neurological Deterioration Among Patients With Cervical Myelopathy or Asymptomatic Cervical Cord Compression Treated Nonoperatively?
We found no studies evaluating the effect of specific activities on myelopathy progression. However, 2 retrospective cohort studies examined the relationship between minor traumatic events and the development or worsening of myelopathy in patients with asymptomatic cervical stenosis or cervical myelopathy (Table 3).13,14 Bednarik et al13 looked at a group of patients with asymptomatic stenosis initially and, during a median follow-up of 3.7 years, examined the risk of developing CSM in those with and without trauma. Seven percent (1/14) of patients who reported a traumatic event to the head, spine, trunk, or shoulder region developed myelopathy. This is compared with 24% (44/185) of patients who reported no trauma who developed myelopathy (relative risk, 0.30; 95% confidence interval, 0.04–2.02; P = 0.151). In this study, therefore, the presence of trauma was not associated with progression of myelopathy, although this could be due to low sample size.
Another study evaluated 27 patients with OPLL with or without myelopathy who sustained minor trauma to the neck.14 Seven of 8 (87%) of the patients who had myelopathy prior to trauma had neurological deterioration as measured by the JOA. Among those who had no myelopathy prior to trauma, 68% developed myelopathy after mild trauma.
Nonoperative treatment of patients with milder degrees of CSM (JOA score of ≥12) was similar to surgical treatment with respect to a change in neurological status or daily activities (improvement or deterioration) but resulted in a faster 10-m walk 1, 2, and 3 years after treatment. The evidence for this is “low.” A strength of evidence of “low” means that our confidence in the percentages presented for these outcomes is limited, and the true effect may be substantially different from the estimates presented (Table 7).
In patients who have greater degrees of myelopathy, nonoperative treatment was less effective than surgery with respect to functional outcomes, pain, and neurological function. Neurological deterioration also seems more likely with nonoperative treatment.
There is insufficient evidence that
- rigorous nonoperative treatment results in different mJOA scores than nonrigorous treatment in patients with mild CSM (JOA score of ≥13);
- outcomes of nonoperative treatment vary according to severity of myelopathy; and
- activity or mild trauma increases the risk of developing or worsening cervical myelopathy;
- however, in patients with OPLL, minor trauma may be associated with neurological deterioration.
There is currently a paucity of high-level evidence to support the use of nonoperative treatment for cervical myelopathy. Regarding KQ 1 on the efficacy, effectiveness, and safety of nonoperative treatment, we identified 3 comparative studies meeting inclusion criteria, 2 retrospective cohort studies, and 1 small randomized controlled trial. The randomized controlled trial showed that nonoperative treatment may be associated with similar, or perhaps even slightly better, neurological outcomes in patients with mild myelopathy (JOA score of >13 at baseline). However, the strength of that evidence is low and must be weighed against the possibility of progressive neurological deterioration. Two of the included studies were subject to confounding by indication, because the treatment groups were allocated according to the severity of myelopathy or patient choice. These studies reported that surgery was associated with higher rates of neurological improvement, although the surgically treated patients had more severe myelopathy at baseline. Thus, surgery was superior in those with moderate to severe myelopathy. There were insufficient data on whether the various types of nonoperative treatment (e.g., chiropractic, bracing, spinal injections) affect outcomes, or whether active nonoperative treatment is better than observation alone. The literature does include anecdotal reports of neurological worsening after certain types of nonoperative management, such as manipulation, traction, and even massage.15,16 Although these reports do not meet inclusion criteria for analysis in this article, in light of the lack of proven benefit for these modalities and the potential for neurological worsening, caution should be exercised before prescribing them for patients with either asymptomatic cervical cord compression or myelopathy.
Regarding KQ 2, as to whether the outcomes of nonoperative treatment vary according to baseline severity, we could find no studies meeting inclusion criteria that addressed this question directly. However, based on the findings with respect to KQ 1, one might speculate that surgical treatment may be more effective as the severity of myelopathy increases, but there is no direct evidence to support that assertion.
Finally, regarding KQ 3 on whether specific activities or the occurrence of minor injuries alters the risk of neurological decline in patients being treated nonoperatively, the level of evidence in the 2 articles meeting inclusion criteria were insufficient. In 1 study, the percentage of patients who developed myelopathy was actually lower in those sustaining minor trauma. In the other study, those with myelopathy at baseline were more likely to worsen after injury than those who were asymptomatic at baseline, suggesting that a spinal cord that is already demonstrating clinical dysfunction is more sensitive to injury than a cord that is tolerating the current level of compression without symptoms. However, this study was performed in patients with OPLL and its findings may not necessarily apply to patients with myelopathy without OPLL. We could not find any studies directly comparing neurological outcomes in patients with myelopathy with restricted versus ad libitum activities.
Further studies are needed to determine the appropriate role of nonoperative treatment in cervical myelopathy more definitively. In particular, randomized studies comparing nonoperative with surgical treatment in those with milder degrees of myelopathy would be particularly helpful, as would trials comparing specific types of nonoperative treatments with the natural history of myelopathy. In addition, further studies directly comparing nonoperative with operative care are necessary to answer fundamentally important questions regarding safety, cost, mortality, neurological progression, and complications associated with each approach.
Evidence-Based Clinical Recommendations.
Recommendation 1. Because myelopathy is known to be a typically progressive disorder and there is little evidence that nonoperative treatment halts or reverses its progression, we recommend not routinely prescribing nonoperative treatment as the primary modality in patients with moderate to severe myelopathy.
Overall Strength of Evidence. Low
Strength of Recommendation. Strong
Recommendation 2. If there is a role for nonoperative treatment as a primary treatment modality, it may be in the patient with mild myelopathy. However, it is not clear which specific forms of nonoperative treatment provide any benefit over the natural history. If nonoperative treatment is selected, we suggest care be taken to observe for neurological deterioration.
Overall Strength of Evidence. Low
Strength of Recommendation. Weak
Recommendation 3. In those with asymptomatic spondylotic cord compression but no clinical myelopathy, the available literature neither supports nor refutes the notion that minor trauma is a risk factor for neurologic deterioration. We suggest that patients should be counseled about this uncertainty.
Overall Strength of Evidence. Low
Strength of Recommendation. Weak
Recommendation 4. In those with a clinical diagnosis of CSM but no OPLL, the available studies did not specifically address the issue of neurologic deterioration secondary to minor trauma. However, in those with underlying OPLL, trauma may be more likely to cause worsening of existing myelopathy or even initiate symptoms in those who were previously asymptomatic. We suggest that patients be counseled about these possibilities.
Overall Strength of Evidence. Low
Strength of Recommendation. Weak
- There is a paucity of evidence regarding the effectiveness of nonoperative treatment of cervical myelopathy.
- Given this overall lack of evidence and in light of the generally progressive nature of cervical myelopathy, nonoperative treatment cannot be routinely recommended.
Supplemental digital content is available for this article. Direct URL citations appearing in the printed text are provided in the HTML and PDF version of this article on the journal's web site (www.spinejournal.com).
1. Wright JG, Swiontkowski MF, Heckman JD. Introducing levels of evidence to the journal. J Bone Joint Surg Am 2003;85-A:1–3.
2. West S, King V, Carey TS, et al. Systems to Rate the Strength of Scientific Evidence. Rockville, MD: Agency for Healthcare Research and Quality; 2002. Evidence Report/Technology Assessment No. 47 (Prepared by the Research Triangle Institute-University of North Carolina Evidence-based Practice Center, Contract No. 290-97-0011).
3. Methods Guide for Effectiveness and Comparative Effectiveness Reviews. Rockville, MD; 2012.
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: 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. StataCorp. Stata Statistical Software: Release 9. College Station, TX: StataCorp LP; 2005.
9. Kadanka Z, Bednarik J, Novotny O, et al. Cervical spondylotic myelopathy: conservative versus surgical treatment after 10 years. Eur Spine J 2011;20:1533–8.
10. Kadanka Z, Mares M, Bednanik J, et al. Approaches to spondylotic cervical myelopathy: conservative versus surgical results in a 3-year follow-up study. Spine (Phila Pa 1976) 2002;27:2205–10; discussion 10–1.
11. Sampath P, Bendebba M, Davis JD, et al. Outcome of patients treated for cervical myelopathy. A prospective, multicenter study with independent clinical review. Spine (Phila Pa 1976) 2000;25:670–6.
12. Yoshimatsu H, Nagata K, Goto H, et al. Conservative treatment for cervical spondylotic myelopathy. Prediction of treatment effects by multivariate analysis. Spine J 2001;1:269–73.
13. Bednarik J, Sladkova D, Kadanka Z, et al. Are subjects with spondylotic cervical cord encroachment at increased risk of cervical spinal cord injury after minor trauma? J Neurol Neurosurg Psychiatry 2011;82:779–81.
14. Katoh S, Ikata T, Hirai N, et al. Influence of minor trauma to the neck on the neurological outcome in patients with ossification of the posterior longitudinal ligament (OPLL) of the cervical spine. Paraplegia 1995;33:330–3.
15. Cheong HS, Hong BY, Ko YA, et al. Spinal cord injury incurred by neck massage. Ann Rehabil Med 2012;36:708–12.
16. Hsieh JH, Wu CT, Lee ST. Cervical intradural disc herniation after spinal manipulation therapy in a patient with ossification of posterior longitudinal ligament: a case report and review of the literature. Spine (Phila Pa 1976) 2010;35:E149–51.