Research Review By Gary J. Maguire©


Download MP3

Date Posted:

November 2011

Study Title:

Effectiveness of manual physical therapy in the treatment of cervical radiculopathy: A systematic review


Boyles R, Toy P, Mellon J et al.

Author's Affiliations:

Program of Physical Therapy, University of Puget Sound, Tacoma, WA, USA

Publication Information:

Journal of Manual and Manipulative Therapy 2011; 19 (3): 135-142.

Background Information:

Cervical radiculopathy (CR) is the result of cervical nerve root pathology secondary to a space-occupying lesion, usually a cervical disc herniation, osteophyte or other spondolytic process. The annual incidence rate or CR is 83.2 per 100 000, and it most commonly affects the C6 and C7 nerve roots. Symptoms occur primarily in the upper extremity and can include pain, numbness, weakness or paresthesia. Pain generators can be of bony, ligamentous or neural origin within the cervical spine.

While EMG and/or nerve conduction velocity (NCV) testing are the current diagnostic gold standard to confirm the presence of CR, not all clinicians have ready access to these assessment tools. Therefore, Wainner et al. developed and published a clinical prediction rule consisting of 4 variables to assist clinicians in diagnosing CR (1):
  1. positive Spurling test;
  2. positive distraction test;
  3. ipsilateral cervical spine rotation (< 60°); and
  4. positive upper limb tension test A – median nerve bias.
If a patient is positive on 3 out of 4 of these variables the prediction rule has sensitivity rating of 94% and a positive likelihood ratio of 6.1. When all 4 variables are positive the specificity ratio increased to 100%, while the positive likelihood ratio increased to 30.3.

The optimal treatment approach for CR has always been a source of controversy. Recent evidence reveals that conservative treatment appears to be more effective than surgical options for CR (2). Such treatment typically includes some combination of therapeutic exercise, manual therapy, modalities, medication and cervical collars. The focus of this study was to perform a systematic review to evaluate the literature regarding the effectiveness of using manual physical therapy in the treatment of CR.

Pertinent Results:

While the four studies (meeting the PEDro score criteria) utilized manual therapy only one study performed by Ragonese utilized manual therapy as a stand-alone treatment approach for CR (3). The other three implemented manual therapy with multimodal treatment approaches including therapeutic exercises focusing on the cervical and or thoracic muscles.

Since there were only 4 papers included, we will summarize the findings of each study below:
  1. Persson et al. (5) split 110 subjects into 3 groups: surgery, physiotherapy (15 sessions over 3 months, 30-45 minutes each), and a rigid cervical collar group. The authors concluded that no significant differences in outcomes existed among the 3 groups. From this, they surmised that physical therapy (manual therapy and exercise) is at least as effective as surgery. The results also suggested that cervical collars may be a valid treatment, representing a more cost-effective option compared to the other treatments.
  2. In the study by Young et al. (4), the focus was to determine the effect of adding cervical traction to a treatment approach of manual therapy and therapeutic exercise. 81 subjects were split into two groups: each group received ~8 treatments including manual therapy and exercise over the course of ~4 weeks. One group received real intermittent traction and the second group received sham intermittent traction. Overall, there appeared to be no additional functional outcome. Subjects in each group demonstrated favorable benefit in both primary and secondary outcomes (NRPS as well as the NDI) after 4 weeks of treatment.
  3. Cleland et al. (6) conducted a descriptive outcome study on 96 subjects, who received physical therapy treatment at the discretion of the treating therapist (including thrust and non-thrust manipulation). They reported that 53% of their subjects experienced at least a minimal clinically important change (MCIC) on 4 outcome measures (PSFS, GROC, NPRS and NDI) when provided a multimodal approach (including manual therapy techniques). This study was deemed ‘high quality’, but lacked a control group as well as defined interventions. This resulted in difficulty analyzing treatment effectiveness and limiting the reproducibility of the results. (EDITOR’S NOTE: the goal of this study was to begin development of a clinical prediction rule for directing treatment for CR by identifying factors that predict a successful outcome. In that regard it seems like a strange inclusion in this paper although it did meet the inclusion criteria.)
  4. Ragonese (3) conducted a between group comparison on only 30 subjects. Overall, they were able to demonstrate statistically significant improvements in NPRS pain scores after 4 weeks of intervention (P < 0.01) and also improved NDI scores (P < 0.05). This was noted in all three treatment groups which were as follows: manual therapy only (cervical lateral glides, thoracic mobilizations, median nerve mobilizations – treatment was 3x/week for 3 weeks), therapeutic exercise only (strengthening of deep neck flexors, lower trapezius and serratus anterior) and a combination of both (this group had the best results of the 3 groups).

Clinical Application & Conclusions:

Each of the four included studies was able to demonstrate a reduction in pain with various forms of manual therapy, however only one study randomized patients to such treatment. Three of the four studies were able to measure statistically significant functional improvement (via the NDI). Range of motion was only measured in one study (Ragonese). This study reported equal and statistically significant improvement in cervical range of motion across all treatment groups (but remember, this was the trial with only 30 subjects).

The interesting part of this systematic review was that four different types of manual therapy techniques were used in the various studies consisting of: thrust mobilizations, non-thrust mobilizations, neural dynamic techniques and muscle energy techniques. Thoracic thrust mobilizations (performed in 2 studies) were effective with one reporting a 66.7% successful outcome (minimally clinically important change) at re-examination for NDI, PSFS, NPRS and GROC. When a thoracic thrust mobilization was directed at hypomobile mid to upper thoracic segments in CR patients the outcome was also statistically significant for the NDI, PSFS and NPRS.

Manual therapy involving non-thrust mobilizations of the cervical spine were also provided in all 4 studies. Techniques varied, but the most detailed study indicated that treatment included cervical lateral glides (grade III-IV) for 30-45 seconds for all segments of C2-C7 (in each treatment). One study also allowed therapists to include retractions, rotations, lateral glides in the ULTT1 position or P-A glides at their discretion.

Neural dynamic techniques (nerve ‘flossing techniques as described by Butler) also proved to be successful for treatment of CR. Two studies provided this treatment approach with one using a “sliding technique” (slow and oscillatory). With symptom reduction treatment was then progressed to a “tension” technique. One study reported a 56.5% successful outcome with either neural dynamic or neural mobilization techniques applied.

One study provided muscle energy techniques (no specific techniques provided) and was able to demonstrate a 46.4% successful outcome. This was classified using NDI, PSFS, NPRS and GROC at re-examination surpassing the minimal clinically important change criteria.

Of the four included studies, the one by Ragonese et al. was the only one that used manual therapy as a stand-alone intervention. This highlights a deficiency in the literature and more research is certainly required. The literature, as we see commonly, is limited overall by a lack of clear descriptions of the treatments used in these studies. Also, none of the included studies were randomized, another glaring deficit that could be improved upon. Having said all that, there is some promise for the role of manual therapy for CR – but the literature to date simply cannot definitively guide us at this point. Further research is required and we will keep you posted on pertinent updates!

Study Methods:

94 studies were identified through various database searches (Medline, CINAHL, Cochrane Library and Google Scholar). Studies could range from randomized trials to case series, but had to be conducted on adults > 18 yoa and evaluate the efficacy of a conservative treatment on patients with imaging-confirmed (CT, MRI or myelography) CR, or CR as indicated using the clinical prediction rule set out by Wainner and colleagues (see above). A total of 25 abstracts were screened for eligibility and 15 of these were deleted. Ten full-text articles were assessed for eligibility and of these, 6 were found to be insufficient. This resulted in the inclusion of four studies. Each study had to utilize at least one of the following outcome measures: active or passive range of motion, a functional outcome measure specific to the neck [Neck Disability Index (NDI) or Patient-Specific Functional Scale (PSFS)], a quality of life measure [Global Rating of Change (GROC)], or Sickness Impact Profile (SIP), and a pain measure [Numeric Pain Rating Scale (NPRS) or Visual Analogue Scale (VAS)].

Studies were ineligible is the included subjects had received any surgical intervention (< one year previous), manual procedures performed by clinicians outside of physical therapy, use of cervical collars, mechanical cervical traction, or any other external and/or mechanical devices.

The 11-item PEDro scale was used to evaluate all included studies. A cutoff of 5/11 was employed by the authors for studies to be included. Of the 4 that made it, the average PEDro score was 7.25 (range was 5-9). (EDITOR’S NOTE: remember that one of the items on the PEDro scale is ‘blinding of therapists’ which in manual therapy trials is nearly impossible – all trials in this review scored zero on this factor.)

Study Strengths / Weaknesses:

The authors correctly identified that the value of this study was limited to some degree because only one article specifically described their treatment intervention, while the other 3 studies allowed the clinician to direct treatment at their discretion (hence, the lack of specific descriptions). Another limitation was the absence of randomized control trials, which prevented the evaluation of the potential cause and effect relationship between manual therapy and reduction of CR symptoms. One other factor that may have improved the findings of this study would have to been to broaden the search criteria to include studies published in other languages pertaining to physical therapy intervention and CR.

This is one of the only studies of its kind (if not the only) and the authors should be commended for this. Manual therapy is supported by a small body of research as an effective intervention for the treatment of CR: improving AROM and function while reducing pain and disability. This provides a foundation for future research to focus on the use of specific interventions with RCTs to develop clear and effective treatment protocols.

Additional References:

  1. Wainner RS et al. Reliability and diagnostic accuracy of the clinical examination and patient self-report measures for cervical radiculopathy. Spine 2003; 28: 52-62.
  2. Costello M. Treatment of a patient with cervical radiculopathy using thoracic spine manipulation, soft tissue mobilization and exercise. J Man Manip Ther 2008; 16: 129-135.
  3. Ragonese J. A randomized trial comparing manual physical therapy to therapeutic exercises, to a combination of therapies, for the treatment of cervical radiculopathy. Orthop Prac 2009; 21 (3): 71-77.
  4. Young IA et al. Manual therapy, exercise, and traction for patients with cervical radiculopathy: a randomized clinical trial. Phys Ther 2009; 89: 632–42.
  5. Persson LC, Carlsson CA, Carlsson JY. Long lasting cervical radicular pain managed with surgery, physiotherapy, or a cervical collar. A prospective, randomized study. Spine 1997; 22(7): 751–8.
  6. Cleland JA, Fritz JM, Whitman JM, Heath R. Predictors of short-term outcome in people with a clinical diagnosis of cervical radiculopathy. Phys Ther 2007;87:1619–32.