Design
This was a 2-armed parallel randomized controlled trial for patients with cervical spondylotic radiculopathy accompanied with symptoms including neck pain, numbness, radiating symptoms on the upper extremity, etc.
Before starting this trial, we classified the technique into four categories, including pain-free, mild pain, pain and mobilization under anesthesia (MUA). For pain-free technique, therapists mobilized within physiological range and would not reach either the pain trigger point nor the stiffness point, so patients won’t feel any discomfort. For mild pain technique, it was performed under mild pain (Visual Analogue Scale; VAS score ≤ 3) that can be tolerated, and the discomfort would gone immediately once the force was removed. For pain group, a pain intensity of VAS score at 4 to 6 would be reached, and usually accompanied with upper extremity avoidance behaviors during mobilization with more soft tissue damages compared with mild pain group, which usually required dozens of minutes to recover. While for the MUA, techniques were conducted under anesthesia. In this study, we selected the first two categories, pain-free and mild pain, for further examination, which were more common in practice.
Participants
Participants were recruited in inpatient and outpatient departments from August 1st to July 31th in 2021. The inclusion criteria were: (1) patients diagnosed with cervical spondylotic radiculopathy with symptoms of neck pain accompanying with radiating symptoms and with diagnostic imaging confirmation; (2) patients aged between 18 and 60 years old; (3) patients with pain severity screened by VAS score ≥ 4/10; (4) patients who demonstrated positive results in nerve tension test and responded positively when applying the neurodynamic mobilization techniques; (5) patients agreed and provided written consent for treatment and follow-up process.
The exclusion criteria were: (1) patients who have been diagnosed with other diseases; (2) patients confirmed with serious neuro-psychological problems, such as Parkinson’s disease, stroke, anxiety, etc.; (3) patients have currently or already undergone any type of pain treatments like nonsteroidal anti-inflammatory drugs (NSAIDs), which may blind the pain severity; (4) patients participated in other physiotherapeutic treatment in last 3 months.
Sample size
G-power software was used for the calculation process. The sample size was calculated by the VAS scale, the primary outcome measure. According to previous research, t test was used for computer required sample size to achieve statistical power of 80%. Assuming an α error of 0.05, and the allocation ration of group 1/ group 2 was 1. Assuming a standard deviation of 9.8, and a mean difference of 10. 40 participants (20 per arm) were recruited in total.
Randomization and allocation concealment
participants were randomized into either pain-free group or mild pain group, and for each group, the number of patients was identical.
The randomization was performed by a statistician using the SPSS 22.0 computer-generated random numbers table, and each group included 20 individuals.
Before receiving treatment, participants were explained the neurodynamic mobilization in detail by researchers responsible for this study. Eligible participants were allocated to either group at the ratio of 1:1 by sealed envelopes.
Blinding
As participants and therapists performing the intervention were impossible to be blinded due to the classification of mobilization grades that was based on patients’ feedback, which cannot satisfy the blinding process of patients. And therefore, the blinding process was only performed in researchers who were responsible for patient evaluation and statistical analysis.
The evaluation was performed by the same evaluator throughout the whole program. Computerized statistics didn’t include any information relating to patients' allocation in either group.
Intervention
All participants with cervical spondylotic radiculopathy were examined both through imaging and functional tests to ensure they met the eligibility criteria. Each group received manual therapy, including joint mobilization, myofascial release technique, cervical distraction, and neurodynamic mobilization, together with modalities targeting patients’ pain or other symptoms if necessary. The patients were administered for a total of 10 sessions, twice a day for 5 days.
For the pain-free group, gliding technique was applied, and no increases in pain level was noticed while operating. The pain free group were mobilized under the physiological range where the patient feels no discomfort in the motion. Additionally, the mobilization never reached either the pain trigger point or the stiffness point, making sure the safety of tissues during the operation.
As for the mild pain group, using similar gliding technique mentioned above, the mobilization was performed under the range of a mild increase in pain that can be tolerated (≤ 3 scores on VAS scale), and the discomfort can get released immediately once the force was removed. When reaching the pain trigger point or stiffness point, a small range of mobilization was applied, which might increase the range of motion accompanied by some micro damages, such as small soft tissue injury. The neurodynamic mobilization of both groups were performed 6–8 repetitions per set, with 3 sets per session, two sessions per day, and after each set, a one-minute pause for a break was given, and patients' reactions towards treatment were asked and recorded[10, 25].
Outcome measures
Clinical outcomes were measured before the treatment, right after the treatment, and followed up for 1, 2, and 4 weeks after the treatment, with primary outcome measurement of the VAS scale. Secondary outcomes included the cervical range of motion and Vernon-Mior neck disability index (NDI).
Pain severity was regarded as one of the most meaningful indexes and was evaluated by the VAS scale. The pain intensity was measured on a 10 cm line, on which patients were asked to place a mark on the line representing their pain level, where 0 indicated pain-free and 10 indicated the maximum pain level. Particularly, 0–3 represented mild pain, 4–6 represented moderate pain, and 7–10 represented severe pain; patients score 7 and above also suggested the difficulty of sleep due to neck pain[20, 26]. The minimal clinically important difference of this scale was considered 1.3 points based on a previous study reported by Cleland et al.[27].
Apart from the pain level, Vernon-Mior neck disability index (NDI) was a measurement to assess patients' functional disability resulting from neck pain. It contained 10 items, including pain, concentration, headache, and functional activities (personal care, lifting, sleeping, recreation, reading, driving, and work). Scores for each item range from 0 to 5, and the higher the scores were, the more severe the disability will be. The self-reported scale was considered the most valid and reliable as a responsive clinical tool. The test-retest reliability for cervical radiculopathy was reported fair (Intraclass correlation coefficient (ICC) = 0.55) to moderate (ICC = 0.68)[20, 28, 29].
Statistical analysis
SPSS 22.0 was used for statistical analysis. After examining for normal distribution, a two-way repeated measure ANOVA (two-tailed) was used between groups and within subjects for all outcome measures. Variables before treatment, immediately after treatment, and 1, 2, 4 weeks after treatment were analyzed. T test was used to compare baseline data between groups.
Primary outcomes between pain free group and mild pain group were used for effect size’s demonstration and evaluation. And more specifically, 0.2 represents a small effect size, 0.5 represents mediate and 0.8 represents large size. An approach of intentional treatment was used for the main comparative analysis with primary and secondary outcomes. A sensitive analysis was made for documenting missing data with a significance level of 0.05.