Table 2.
Details of included studies for radicular pain.
| Participants | N (at follow-up) | Intervention | Control | Disability | Results (pain) | Risk of bias | |
|---|---|---|---|---|---|---|---|
| Persson et al.63 | People with chronic cervical radiculopathy with evidence of nerve compression on MRI | 82 (79) | 3 mo of twice per week physiotherapy including TENS, heat, cold, massage, active and aerobic exercise chosen at the discretion of the 25 physiotherapists involved. Another intervention arm received decompression surgery |
Comfortable rigid collar, worn for 3 mo | Not measured | Fourteen weeks after treatment began, mean pain intensity in the surgery arm had improved by −20 points on a 0–100 VAS. Improvement was −9 points in the physiotherapy group, and −1 point in the neck collar group. Between-group differences were P < 0.01. | Some concerns |
| Hofstee et al.35 | People with acute lumbar radicular pain | 250 (225) | Four to 8 wk of twice per week physiotherapy including manual mobilisations and spinal exercise. Another intervention group received advice to rest in bed for 7 d, and to rest as much as possible thereafter |
Advice to continue with normal activities of daily living | After 2 mo, the difference between the bed rest and the control group was a mean −2.7 (−9.9 to 4.4) on the 0–100 Quebec Disability Scale in favour of the control. There was no mean difference between the physiotherapy and the control group. 2 m: −0.0 (−7.2 to 7.3). | After 2 mo, the difference between the physiotherapy group and the control group was a mean 0.8 points on a 0–100 VAS (95% CI: −8.2 to 9.8). The difference between the bed rest and control arms was a mean 2.5 (95% CI: −6.4 to 11.4) in favour of the control group. | High |
| Bakhtiary et al.4 | People more than 2 mo of radicular pain from a herniated lumbar disk, referred from orthopaedic care to physiotherapy | 60 (52) | One group performed stabilisation exercises at home twice per day for 4 wk, whereas the other group did not exercise. Groups then crossed over. | None (2 intervention arms) | Not measured | After the first 4 wk, the exercising group improved by a mean difference from baseline of −3.2 points on a 10-cm VAS, whereas the group not exercising improved by −0.5 points; the between-group difference was −2.7 points (95% CI: −3.5 to −1.9). After 8 weeks, when the second group had also exercised, the between-group difference was −0.9 (95% CI: −1.7 to −0.01). |
Some concerns |
| Luijsterburg et al.53 | People with acute lumbar radicular pain in primary care | 135 (117) | GP and physiotherapist care. Physiotherapy based on exercise and return to activity, with no manual therapy or electrotherapy. | Usual guideline-based GP care | After 12 wk, the difference between groups on the 24 point RMDQ was 0.8 (95% CI: −1.6 to 3.2) in favour of the control group. After 1 y, the difference was −0.9 (95% CI: −3.0 to 1.3) in favour of the intervention group. |
After 12 wk, the intervention group reported an improvement in leg pain of −3.9 on a 10-point NRS and the control group reported an improvement of −3.7; the mean difference between the 2 groups of 0.3 (95% CI: −0.06 to 1.2). After 1 y, the intervention group reported an improvement of −4.4 and the control group −3.7; the mean difference between the 2 groups was −0.7 (95% CI: −1.7 to 0.2). The primary outcome measure was global perceived effect. |
Low |
| Kuijper et al.48 | People with acute cervical radicular pain | 205 (192) | One intervention group wore a semihard collar during the day for 3 wk and were advised to rest as much as possible. They were then weaned from the collar for 3 wk. The other intervention group engaged in twice-weekly supervised physiotherapy for 6 wk, with a focus on mobilising and stabilising the spine, along with home exercises. | Reassurance. | After 6 weeks, the beta-coefficient for weekly change in the NDI was 0.8 points per week (95% CI: −1.8 to 0.2) in the physiotherapy group when compared to the control group. | The authors used generalised estimating equations to show that both intervention groups reported a benefit of 1.9 mm on a 100-mm VAS per week in arm pain for the first 6 wk (95% CIs −3.3 to −0.5 for neck collar; −3.3 to −0.8 for exercise). After 6 mo, all groups reported a median of 0 points. |
High |
| Young et al.79 | People with mixed acute and chronic cervical radicular pain | 81 (69) | Manual therapy, exercise, and intermittent cervical traction | Manual therapy, exercise, and sham traction | After 4 weeks, there was a mean difference of 1.5 points (95% CI: −6.8 to 3.8) on the neck disability index and 0.29 points on the PSFS 0.29 (95% CI: −1.8 to 1.2) in favour of the intervention group. | After 4 weeks, there was an adjusted mean between-group difference of 0.52 points on a 10-point NRS (95% CI −1.8 to 1.2) in favour of the intervention group | Some concerns |
| Huber et al.36 | People with acute lumbar radicular pain, caused by a herniated disk, in primary care | 52 (52) | Three sessions per day for 20 d of supervised supine isometric exercises. Not specified what proportion of the sessions was supervised. | Advice to continue with activities of daily living | Not measured | After the intervention, the intervention group reported 1.7 points less pain on a 10-point VAS but the authors did not report the between-group difference or provide exact P-values for the within- or between-group differences, so we are unable to say how precise this estimate is. | High |
| Albert and Manniche2 | People with mixed acute and chronic lumbar radicular pain, in secondary care | 181 (170) | Information, advice, and symptom-guided spinal exercises based on McKenzie method of directional preference, along with stabilizing exercises. Eight weeks with 4–8 treatment sessions. | Information, advice, and low-dose general exercises to stimulate circulation | The authors reported no significant between-group differences on the RMDQ but did not give values for this. | After treatment, there was a 0.8-point mean difference between the groups on a 10-point NRS for patients' current leg pain (P = 0.06), in favour of the intervention group. On a “total leg pain” score, including current leg pain, worst leg pain, and average leg pain, there was no difference in mean ratings between the 2 groups (P value not provided). | High |
| Nee et al.61 | People with more than 4 wk of nerve-related arm pain. With a positive median nerve tension test and without more than 2 abnormal neurological findings. Recruited from the community through advertisements in newspapers and e-newsletters | 60 (56) | A standardised programme of 4 sessions of neural tissue management, including exercise, manual therapy, and education, over 4 wk. | Advice to continue with activities of daily living, with complementary treatment after the trial | After treatment, there was a mean 3.4-point difference between groups (95% CI: −0.6 to 6.3) on the NDI and a 2.1-point difference (95% CI: 0.9–3.2) on the PSFS, both in favour of the intervention group. | After treatment, there was a mean 1.5-point difference on a 0–10 NRS favouring the intervention group (95% CI: −0.5 to −2.6). Primary outcome measure was Global Rating of Change. | Low |
| Fritz et al.26 | People with mixed acute and chronic cervical radicular pain | 86 (54) | One intervention group performed exercise (scapula and neck muscle strengthening) and received mechanical traction. The other intervention group performed exercise and overdoor traction. 10 sessions over 4 wk. | 10 sessions over 4 wk of exercise only | Immediately after treatment, there were small, not clinically or statistically significant differences between groups on the NDI. At 6-mo follow-up, the group receiving mechanical traction reported a mean of 13.3 points (95% CI: 5.6–21.0) less disability. The group performing overdoor traction reported a mean 5.2 points (95% CI: −2.6 to 13.0) less disability. | Immediately after treatment, there were small, not clinically or statistically significant differences between groups. At 6-mo follow-up, the group receiving mechanical traction reported a mean of 2.3 points (95% CI: 0.9–3.8) less arm pain on a 10-point NRS scale than the exercise-only control group. The group performing overdoor traction reported 2.5 points' less pain (95% CI: 1.0–4.0). | Some concerns |
| Langevin et al.49 | People with cervical radicular pain | 36 (36) | A 4-wk programme of manual therapy, exercises and stretches aimed at increasing space in the intervertebral foramen. | A similar programme not aimed at increasing space in the intervertebral foramen | After 4 wk there was a mean 2.3-point difference (95% CI: 10.1 to −5.5) between groups on the NDI and 3.9 points (95% CI: 14.0 to −6.2) on the QuickDASH, both in favour of the intervention group. After 8 wk, this was 4.6 points (12.1, −2.8) and 5.6 points (95% CI: 20.0 to −8.9), respectively. | After 4 wk there was a mean −0.1 difference (95% CI: −1.9 to 1.8) in mean arm pain on a 10-point NRS. After 8 wk, the difference between the groups was −1.3 points (95% CI: −2.8 to 0.2). | Some concerns |
| Moustafa and Diab59 | People with chronic lumbar radicular pain from a disk herniation and anterior head translation as measured by cervical radiograph | 154 (131) | A 2-y programme of gym-based “functional restoration” exercises in phases of decreasing independence and increased supervision. Some exercises were intended to encourage upright neck posture. | The same exercise programme without the exercises aimed at neck posture | After 10 weeks, the control group reported a mean 2.82 points lower score on the ODI (P = 0.08). After 2 y, the mean difference was 11.8 (P = 0.005). | After 10 weeks, the control group reported 0.2 points less leg pain on a 10-point NRS (95% CI: −0.73 to 0.14). After 2 y, the intervention group reported 1.6 points less leg pain (95% CI: −2.5 to −1.58). The primary outcome measure was the Oswestry Disability Index. |
High |
| Ferreira et al.22 | People with chronic nerve-related leg pain. Recruited from the community through newspaper and social media advertisements. | 60 (54) | Four sessions in 2 wk of manual therapy and exercises aimed at managing neural mechanosensitivity. | Advice to remain active | After 2 wk, the intervention group reported a mean 3.3 points less disability on the ODI (95% CI: 9.6 to −2.9) and 5.3 points on the PSFS (95% CI: 2.2–8.2). After 4 wk, this was 5.0 points on the ODI (11.0 to −1.1) and 4.7 points on the PSFS (1.7–7.8) | After 2 weeks, the intervention group reported a mean −1.1 points less leg pain on a 0–10 NRS scale (95% CI: −2.3 to 0.1). After 4 wk, they reported −2.4 points less pain (95% CI: −3.6 to −1.2). | Low |
| Kim et al.43 | People with chronic cervical radicular pain | 30 (30) | Eight week, 3 times per week programme of manual cervical traction with neural mobilisation | Manual cervical traction only | After 8 wk, there was a mean 3.27 points difference between the groups on the NDI (P = 0.004) | After 8 wk, there was a mean 1 point difference between the groups on a 0–10 numeric rating scale (P = 0.006) | High |
| Hahne et al.32 | People with chronic lumbar radicular pain from a disk herniation | 54 (49) | 10 sessions in 10 wk of a multimodal individualised functional restoration programme with a behavioural component | Two 30-min advice sessions | After treatment, patients in the intervention group reported a mean 7.7 points less disability (95% CI: 0.3–15.1) on the ODI compared to the control group. At 26 wk, the difference was 5.7 (95% CI: −1.7 to 13.1) and at 1 y, 8.2 (95% CI: 0.7–15.6) All in favour of the intervention group |
After treatment, patients in the intervention group reported a mean 1.1 points less in leg pain on a 0–10 NRS scale (95% CI: −0.3 to 2.4) compared to the control group. At 26 wk, the difference was 1.2 (95% CI: −0.2 to 2.6) and at 1 y, 0.9 (95% CI: −0.5 to 2.3) | Low |
| Akkan and Gelececk1 | People with cervical radicular pain | 46 (32) | Fifteen sessions in 4 wk of neck stabilisation exercises, generic neck exercises, hot pack, TENS, and ultrasound treatment with training on postural alignment | As in the intervention group but without stabilisation exercises | After 4 wk, there was a 0.39 points mean difference between groups on the NDI in favour of the intervention group. At 12 wk, this was 0.24 points. Between-group difference was P > 0.05 (exact value not stated) | After 4 wk, there was no mean difference between the groups on a 0–10 VAS. After 12 wk, the control group reported 0.21 points less pain. Between-group differences P > 0.05 (exact value not stated). | High |
| Calvo-Lobo et al.12 | People with nerve related arm pain | 105 (75) | One intervention group received 5 sessions per wk for 6 wk of median nerve neural mobilisation. Another intervention group received the same frequency of a cervical lateral glide technique. | Oral ibuprofen. | After 6 wk, the oral ibuprofen group reported 14.4 points less disability on the QuickDASH (95% CI: 8.48–20.23) than the group receiving median nerve mobilisations; and 19.2 points less disability (95% CI: 13.79–24.67) than the group receiving cervical lateral glides. | After 6 wk, measured 1 hr after treatment, the oral ibuprofen group reported a mean 1.8 points less pain on a 0–10 NRS (95% CI: 1.12–2.42) than the group receiving median nerve neural mobilisations; and 2.2 points less pain (95% CI: 1.61–2.69) than the group receiving cervical lateral glides. | High |
| Dedering et al.15 | People with mostly chronic cervical radicular pain recruited from a neurosurgical department | 144 (73) | A 3 mo, 3 times per week programme of progressive neck-specific training with multiple sessions of cognitive behavioral therapy coaching | A general exercise programme with a single session of cognitive behavioural therapy coaching | After 3 mo, the intervention group reported a mean 2 points less disability on the NDI (95% CI: −6 to 10). After 1 y, this difference was 1 point in favour of the control group (95% CI: 1 to −7). | After 3 mo, the intervention group reported a mean 8 points less arm pain on a 0–100 mm VAS (95% CI: −2 to 18). After 1 y, the difference was 2 points in favour of the intervention group (95% CI: −10 to 14). | Some concerns |
| Rodriguez-Sanz et al.65 | People with cervical radicular pain | 60 (51) | A 6-week programme, 5 d per week of manual therapist-applied median nerve mobilisation. | Waiting list | After treatment, there was a mean 26.97 points difference on the Quick DASH in favour of the intervention group (95% CI: 33.75–20.20) | After treatment, there was a mean 3.70 point difference on a 0–10 NRS in favour of the intervention group (95% CI: 4.29–3.10) | High |
| França et al.25 | People with chronic lumbar radicular pain caused by disk herniation | 40 (40) | An 8-week, twice-weekly programme of lumbar stabilisation exercises | The same frequency of treatment with TENS | After treatment, there was a mean 8.4 point difference between the groups on the 44-point functional disability version of the ODI (95% CI: 5.44–11.36) in favour of intervention group. | After treatment, there was a mean 3.3 point difference between the groups on a 0–10 VAS (95% CI: 2.12–4.48) in favour of the intervention group. | Some concerns |
| Plaza-Manzano et al.64 | People with lumbar radicular pain caused by disk herniation | 32 (32) | An 8-wk, 4 times per week programme of neural mobilisation plus motor control exercises | The same frequency of motor control exercises alone | After treatment, the intervention group reported a mean 0.7 points lower score on the RMDQ but between-group analysis was not provided. | After treatment, the intervention group reported a mean 2.6 points of leg pain on a 0–10 NRS (95% CI: 2.2–3.0). The control group reported 3.2 points leg pain (95% CI: 2.8–3.6). Between-group data not provided. | Some concerns |
| Satpute et al.69 | People with subacute and chronic lumbar radicular pain, excluding patients with neuropathic pain on S-LANSS | 60 (39) | Six sessions over 2 wk of spinal mobilisation with leg movements plus usual care or TENS and active nerve mobilisations. | Usual care of TENS and active nerve mobilisations alone | After treatment, the intervention group reported a mean 3.9 points less pain than the control group on the ODI (95% CI: 5.5–2.2). At 3 mo, the difference was 5 points (95% CI: 6.5–3.4) and at 6 mo, 4.7 points (95% CI: 6.3–3.1). | After treatment, the intervention group reported a mean 2.0 points less pain than the control group on a 0–10 VAS (95% CI: 1.4–2.6). At 3 mo, the difference was 2.8 (95% CI: 2.2–3.4) and at 6 mo, 2.6 (95% CI: 1.9–3.2). | Some concerns |
| Basson et al.7 | People with acute and subacute nerve-related arm pain, without “serious neurological signs.” | 86 (78) | Usual care of spinal mobilisations, exercise, and advice to stay active plus therapist-applied neural mobilisation. | Usual care of neck mobilisations, exercise, and advice to stay active. | After 6 wk, the intervention group reported a mean 0.5 points more disability on the PSFS (95% CI: −3.9 to 3.1). After 6 mo, the difference was −0.1 (95% CI: −2.7 to 2.5) and at 12 mo, −0.3 (95% CI: −2.7 to 2.5). | After 6 wk, the intervention group reported a mean 0.6 points less pain on a 0–10 point NRS (95% CI: −0.6 to 1.8). After 6 mo, the difference was 1.1 (95% CI: 0.1–2.2) and at 12 mo, 1.1 (0.1–2.0). | High |
Unless otherwise specified, studies used pain as a primary outcome measure. The number of participants shown at follow-up is for the longest-term follow-up in each trial.
95% CI, 95% confidence interval; GP, general practitioner; NDI, Neck Disability Index; NRS, Numeric Rating Scale; ODI, Oswestry Disability Index; PSFS, Patient-Specific Functional Scale; QuickDASH, Quick disabilities of the Arm, Shoulder and Hand Score; RMDQ, Roland Morris Disability Questionnaire; S-LANSS, Self-Administered Leeds Assessment of Neuropathic Symptoms and Signs; VAS, visual analogue scale; TENS, transcutaneous electrical nerve stimulation.