Nerve Gliding for Cervical Radiculopathy: What We Know

Nerve gliding exercises appear to reduce pain and disability in cervical radiculopathy when compared with no treatment or sham care. The best available pooled analysis found large standardized effects for both outcomes. That makes neuromobilization—a category that includes nerve "sliders" and "tensioners"—a reasonable addition to conservative care. It does not make nerve flossing a proven replacement for exercise, manual therapy, medication, or a clinician's assessment.

The distinction matters. Cervical radiculopathy can cause neck pain, arm pain, tingling, numbness, or weakness when a cervical nerve root becomes irritated or compressed. Online demonstrations often make nerve flossing look like a universal fix: bend here, tilt there, repeat until enlightened. The evidence supports a narrower claim. Neuromobilization beats inactive care in a small group of trials, but researchers cannot yet say it beats other active treatments or identify one reproducible protocol.

What the meta-analysis found

A 2023 systematic review and meta-analysis by Paraskevopoulos et al. identified seven randomized controlled trials of neuromobilization for cervical radiculopathy; five contributed data to the quantitative analyses. Against no-treatment or sham controls, neuromobilization improved every pooled outcome the review reported.

• Pain improved with a standardized mean difference (SMD) of −1.33, with a 95% confidence interval of −1.80 to −0.86 and P<0.01.
• Neck Disability Index scores improved with an SMD of −1.21, 95% CI −1.67 to −0.75 and P<0.01.
• Cervical flexion range of motion improved with an SMD of 0.66, 95% CI 0.23 to 1.10 and P<0.01.
• Cervical extension range of motion improved with an SMD of 0.47, 95% CI 0.04 to 0.90 and P<0.01.

Statistical heterogeneity was reported as I²=0% for each of these outcomes. The review rated the evidence for neuromobilization versus control as moderate certainty using GRADE.

An SMD compares groups in units of their variability rather than in familiar pain-scale points. By the conventional Cohen benchmark, an effect around 0.8 is large; the pooled pain effect of −1.33 exceeds that threshold. The disability effect does too. Those are substantial statistical signals against inactive control conditions.

They are not a promise that every patient will experience a large drop on a 0-to-10 pain scale. The review does not provide the information needed to convert the pooled SMD into a specific VAS or numerical-rating-scale change, and the available evidence does not establish cervical-radiculopathy-specific minimum clinically important differences for pain or the Neck Disability Index. "Large" describes the standardized result, not a confirmed threshold of meaningful improvement for each individual patient.

Beating inactivity is not the same as beating treatment

The same meta-analysis found no statistically or clinically significant superiority for neuromobilization over other active conservative interventions for pain, disability, or range of motion. The certainty of this head-to-head evidence fell to low or very low on GRADE. That leaves two plausible explanations: nerve gliding may perform similarly to other active options, or the comparative trials may have been too small or too uncertain to detect a real difference.

Neuromobilization has evidence as an active component of conservative care. It does not yet have evidence that makes it the preferred component.

This limitation is easy to lose in social-media translations of the research. A favorable comparison with doing nothing can justify offering an intervention. It cannot establish that the intervention should displace a treatment that already has evidence, especially when direct comparisons are imprecise.

The broader neck-pain literature shows why those comparisons matter. In a randomized trial by Akgüller et al., 60 people with mechanical neck pain were assigned to cervical thrust manipulation, exercise, or both treatments. All three groups improved, while the combined manipulation-and-exercise group had greater improvement on several outcomes and was the only group to achieve the study's minimum clinically important difference for the Neck Disability Index. Mechanical neck pain is not cervical radiculopathy, so this trial does not test nerve gliding. It does show that treatment combinations and comparator choice can change results in cervical pain research.

Other studies reinforce that point without settling the question for radiculopathy. A 2025 trial by Aygul et al. assigned 66 people with cervical degenerative disease to home exercise, manual therapy, or telerehabilitation-assisted treatment; every group completed an eight-week exercise program, with added twice-weekly sessions in the manual-therapy and telerehabilitation groups. All groups improved pain and function. Meanwhile, a 2024 randomized trial by Osama et al. compared muscle-specific and movement-specific muscle energy techniques in 66 adults with mechanical neck pain. These are adjacent populations, not substitute evidence for cervical radiculopathy, but they illustrate how many active approaches occupy the same clinical territory.

Guidelines have not made nerve gliding a standalone first choice

A 2024 systematic review by Feng et al. summarized international clinical practice guidelines for cervical radiculopathy. The guidelines it reviewed included patient education, exercise therapy, NSAIDs, manual therapy, cervical steroid injections, and cervical collar use among first-line conservative options. They did not designate neuromobilization as a standalone recommended treatment.

That omission does not prove nerve gliding fails. Guidelines weigh evidence quality, comparative effectiveness, feasibility, and clinical applicability. Here, the position fits the available data: neuromobilization has moderate-certainty evidence versus inactive control, but weak comparative evidence against active care.

Non-manual treatments also compete for attention. A 2024 systematic review of high-intensity laser therapy examined randomized trials in neck-pain disorders and assessed pain, disability, and cervical range of motion. It does not answer whether laser therapy is better than nerve gliding in cervical radiculopathy, because the population and comparison differ. It does show why declaring any single modality the winner would outrun the evidence.

Why the pooled estimates need restraint

Five contributing randomized trials are enough to calculate a pooled result, but not enough to make that result immune to revision. Small intervention literatures can produce large point estimates that shrink as larger trials arrive. They also make heterogeneity statistics fragile. An I² of 0% may mean the trials produced similar estimates; with only five trials, it may also mean the analysis had limited ability to detect meaningful differences in technique, patient selection, treatment dose, or comparator.

The review's June 2022 search cutoff is another constraint. The 2023 publication summarizes a literature snapshot that may exclude later trials. It also does not provide a usable clinical recipe. The available report does not establish whether sliders or tensioners work better, whether the median, ulnar, or radial nerve should be targeted, how many repetitions to use, how often to perform them, or how long treatment should continue.

Long-term outcomes remain especially uncertain. The evidence set does not provide follow-up beyond three months, and the individual trial follow-up periods are not reported here. A short-term gain in pain or motion can matter, but it does not tell us whether benefits persist, whether patients need continuing treatment, or which people respond best by symptom duration, severity, or affected nerve root.

How to use this evidence without overreading it

If a clinician has diagnosed cervical radiculopathy and conservative treatment is appropriate, neuromobilization is a defensible adjunct to an individualized plan. The pooled evidence supports benefit compared with inactive care, and it may fit alongside therapeutic exercise, education, manual therapy, and medication when indicated.

It should not become a self-directed substitute for evaluation, particularly when weakness is progressing, sensation is changing, pain is severe, or symptoms follow trauma. Generic online routines may not match the interventions tested in the trials, and the evidence does not identify a single dose or nerve target that readers can safely copy.

The practical verdict is modest but useful: nerve gliding has a real signal in cervical radiculopathy, especially against doing nothing. Its place is inside multimodal conservative care until larger, well-described head-to-head trials show whether it offers an advantage over the other treatments already on the table.