The effects of exercise on neuromuscular function in people with chronic neck pain: a systematic review

Key points

1. 01A meta-analysis of 6 studies (9 interventions) showed a statistically significant reduction in sternocleidomastoid (SCM) EMG amplitude after exercise versus control (effect size -1.67, 95% CI -2.40 to -0.94, Z=4.47, p<0.00001).
2. 02Craniocervical flexion (CCF) training and resistance training each carry moderate-certainty evidence for inducing neural adaptations in neck muscles; combined CCF plus resistance training also rates moderate certainty.
3. 03Neural adaptations appear to be task-specific: CCF training preferentially improves deep cervical flexor activation and reduces superficial muscle overactivity during the CCFT, while resistance training reduces superficial muscle activity during loaded contractions.
4. 04Neuromuscular changes can occur rapidly, with 2 to 4 weeks of training sufficient to reduce SCM and upper trapezius EMG during the craniocervical flexion test.
5. 05All studies except two reported concurrent improvements in pain and disability, suggesting neuromuscular changes track clinical benefit.

How it was conducted

Design

Systematic review and meta-analysis; PROSPERO registration CRD42021298831; reported per PRISMA

Databases searched

MEDLINE, CINAHL, Web of Science, Scopus, AMED, Google Scholar, Open Grey, Zetoc; searched from inception to 21 October 2023

Included studies

14 articles (10 RCTs, 1 preliminary RCT, 3 randomised trials) published 1999-2023

Participants

Adults with chronic non-specific neck pain; most studies enrolled women aged 25-50 years; sample sizes per study ranged from 20 to 100

Interventions

Exercise programs targeting neck muscles for at least 2 weeks: CCF training, resistance/endurance training, neck-specific programs, scapular stabilisation, suspension training, global postural re-education, Feldenkrais, and mixed approaches

Primary outcomes

Neuromuscular function via EMG (amplitude, timing, fatigability), muscle strength, and endurance; secondary outcomes were pain intensity and disability

What they found

• Meta-analysis of SCM EMG amplitude (6 studies, 9 interventions): overall effect size -1.67 (SE 0.73), 95% CI -2.40 to -0.94, Z=4.47, p<0.00001, favouring exercise over control.
• CCF training (Jull 2009): significant decrease in SCM EMG during CCFT at 30 mmHg post-training (p<0.001); no significant change after strength training on the same task (p>0.05).
• Resistance training (Falla 2006): significant reduction in SCM EMG amplitude at 10%, 25% and 50% MVC (p<0.05); mean reductions approximately 7.8-11.8 units.
• CCF training (Falla 2006): no significant change in SCM or AS EMG amplitude during isometric contractions (p>0.05).
• Resistance training (Mehri 2020): significant reduction in UT RMS (48.32 to 39.2, p=0.005), SCM RMS (41.4 to 35.3, p=0.024), and ES RMS (55.22 to 43.1, p=0.019); earlier activation onset for UT (165.42 to 147.65 ms, p=0.008), SCM (142.38 to 118.34 ms, p=0.005), and ES (156.38 to 142.85 ms, p=0.001) versus control.
• Shoulder lateral raise training (Lidegaard 2013): training increased average duration of EMG gaps by 71%, EMG gap frequency by 296%, and percentage time below 0.5% and 1.0% EMGmax by 578% and 242% respectively in splenius capitis during the workday (p<0.05).
• Neck-specific exercise program (Falla 2013): significantly less SCM and splenius capitis EMG amplitude during MVCs post-training (p<0.05) versus control.
• Global postural re-education (Mendes-Fernandes 2023): significant reduction in SCM and anterior scalene RMS during CCFT at all pressure stages (p<0.05); GRADE high certainty.
• Craniocervical flexion endurance (Ghaderi 2017 CCF group): endurance time improved from 19.53 to 73.59 seconds (p<0.001); between-group difference not significant (p=0.365).

Limitations

• Most included studies had high or some-concerns overall risk of bias, largely because four trials lacked a control group and several had small sample sizes (fewer than 24 participants).
• Wide heterogeneity in exercise type, dosage, duration (2 to 16 weeks), and EMG tasks assessed made quantitative pooling possible only for SCM amplitude, limiting broader meta-analytic conclusions.
• Almost all studies enrolled exclusively or predominantly women, restricting generalisability to men and mixed-sex populations.
• No study measured corticospinal excitability via transcranial magnetic stimulation, leaving central neural mechanisms uncharacterised; longer-term follow-up data were also largely absent.

Why it matters

For patients

People with chronic neck pain can expect that a structured neck exercise program, particularly craniocervical flexion or resistance training, is likely to reduce overactivity of superficial neck muscles and may also ease pain and disability.

For clinicians

Exercise prescription should be matched to the specific neuromuscular deficit: CCF training is best supported for improving deep flexor activation and reducing superficial muscle overactivity during low-load tasks, while resistance training better addresses strength deficits and high-load tasks.

For readers

This is the first systematic review to comprehensively cover all exercise types and their effect on cervical neuromuscular function; moderate-certainty evidence supports CCF and resistance training, but larger RCTs measuring deeper neck muscles and longer follow-up are still needed.