The immediate effects of passive joint mobilisation on local muscle function. A systematic review of the literature

Key points

1. 01Moderate evidence supports immediate reduction in superficial muscle activation during low-load tasks in people with pain, suggesting a shift toward deeper muscle recruitment.
2. 02Low evidence suggests mobilisation can improve maximal muscle strength in healthy individuals, with five of five asymptomatic studies showing positive effects.
3. 03Very low evidence for any strength benefit in symptomatic individuals; the one low-risk trial in this group found no effect.
4. 04Improvements in muscle function do not appear to require simultaneous pain reduction, pointing to non-pain-related mechanisms such as mechanoreceptor stimulation.
5. 05Neither of two trials found an effect of mobilisation on spinal reflex excitability, making motoneuron recruitment an unlikely mediator.

How it was conducted

Design

Systematic review following PRISMA guidelines; no meta-analysis possible due to heterogeneity

Included studies

17 controlled experimental trials (10 on asymptomatic individuals, 7 on symptomatic individuals)

Databases searched

MEDLINE, CINAHL, Web of Science, PEDro, ScienceDirect, CENTRAL; search cut-off December 26, 2018

Intervention

Passive oscillatory joint mobilisation (grades I-IV) at spinal and peripheral joints

Outcome measures

Surface EMG, ultrasound imaging, manual muscle strength tests, Hoffmann reflex, isokinetic dynamometry

Evidence grading

GRADE approach; risk of bias assessed with Cochrane RoB 2.0

What they found

• Chesterton and Payton (2017): lumbar mobilisation reduced erector spinae activity by -4.7% (90% CI [-10.5; 1.4%]) during active lumbar flexion and -18.3% (90% CI [-27.7; -7.6%]) during active knee extension in asymptomatic individuals.
• Krekoukias et al. (2009): lumbar mobilisation reduced erector spinae sEMG by 15.5% (95% CI 8.0-22.5%) vs control and 17.8% (95% CI 12.9-22.4%) vs placebo during standing in asymptomatic individuals.
• Chi-ngai et al. (2016): lumbar mobilisation increased hip flexor strength by +15.65 (SD 13.32)% vs -1.89 (SD 5.44)% in the control group in asymptomatic individuals with weak hip flexors.
• Makofsky et al. (2007): hip mobilisation produced +17.35% (95% CI [-0.01; 31.5]) increase in hip abductor torque vs -3.68% in sham group (p = 0.03) in asymptomatic individuals.
• Yerys et al. (2002): hip mobilisation produced +14% (95% CI [9.3; 16.8]) increase in hip extensor torque vs +4% in sham group (p = 0.002) in asymptomatic individuals.
• Yuen et al. (2017): lumbar mobilisation produced a significantly higher torque change (+3.4 (SD 5.4)%) vs sham (-2.36 (SD 5.81)%; p = 0.02), though within-group change was not significant (p = 0.076) in asymptomatic individuals.
• Liebler et al. (2001): thoracic mobilisation produced a 6.0% (95% CI [1.7; 10.2]) increase in lower trapezius torque vs control (p = 0.047) in asymptomatic individuals.
• Sterling et al. (2001): cervical mobilisation significantly reduced superficial neck flexor sEMG at 22 mmHg (-28 (SD 3.2)%), 24 mmHg (-34 (SD 2.6)%) and 26 mmHg (-21 (SD 1.8)%) vs placebo/control in patients with cervical pain (p < 0.0002).
• Pecos-Martin et al. (2017): thoracic mobilisation reduced erector spinae activity by -34.1% (95% CI [3.7; 63.4]) vs placebo in patients with non-specific thoracic pain (p < 0.05).
• Taylor et al. (1994): TMJ mobilisation reduced masseter sEMG at rest by 17% (SD 21.1%) and during clenching by 24% (SD 26.8%) vs placebo in patients with temporomandibular pain.
• Grindstaff et al. (2014) and Harkey et al. (2014): no immediate effect on spinal reflex excitability in knee injury or chronic ankle instability populations.
• Lluch et al. (2018, low risk of bias): no within- or between-group differences in shoulder rotator strength in overhead athletes with chronic shoulder pain.
• Interrater agreement for study selection: kappa = 0.79 (95% CI [0.57-1.0]; p < 0.001); for risk of bias: kappa = 0.85 (95% CI [0.75-0.95]; p < 0.001).

Limitations

• Only 17 studies were included, with small sample sizes across most trials; 10 of 17 did not report a power calculation, increasing the risk of type II error.
• Clinical heterogeneity across populations, joint regions, mobilisation grades and outcome measures prevented meta-analysis and limits generalisability.
• None of the asymptomatic studies reported patient blinding assessment, so placebo-related mechanisms (treatment expectation) cannot be excluded as an explanation for strength gains in healthy participants.
• The review focused only on immediate effects and did not capture long-term clinical relevance; changes in muscle function were not consistently linked to patient-reported outcomes.

Why it matters

For patients

People receiving joint mobilisation for musculoskeletal pain may experience an immediate change in how their local muscles activate, but it is not yet clear whether this translates to meaningful functional improvement.

For clinicians

Mobilisation can acutely reduce superficial muscle overactivity in painful conditions, which may support its use as part of motor retraining, but evidence for improving maximal strength in patients is very weak and based on a single low-risk trial showing no effect.

For readers

This review highlights that the muscle-function effects of mobilisation likely occur through proprioceptive and non-pain pathways, but high-quality RCTs linking these immediate neurophysiological changes to patient outcomes are still needed.