No differences between individuals with chronic idiopathic neck pain and asymptomatic individuals on seven cervical sensorimotor control tests: a cross-sectional study

Key points

1. 01No significant between-group differences were found on any of the seven tests, with p-values ranging from 0.203 to 0.981
2. 02Test scores correlated only weakly with neck pain intensity (rho 0.010 to 0.294) and neck disability (rho 0.007 to 0.316)
3. 03Poor performers on each test included both people with and without neck pain in similar proportions
4. 04Cervical range of motion was reduced in all directions in the neck pain group, confirming the sample had genuine impairment
5. 05Results question whether sensorimotor control disturbances are a meaningful feature of chronic idiopathic neck pain

How it was conducted

Design

Case-control cross-sectional study

Participants

50 adults with chronic idiopathic neck pain and 50 age- and sex-matched asymptomatic controls

Neck pain criteria

Neck pain at least 4/10 on numeric rating scale, duration at least 12 weeks, spontaneous idiopathic onset

Tests administered

Seven cervical sensorimotor control tests: joint position error, joint position error torsion, postural balance, subjective visual vertical, head tilt response, The Fly, smooth pursuit neck torsion, and head steadiness

Primary analysis

Mann-Whitney U tests for between-group differences; Spearman rho for correlations with pain and disability

Setting

University of Newcastle, Australia; conducted July 2016 to June 2017

What they found

• No significant differences between groups on any of the seven tests; p-values ranged from p=0.203 to p=0.981
• Correlations between sensorimotor tests and current neck pain intensity ranged from rho=0.010 to rho=0.294
• Correlations between sensorimotor tests and neck disability ranged from rho=0.007 to rho=0.316
• Only two significant correlations were found: head tilt response with current neck pain intensity (rho=0.294, p=0.042) and with neck disability (rho=0.316, p=0.029)
• Proportions of poor performers (top 10% scores) were not significantly different between groups for any test
• Mean number of tests with poor performance was 1.4 (SD 1.5) in the neck pain group versus 1.6 (SD 1.6) in the asymptomatic group (p=0.660)
• Cervical range of motion was significantly less in the neck pain group in all measured directions
• Number of poor-performing tests correlated only weakly with demographics, ROM, and pain or disability levels

Limitations

• Assessor blinding was not possible, though most tests used computerized data collection reducing manual bias
• Participants were mainly recruited through advertising, so the sample may not fully represent treatment-seeking patients
• All tests were performed in a single approximately 75-minute session, which may have allowed practice effects despite randomization of test order
• Findings may not generalize to other neck pain populations such as acute, traumatic, or whiplash-related neck pain

Why it matters

For patients

People with chronic idiopathic neck pain may not have measurable differences in head and neck movement control compared to those without pain, so tests focused on this area may not add useful information about their condition.

For clinicians

Using cervical sensorimotor control tests routinely in people with chronic idiopathic neck pain is not well supported by this evidence, as none of seven tests differentiated patients from controls or correlated meaningfully with pain or disability.

For readers

This well-matched case-control study with a battery of seven tests provides the most comprehensive challenge yet to the assumption that sensorimotor impairment is a defining feature of chronic idiopathic neck pain.