Research Review By Dr. Ceara Higgins©


Download MP3

Date Posted:

October 2017

Study Title:

Sensorimotor control in individuals with idiopathic neck pain and healthy individuals: A systematic review and meta-analysis


de Zoete RMJ, Osmotherly PG, Rivett DA, et al.

Author's Affiliations:

School of Health Sciences, Faculty of Health and Medicine, The University of Newcastle, NSW; Recover Injury Research Center, Menzies Health Institute Queensland, Griffith University, QLD, Australia

Publication Information:

Archives of Physical Medicine and Rehabilitation 2017; 98: 1257-1271.

Background Information:

An estimated 70% of the population will suffer from neck pain at least once during their lives (1). This is a significant number, since neck pain has been related to high health care costs (2) and significant disability (3). Traditional rehabilitation programs have been shown to be unsuccessful in treating a lot of neck pain problems (4) and common treatments such as joint mobilization, massage, traction, and exercise focused on overall muscle strength show short term effects, but no significant, consistent impact on long-term patient symptoms (5).

Sensorimotor control, also referred to as proprioception, kinesthesis, and somatosensory control, includes sensory input and motor output for adequate motor control. In this review, sensorimotor control testing is used to describe tests that include all afferent and efferent information streams and the central integration components that contribute to joint stability (6). Research in this area has most commonly looked at individuals with whiplash associated disorders (WAD). The authors of this review chose to look at idiopathic neck pain, as it presents more commonly in clinical practice (7). There is currently no evidence to support a single method to measure deficits in the cervical sensorimotor system (8). The authors of this review focused on two main questions:
  1. What tests are used to assess cervical sensorimotor control in patients with idiopathic neck pain?
  2. Can sensorimotor control tests identify and quantify differences between patients with idiopathic neck pain and healthy individuals?

Appendix A: Test Descriptions

The following tests will be referred to in the Literature Summary section below:
  • Joint position error (JPE) – cervical spine: The participant actively moves the cervical spine in the transverse, sagittal, or coronal plane, and then returns to neutral head position or a pre-set target position. After each trial, the examiner repositions the head to the neutral head position. Error is measured using a laser pointer or electromagnetic motion tracking system.
  • Postural sway: Participants stand on a force platform in both eyes open and eyes closed conditions. The amount of movement during quiet stance (sway) is measured as the area or distance from the center of pressure.
  • Subjective visual vertical: A computer screen or virtual reality device is used to present a titled line. Participants use the computer mouse or knob to reposition the line to true vertical.
  • Smooth pursuit neck torsion: Keeping the head still, in a neutral position, participants focus their eyes on a moving target. This is then repeated with the head held so that the torso is rotated relative to the head (torsion) to either side. The velocity of eye movements while following the target is recorded and the gain (ratio between eye movement and target movement) is calculated. The outcome measure is the difference between the gain in neutral and the average gain in torsion (right and left).
  • The Fly: Participants follow a target (the fly) on a computer screen by moving their head and neck. A Fastrack system is used to measure their accuracy.
  • Head steadiness: An electromagnetic motion tracing system is used to measure angular velocity to evaluate the participant’s ability to hold their head still in a given position over a certain time. Generally, this involves the individual holding their head up off the treatment table while in the supine position.

Literature Summary:

Forty-three studies were included, with 30 suitable for meta-analysis.

Sensorimotor outcomes in individuals with neck pain and healthy individuals:

Six different tests were used, with joint position error (JPE) and postural sway tests most commonly used. The average difference for JPE to the neutral head position after rotation ranged from 2.2-9.8 degrees in individuals with neck pain and 1.66-5.1 degrees in healthy individuals. After flexion-extension movements the average range was measured at 3.2-3.5 degrees in neck pain patients and 1.72-4.4 in healthy individuals. This indicates a significant difference between individuals with neck pain and those without (particularly in rotation).

The validity of JPE testing must be questioned, as the neutral neck position is a commonly adopted position, it may be easier for the subject to return their head to neutral purely based on memorization (13) and recall rather than overt proprioception. Studies using a pre-set target in the transverse plane may be more useful, as recalling a head position from previous experience is less likely. One advantage to JPE testing is the use of blindfolds to exclude the visual system. This allows testing to be more specific to the proprioceptive and vestibular subsystems without interference from the visual-motor system.

Studies involving postural sway testing ranged greatly in size, from 9-107 subjects. The average sway area with eyes open ranged from 4.85-10.5cm2 in the neck pain group and from 3.5-6.6cm2 in healthy individuals. With eyes closed, the range went from 2.51-16.6cm2 in the neck pain group and from 2.74-10.9cm2 in the healthy group. Once the data was pooled and run through meta-analysis, no significant differences were found between individuals with idiopathic neck pain and healthy individuals under either the eyes open nor eyes closed conditions. The motor output during quiet stance in postural sway testing is not specifically controlled by the cervical spine, but due to the high density of proprioceptive receptors and afferent pathways for the vestibular system that are found in the neck (15), it is reasonable to apply the outcomes of postural sway testing to cervical sensorimotor control (6).

Four additional tests were less commonly studied. One study (14) looked at the subjective visual vertical and found a significant difference between individuals with idiopathic neck pain, with an average error of 2.43 degrees, and healthy individuals, with an average error of 1.92 degrees. Although subjective visual vertical testing’s motor output component and relevance to the neck is questionable, a variation called the head tilt response test has been proposed by Geisinger et al. (16). In this test, a line is presented on a virtual reality device. The line is then tilted and the participant repositions the line along the vertical by laterally flexing the cervical spine. This presents a better representation of the motor output component of sensorimotor control, specific to the cervical spine. However, this study was excluded from this review because it did not examine individuals with idiopathic neck pain. Future research is needed to examine the effectiveness of this test in an idiopathic neck pain population.

Four studies were found that assessed the smooth pursuit neck torsion test. No study compared idiopathic neck pain and healthy individuals. One study looked only at individuals with idiopathic neck pain, reporting a gain difference (ratio of eye movement to target movement) between neutral and a rotated position at 0.06. The other three studies looked at healthy individuals and reported a gain difference from 0.01-0.02. The validity of this test is questionable in terms of cervical motor output because it is mainly focused on sensory input and the response of the eyes to that input. However, it assesses cervical proprioceptive reflexes (7) and is believed to isolate the vestibular system, so may be a valuable part of sensorimotor testing.

Three studies looked at the Fly test and found significant differences between healthy individuals and those with idiopathic neck pain, although not all studies showed these differences in all patterns. This test appears to involve both sensory and motor components via the visual input of the moving target and the motor output of the movement of the neck. As well, due to the constantly changing pattern of target movement, there is no possible issue related to memorization of a certain head position. However, this random pattern doesn’t allow for differentiation between cervical movement in different directions.

Finally, two studies looked at head steadiness, with one study finding significant differences between individuals with idiopathic neck pain and healthy individuals, but the other study showing no significant differences. This test may be considered more of a test of muscular endurance than sensorimotor control.

More recently, Chen and Treleaven (12) have reported on new sensorimotor tests. The first is the JPE torsion test, where the head remains still and the trunk is rotated relative to the head while the repositioning error of the trunk is measured. This avoids the influence of the vestibular system, making it potentially useful for differentiating the sensorimotor subsystems. The second is the Enbloc test, where the repositioning error of the head and trunk is measured as a block. There is no literature available yet on these tests in individuals with idiopathic neck pain. Thus, more research is needed.

Clinical Application & Conclusions:

Due to the significant difference seen between idiopathic neck pain and healthy groups in the JPE, this test may be clinically useful in examining individuals with idiopathic neck pain. The postural sway test did not show significant differences and other individuals test, although demonstrating some significant differences, were not reported on in enough studies to allow for meta-analysis. Thus, no conclusions can be drawn on their clinical utility.

Although this study found some tests to be potentially of use clinically, it is important to note that it is difficult to differentiate between the subsystems being tested, as they would theoretically all affect the test outcomes. Therefore, these tests may be able to assist in testing sensorimotor control in general, but not individual subsystems.

New tests have been more recently reported in the literature. These include a variation on the subjective visual vertical test, the JPE torsion test, and the Enbloc test. However, more research is needed on these tests before their utility can be determined in an idiopathic neck pain population.

Study Methods:

A search of Allied and Complementary Medicine Database, CINAHL, Cochrane Central Register of Controlled Trials, Embase, MEDLINE, Physiotherapy Evidence Database, Scopus, and SPORTDiscus was performed up to July 2015, utilizing a search strategy agreed upon by the authors. Search terms used were consistent with strategies recently used by the Cochrane Back Review Group (25). Inclusion and exclusion criteria were as follows:

Inclusion Criteria:
  • Studies reporting an outcome measure of the sensorimotor system
  • Studies on individuals with idiopathic neck pain, defined as being perceived in either the upper or lower cervical spine with no trauma at the onset of pain development (10), or healthy individuals
  • Studies excluding individuals with traumatic neck pain
  • Studies involving individuals 18 years of age or older
  • Studies in English or Dutch
  • Observational studies and intervention studies
Exclusion Criteria:
  • Studies on specific neck pathologies and degenerative or inflammatory diseases
  • Studies describing individuals with only headache or radiating pain to the arms, with no neck pain
  • Meetings or conference abstracts
  • Studies not involving living human subjects
2 reviewers screened titles and abstracts first and then full texts. A third reviewer was consulted in cases of disagreement.

The US National Institute of Health’s Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies (11) was used by the same 2 researchers to assess methodological quality and risk of bias. Again, a third reviewer was consulted in cases of disagreement. Studies were not excluded due to a high risk of bias. The first author extracted outcome measurement data. In the case of intervention studies, only the cross-sectional baseline data were extracted. In studies with 2 groups, baseline data were combined according to the Cochrane Handbook for Systematic Reviews of Interventions (11) guidelines. Studies which reported outcomes as medians and interquartile ranges were converted to means and standard deviations (SDs). For the joint position error (JPE) test, findings were reported as the error in degrees, with other measurements converted to degrees.

Study Strengths / Weaknesses:

  • Combining data from different studies resulted in a larger pool of data being available for analysis.
  • The included studies had a low to high risk of bias, most commonly due to a lack of assessor blinding. This could affect the quality of the data.
  • Although tests were described in all studies, the procedures may have differed slightly, leading to the possibility of statistical heterogeneity.
  • There was insignificant data available to perform meta-analysis for most sensorimotor tests.
  • The small sample sizes in some of the pooled studies limits their generalizability.
  • The validity of the tests evaluated is unknown due to the absence of a criterion (or gold) standard.

Additional References:

  1. Brontfort G, Evans R, Anderson AV, et al. Spinal manipulation, medication, or home exercise with advice for acute and subacute neck pain. Ann Intern Med 2012; 156: 1-10.
  2. Driessen MT, Lin CW, van Tulder MW. Cost-effectiveness of conservative treatments for neck pain: a systematic review on economic evaluations. Eur Spine J 2012; 21: 1441-1450.
  3. Hoy DG, March L, Woolf A, et al. The global burden of neck pain: estimates from the global burden of disease 2010 study. Ann Rheum Dis 2014; 73: 1309-1315.
  4. Falla D, Jull G, Hodges P. Training the cervical muscles with prescribed motor tasks does not change muscle activation during a functional activity. Man Ther 2008; 13: 507-512.
  5. Kristjansson E, Treleaven J. Sensorimotor function and dizziness in neck pain: implications for assessment and management. J Orthop Sports Phys Ther 2009; 39: 364-377.
  6. Riemann BL, Lephart SM. The sensorimotor system, part i: the physiologic basis of functional joint stability. J Athl Train 2002; 37: 71-79.
  7. L’Hereux-Lebeau B, Godbout A, Berbiche D, et al. Eavluation of paraclinical tests in the diagnosis of cervicogenic dizziness. Otol Neurotol 2014; 35: 1858-1865.
  8. Humphreys BK. Cervical outcome measures: testing for postural stability and balance. J Manipulative Physiol Ther 2008; 31: 540-546.
  9. Cochrane Back Review Group. Updated search strategies for Cochrane Back Review Group, June 2011. Available at: Accessed March 1, 2015.
  10. Ahmed TS, Oliver M, Blackburn N. Insidious onset neck pain – a symptom not to be dismissed. Ann R Coll Surg Engl 2007; 89: W6-8.
  11. National Heart, Lung, and Blood Institute. Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies. Available at:
  12. Chen X, Treleaven J. The effect of neck torsion on joint position error in subjects with chronic neck pain. Man Ther 2013; 18: 562-567.
  13. Kristjansson E, Dall’Alba O, Jull G. A study of five cervicocephalic relocation tests in three different subject groups. Clin Rehabil 2003; 17: 768-774.
  14. Treleaven J, Takasaki H. High variability of the subjective visual vertical test of vertical perception, in some people with neck pain – should this be a standard measure of cervical proprioception? Man Ther 2015; 20: 183-188.
  15. Sjostrom H, Allum JH, Carpenter MG, et al. Trunk sway measures of postural stability during clinical balance tests in patients with chronic whiplash injury symptoms. Spine 2003; 28: 1725-1734.
  16. Geisinger D, Ferreira E, Suarez A, et al. Head tilt response: a complementary test to the subjective visual vertical. J Vestib Res 2010; 20: 381-389.