Murdoch University Research Repository

Welcome to the Murdoch University Research Repository

The Murdoch University Research Repository is an open access digital collection of research
created by Murdoch University staff, researchers and postgraduate students.

Learn more

Spinal pain and postural sway. Is there a relationship?

Ruhe, Alexander (2011) Spinal pain and postural sway. Is there a relationship? PhD thesis, Murdoch University.

PDF - Front Pages
Download (411kB)
PDF - Whole Thesis
Download (7MB)


Postural stability is an important component in maintaining upright stance and balance during normal daily movements and activities. Postural stability is also an important factor in the elderly where balance disability may increase the risk of falls and subsequent injury. In sport, problems with balance may lead to serious injuries. Thus, postural stability has important implications in rehabilitation and sports.

Many different methods exist today for assessing postural sway. Centre of pressure (COP) evaluation is a frequently used method of measuring this stability and gain insights into potential pathological mechanisms e.g. in association with pain. This is possible as the COP signal is proportional to ankle torque, a combination of descending motor commands as well as mechanical properties of the musculature around.

Over the last decades, postural sway has been most commonly evaluated by using spatial measures such as sway distance, velocity and area traversed based upon sequential locations of the COP in the plane of the force platform. However, despite its common usage, important clinical aspects of the COP measurements such as its potential suitability for clinical monitoring purposes in pain patients remained unaddressed. Several literature reviews were conducted that identified relevant gaps in current knowledge to focus our research.

This led to the following primary research questions:

a) Can a best evidence experimental setup be identified that is suitable for spinal pain sufferers?

b) Is there a relationship between pain intensity and the COP excursions?

c) Are there alterations in postural sway associated with diminishing pain?

Based on a systematic review of the literature the following experimental protocol was developed: Three measurements of 90sec each were conducted in bipedal narrow stance with closed eyes at a sampling frequency of 100Hz. We selected the COP parameters 90% circle diameter as a descriptor of sway area and mean sway velocity as it has shown its discriminative value for various pain conditions.

The prospective part of this thesis was preceded by pilot studies that confirmed the excellent reliability of the selected experimental setup for mean sway velocity in antero-posterior (AP) and the medio-lateral (ML) direction (ICC2,k 0.85-0.89, 95% CI 0.63-0.97, SEM 0.66-0.78) and 90% circle diameter (ICC2,k 0.80, 95% CI 0.54-0.94, SEM 0.89). Later on, very similar values were observed for sway data obtained from the symptomatic groups.

The experimental setup was found to be safe and a sub-sample of predominantly low back pain patients (n=20) reported no difficulties complying with the postural tasks involved. Furthermore, no effects of learning or fatigue could be demonstrated in 10 healthy individuals either during inter-session (10 consecutive measurements) or intra-session (three times 3 measurements at 2-3 day intervals). No adverse incidents associated with the measurements occurred in approximately 1500 measurements.

By enrolling age matched healthy individuals as a control group (n=77), reference values for the included COP parameters were established to which all subsequent data obtained from symptomatic individuals could be compared.

A total of 210 patients were enrolled subdivided into three groups for non-specific neck, mid back and low back pain. A physical examination was conducted for all pain sufferers, who were asked to rate their pain intensity on a NRS-11 scale. The associated disability was assessed by means of the Disability Rating Index. Depending on the reported severity of their complaint, the symptomatic individuals were subdivided into seven pain intensity groups (NRS 2-8) for each of the painful regions: low back (n=77, n=11/group), mid back (n=63, n=9/group) and neck (n=70, n=10/group).

The symptomatic participants exhibited greater postural sway than healthy controls. As a general trend, a statistically significant increase was reached beginning at about NRS score 4 for all three pain regions. Depending on the COP parameter and painful region, significant differences between individual NRS levels were reached about every 2-3 NRS levels.

Significant differences in COP excursions between mid back, low back and neck pain sufferers could be identified. However, in the light of the expected inter-subject variability in pain perception as well as the low number of participants per NRS group this conclusion warrants caution.

A major finding from a univariate regression analysis was a linear relationship between pain intensity and the COP parameters (p<0.001) for all painful regions, while a multivariate regression analysis showed that other variables such as age, gender, height, weight and BMI did not have a statistically significant effect on postural sway.

This close relationship was maintained even with diminishing pain levels after a course of manual therapy treatments conducted in a group of low back (n=38) and neck pain patients (n=36). In this instance three measurements and interventions were performed at 3-4 day intervals. With few exceptions, the follow-up COP measures in connection with specific pain intensities did not show a significant difference in postural sway compared to reference values for identical NRS levels at baseline.

In addition, a similar linear relationship between pain intensity, the COP sway parameters and the patient's disability ratings was identified for all painful regions.

At the same time, a clear trend towards predominant sway in the medio-lateral direction was observed with increasing pain intensities, until 70% of sway occurred in ML direction at NRS score 8. In comparison, healthy controls showed a nearly equal sway distribution between AP (52%) and ML (48%) direction.

In the absence of learning effects, the reduced COP excursions with decreasing NRS scores in subacute and chronic pain sufferers further suggests that pain interference rather than long-term neuro-physiological adaptations (such as central sensitization) are the primary causative factor for increased sway.

Our findings may have clinical implications for COP measures in patients with significant pain. These include routine sway analyses as an objective outcome measure during the rehabilitation and treatment process. It also stresses the importance of an initial focus on pain regulation rather than proprioceptive training.

Item Type: Thesis (PhD)
Murdoch Affiliation(s): School of Chiropractic and Sports Science
Supervisor(s): Walker, Bruce and Fejer, René
Item Control Page Item Control Page


Downloads per month over past year