Decreased limits of stability in response to postural perturbations in subjects with low back pain

Abstract

Background

Low back pain is associated with abnormal movement strategies due to changes in neuromuscular control. A plausible contributing factor to low back pain is poor control of trunk muscles, thus understanding motor control alterations in this population can guide rehabilitation. Quantification of postural responses following support surface translations is one way to examine motor control impairments in people with low back pain.

Methods

Twenty-four healthy subjects [mean 33 (SD 11) years] who had no low back pain and 26 subjects [mean 39 (SD 13) years] with chronic, recurrent low back pain were instructed to stand with feet placed on separate force plates, which were mounted on a moveable platform. The platform was translated unexpectedly in one of 12 directions for a total of 72 trials. For both the sagittal and frontal planes, the net center of pressure displacement was derived and the total body center of mass was calculated by combining kinematic and anthropometric data.

Findings

For sagittal plane responses, subjects with low back pain had reduced and delayed sagittal plane center of pressure responses (P < 0.01) compared to the subjects without low back pain. In contrast, the sagittal plane center of mass responses were larger in magnitude (P = 0.03) yet similarly delayed in onset (P = 0.04) for the low back pain group. Frontal plane responses did not differ between groups.

Interpretation

Subjects with low back pain have altered automatic postural coordination, both in terms of magnitude and timing of responses, indicating alterations in neuromuscular control.

Introduction

Little is known about the causes of low back pain (LBP), but poor neuromuscular control of trunk muscles has been identified as a plausible contributing factor (Van Dieen et al., 2003). Several studies have described motor control impairments, such as abnormal muscle recruitment and/or activation levels, associated with LBP (Radebold et al., 2001), however, it has yet to be demonstrated whether or not these impairments are the cause or simply the consequence of LBP. Despite this unknown, several treatment regimes designed to ameliorate these motor impairments are currently used for people with recurrent LBP (Koumantakis et al., 2005), chronic LBP (Richardson et al., 1999), and LBP secondary to spondylolisthesis (O’Sullivan and Allison, 1997). One of the underlying assumptions of these treatment programs is that a causal relationship exists between motor control impairments and LBP, thus improvements in control of movement (voluntary and reactive) will ultimately result in reduction of LBP in affected individuals. However, these treatment regimens are not well understood with regard to how they influence motor control impairments and/or reduce LBP, and it remains unclear which treatments have the greatest potential to improve pain and function in persons who are classified into particular subgroups of LBP. Before treatments can be developed to effectively target these motor control impairments we need a clearer understanding of the nature of the relationship between LBP and neuromotor deficits.

In order to gain insight into the neuromuscular control impairments associated with LBP, paradigms that examine automatic, anticipatory and voluntary responses to functional and well-controlled laboratory based tasks are needed in addition to paradigms that manipulate central versus peripheral, feedforward versus feedback, and central set mechanisms. Paradigms used to quantify changes in neuromuscular control in subjects must be independent of any intervention examined (Cacciatore et al., 2005) to avoid improvements simply due to practice and learning of the paradigm. This broader approach could be used to examine altered neuromuscular control in different subpopulations of people with LBP and to determine how the identified motor control impairments are affected, positively or negatively, by different interventions. The authors believe the surface translation paradigm (Henry et al., 1998a, Henry et al., 1998b), such as the one used in the current study, provides an ideal method with which to examine neuromuscular responses in persons with LBP by providing a systematic postural perturbation that is followed by a well-documented automatic postural response not under voluntary control.

To date, alterations in neuromuscular control associated with LBP have been documented using static (Mientjes and Frank, 1999) and dynamic tasks (Radebold et al., 2001) that require reactive and/or anticipatory postural responses. In sitting tasks while maintaining isometric trunk exertions, subjects with LBP demonstrated delayed and variable trunk muscle latencies and a greater degree of trunk muscle co-contraction in response to a sudden load release (Radebold et al., 2000). LBP subjects also demonstrated poorer postural control during unstable sitting tasks, especially with eyes closed as demonstrated by increased center of pressure (CP) excursions (Cholewicki et al., 2000). While in the standing position, subjects with LBP demonstrated asymmetric activation of the rectus abdominis and erector spinae muscles in response to a forward surface translation (Newcomer et al., 2002) compared to the control group; however, the muscle latencies were similar between the two groups. These studies have recruited persons who reported having a recurrence of LBP at the time of testing or persons who reported chronic LBP with no documentation provided with regard to recurrence or remission. The presence of increased pain (above the baseline level) can confound the findings further because one does not know if the documented altered neuromuscular responses are the result of the increased pain (Moseley et al., 2003) or if the alterations are a pre-existing underlying contributor to the LBP.

The authors are unaware of studies that have examined postural responses in people with recurrent, mechanical LBP during a period when the person was in ‘remission,’ or little to no pain (Von Korff, 1994). Documenting neuromuscular responses during a quiescent period of pain can provide a preliminary understanding about whether or not the identified motor control impairments are the cause or the consequence of the LBP. Additionally, little is known about the extent and in what manner automatic postural responses are altered in the LBP population and whether alterations differ across subgroups of LBP subjects. Thus, the purpose of this study was to characterize the differences in automatic postural responses among persons without and with chronic, recurrent mechanical LBP (while the latter had little to no pain) by quantifying center of pressure and center of mass responses following systematic support surface perturbations.