Elsevier

Gait & Posture

Volume 49, September 2016, Pages 73-77
Gait & Posture

Full length article
Pain catastrophizing and trunk muscle activation during walking in patients with chronic low back pain

https://doi.org/10.1016/j.gaitpost.2016.06.025Get rights and content

Highlights

  • Pain catastrophizing is associated with altered neuromotor behavior during gait.

  • Catastrophizers had higher EMG activation in certain trunk muscles during gait.

  • Catastrophizers had less phasic modulation of other trunk muscles during gait.

  • Activation of most muscles was correlated with pain catastrophizing scores.

Abstract

It has been hypothesized that individuals with low back pain (LBP) will have higher trunk muscle activity during gait, in an attempt to limit spine motion, and that this “guarding strategy” may be influenced by the person’s psychological response to pain. This study investigated whether the amplitude of trunk muscle activation differs between persons with chronic LBP and healthy individuals during walking, and whether changes in muscle activation were related to pain catastrophizing. Thirty persons with chronic non-specific LBP, stratified into 2 groups of high (HLBP) and low (LLBP) pain catastrophizing, were contrasted with a control group of 15 healthy individuals during walking on a treadmill at a self-selected speed. Surface electromyographic (EMG) data were recorded from 10 trunk muscles. The effects of Group and gait Sub-phase on EMG activation amplitudes were assessed. The HLBP group exhibited higher activation of certain muscles throughout the gait cycle, and reduced variability of others at specific sub-phases of gait. A significant correlation was found between activation amplitude and pain catastrophizing in most muscles, when controlling for gait speed and pain intensity. These data indicate that altered trunk muscle activation is present in some patients with LBP during walking, but does not represent a universal increase in activation for all muscles. This altered neuromotor control is, however, more strongly associated with pain catastrophizing than with pain intensity, and appears to represent a non-functional, maladaptive behavior, as it alters the normal, phasic pattern of activation in certain trunk muscles.

Introduction

Individuals with low back pain (LBP) tend to walk slowly [1], which may be a means of reducing spine motion [2]. This should normally lead to lower trunk muscle activation [3], but there is evidence that some individuals with LBP increase trunk muscle activity during gait, possibly as a “guarding strategy” to further limit spine motion [4], [5]. Furthermore, it has been suggested that muscle guarding is more closely related to the psychological response to pain than to pain intensity [6].

There are conflicting findings on pain-induced trunk muscle guarding during gait. Arendt-Nielsen et al. [7], for example, reported a decrease in the phasic nature of EMG patterns during gait in individuals with chronic LBP, which could reflect an attempt to splint the spine. Lamoth et al. [8], on the other hand, found that induced pain has only minimal effects on the global pattern of lumbar erector spinae activity. The influence of psychological factors on trunk muscle activity during gait is also unclear. Two previous studies have found no relationship between fear-avoidance beliefs and lumbar EMG [4], [9]. Pain catastrophizing, however, has been linked to increase erector spinae activity during gait [10].

The objective of this cross-sectional study is twofold: first, to investigate whether the amplitude of trunk muscle activation differs between individuals with and without chronic LBP, when walking on a treadmill at a self-selected speed; second, to determine if the amplitude of trunk muscle activation is affected by pain catastrophizing in individuals with chronic LBP. We hypothesized that higher trunk muscle activation is used by persons with LBP during walking as a guarding strategy that can be influenced by pain catastrophizing.

Section snippets

Participants

Thirty individuals with chronic LBP were divided into two age- and gender-matched groups, based on their scores on the Pain Catastrophizing Scale (PCS) [11]: high (HLBP: PCS  21/52) and low (LLBP: PCS  20/52). The cut-off value of 20/52 was chosen based on the median value for the PCS, as reported in the PCS User Manual [12]. Inclusion criteria for both groups were: nonspecific LBP (excluding serious pathology, spinal stenosis and radiculopathy [13]) primarily located between the lower ribs and

Results

A significant main effect of Group was found for the percentage sub-MVIC EMG amplitude of MF, bilaterally (RMF: F = 4.546, p = 0.016; LMF: F = 3.998, p = 0.026) as well as for LRA (F = 4.604, p = 0.016). Post-hoc comparisons revealed that amplitudes for these 3 muscles were significantly higher for the HLBP group than for controls (Fig. 1).

Significant interactions between Group and Sub-phase of gait were found in EO (REO: F = 3.719, p = 0.015; LEO: F = 3.477, p = 0.014) and ESI (RESI: F = 2.749, p = 0.033; LESI: F = 

Discussion

Our findings indicate that, at self-selected gait speeds, the activation amplitude for certain trunk muscles is higher among individuals with chronic LBP who exhibit high pain catastrophizing (HLBP) than for healthy controls, when normalized to a standardized, sub-maximal task. This pattern was significant across the whole gait cycle (main effect of Group) for 3 of the 10 trunk muscles assessed (LRA and MF bilaterally), with another 2 (EO bilaterally) showing similarly elevated levels of

Conclusion

The findings of this study provide evidence for an association between pain catastrophizing and altered neuromotor behavior during gait, among individuals with LBP. This underscores the biopsychosocial nature of LBP, where the psychological response to pain may influence not only the patient’s perception of their condition, but may manifest in physical behaviors that are generally controlled at a subconscious level. The findings of the current study also provide some support to the “guarding

Conflicts of interest

The authors have no relevant financial activities and no conflicts of interest to disclose in relation to this work.

Acknowledgements

This work was supported partially by funding from the Physiotherapy Foundation of Canada, the Quebec Rehabilitation Research Network (REPAR), and the Quebec health research funds (FRQS).

References (27)

Cited by (0)

View full text