Relationship Between Proprioception and Pain and Disability in People With Non-Specific Low Back Pain

Study Design. Systematic review with meta-analysis. Objective. To examine the association between proprioception and pain and disability in people with non-specific low back pain (NSLBP) and the impact of potential moderators on the strength of the association. Summary of Background Data. NSLBP is a common musculoskeletal disorder and a leading cause of disability. Proprioception deficit has been associated with low back pain and correlated with pain and disability; however, the correlation coefficients presented are conflicting. Methods. Systematic literature search was performed in eight databases and Google scholar. Two independent reviewers assessed the literature and extracted data. Data of interest included participant characteristics of NSLBP group, pain and disability, proprioception test protocol, and the correlation coefficient between proprioception and pain/disability. Meta-analysis of data from homogeneous studies was performed according to proprioception test methods. Subgroup of movement directions was analyzed. The methodological quality of studies was assessed. Results. The literature search identified 3067 records of which 14 studies were included for meta-analysis. Proprioception was measured by active joint repositioning sense (AJRS), passive joint repositioning sense (PJRS), and threshold to detect passive motion (TTDPM). Meta-analysis revealed no significant correlation coefficient between any proprioception measures and pain or disability (P>0.05). The subgroup analysis showed little correlation between absolute error (AE) of passive lumbar flexion and pain (r = 0.225, 95% CI 0.040-0.395), and between AE of passive lumbar extension and disability (r = 0.196, 95% CI 0.010-0.369). Conclusion. AJRS and TTDPM are not correlated with pain and disability in people with NSLBP. The AE of passive lumbar movement is correlated to a small degree with pain and disability. However, the degree of correlation is small and should be interpreted with caution. Therefore, the differences in proprioception observed between people with pain and people without pain seem to be independent of the symptoms.