Article Text

FRI0570-HPR Lumbar Repositioning is Influenced by Nonspecific Low Back Pain, Test Setup and Body Mass Index
  1. C. Bauer1,
  2. M. Ernst2,
  3. F. Rast2,
  4. S. Schelldorfer2,
  5. A. Meichtry2,
  6. J. Kool2,
  7. J. Suni3,
  8. M. Kankaanpää1
  1. 1University of Tampere, School of Medicine, Tampere, Finland
  2. 2Departement of Health, ZHAW, Winterthur, Switzerland
  3. 3UKK Institute, Tampere, Finland


Background Low back pain (LBP) affects up to 84% of people in industrialized countries [1]. In 2005 the total direct costs of LBP in Switzerland amounted €2.6 billion [2]. Several studies have investigated deficits in proprioception in non specific LBP (NSLBP) patients by investigating lumbar spine position sense. These studies measured lumbar repositioning error (RE), with conflicting results regarding the association of LBP and RE [3-5]. Comparison between these studies is difficult as there are large differences in the measurement procedure.

Objectives The aim of this study therfore was to analyse if RE is influenced by sub-acute nonspecific low back pain (NSLBP), the test setup and individual factors and to propose a good test setup.

Methods We measured lumbar RE (absolute, constant, variable & mean squared) with three test setups in a cross-sectional, laboratory study. 60 patients with NSLBP and 31 healthy participants were included. All participants were instructed to move from neutral position through half of their maximal lumbar range of motion and to reproduce the neutral position. We analysed Group, Test & GroupXTest effects on RE and adjusted for individual factors such as body mass index (BMI), by fitting a linear mixed model for every main outcome measure.

Results NSLBP patients produced a greater absolute (38% difference between groups), constant (60%) and mean squared error (70%) compared to controls. There was no group effect on variable error (8%). NSLBP patients were less accurate but not less consistent. The test setup influenced the magnitude and direction of RE. BMI influenced absolute and mean squared error. The magnitude and direction of RE depended on the test setup.

Conclusions Tests for lumbar RE discriminate NSLBP patients from healthy participants but should be viewed with caution as their outcome depends on the test setup and BMI. We recommend two test setups and normalization for BMI for further use in clinical studies.


  1. Walker BF. J Spinal Disord. 2000;13:205-17.

  2. Wieser S. et. al. Eur J Health Econ. 2011;12:455-67.

  3. Asell M. et. al. Arch Phys Med Rehabil. 257 2006;87:1170-6.

  4. Brumagne S. Spine. 267 2000;25:989-94.

  5. O'Sullivan K. et. al. Manual Therapy. 2013;18:526-32.

Acknowledgements The Swiss Commission of Technical Innovation (CTI) funded this study. We thank Hocoma AG, (CH) for providing hardware and software.

Disclosure of Interest None declared

DOI 10.1136/annrheumdis-2014-eular.5032

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.