Low-back pain, a consequence of cumulative mechanical loading?

In a recent reaction posted on this website, it was stated that systematic reviews typically find little or no relationship between physical loading and low-back pain. The author cited three reviews, seemingly providing solid evidence in support of this statement. However, these reviews are part of a series produced by the same group, according to the same methodology, reviewing largely the same studies [for an overview see 1] and thus do not provide independent observations to underpin this statement. Moreover, these reviews were quite heavily criticised for methodological errors and biases [2-6] and the results were certainly not typical, as several other reviews [7-11] and an extensive meta-analysis [12] reached a different conclusion.

The book is by no means closed on the role of physical loading in the aetiology of low-back pain. Assessment of exposure-response relationships, i.e. the quantitative relationships between the intensity, frequency and duration of physical loading and the incidence of low-back pain, is key [1, 9], but the way in which exposure-response relationships have traditionally been studied is problematic. For example, the mass of objects lifted has been used as a measure of intensity of exposure to lifting, even though object mass [13, 14] has only limited influence on the magnitude of mechanical loading of the back. Other activities like bending the trunk or pulling may cause similar loading of the back to that caused by lifting.

Methods to estimate mechanical loading of the back suitable for large-scale studies are needed, and have recently been developed and tested [e.g. 15, 16]. They have allowed us to perform the first study in which mechanical back loads were related to the development of low-back pain [17]. Over a three-year period, low back pain was assessed in 1131 participants who had been categorised into 19 groups based on their occupational activities. Video and hand force recordings of four or five workers per group, obtained during their work at the start of the three-year period, were used to determine the inputs for a mechanical model that estimates low-back loads. Mechanical loads were expressed as torques generated by the trunk muscles, and peak torques and cumulative torques (torque integrated over time) were calculated and averaged to obtain estimates of back loads in each group.

A significant exposure-response relationship was found for cumulative low-back torques and low-back pain, with a 2.4 to 5.1 times higher risk of low-back pain in the most exposed as compared with the least exposed participants. The result was strongly determined by the group that experienced the highest back loads, which mainly consisted of road workers. The latter may indicate that the exposure-response relationship is not linear, with increased risk only for people that are exposed to loads at the high end of the spectrum. Overall, this study supports an aetiological model of low-back pain based on cumulative tissue damage due to mechanical loading. No exposure-response relationship was found for peak loading.  Therefore, although the study design was not optimally designed to reveal a role of incidental events, our data do not support an aetiological model based on acute injury.

Our results imply that we should reduce cumulative low-back loads to prevent low-back pain, for example by reducing handling of heavy loads and working in awkward body postures. However, there is a caveat. The load bearing capacity of the spine is largely determined by its bone mineral content, which is a measure of its size and density [18-22]. Thus the risk of injury is expected to be lower in people with a high bone mineral content. In line with this, the probability of low-back pain was found to be lower in men with a high compared to a low bone mineral content [23]. Now, it is well known that unloading decreases [24] and loading increases bone mineral content [25, 26]. So physical activity and resulting low-back load may also have positive effects with respect to low-back pain, as they are known to for recovery from back pain [27]. Perhaps this does not come as a surprise. In sports, the benefits of physical activity (training) as well as the risks (overuse injury) are well recognised. Fortunately, it seems that unloading needs to be fairly dramatic before detectable loss of bone mineral content occurs [24] and our study suggested that cumulative loads need to be quite high before they increase the risk of low-back pain. This implies that we should look for healthy levels of loading and there appears to be enough middle ground in between the extremes of disuse and overuse to be able to do that.

About Jaap

Jaap van DieenJaap van Dieën is professor of biomechanics at the Faculty of Human Movement Sciences of the VU University Amsterdam. He has been active in biomechanics since 1990, initially with a focus on the biomechanics of the spine in relation to low-back pain and more recently on the biomechanics of posture and gait in healthy and impaired populations. His main interest is how neural control interacts with the musculoskeletal system in functional activities and how this affects musculoskeletal loading. In recent years, he has collaborated extensively with epidemiologists in applying biomechanical methods outside the laboratory, to obtain insight in risk factors for disorders and injuries.

 References

  1. Kwon DK, Roffey DM, Bishop P, Dagenais S, Wai EK. Systematic review: occupational physical activity and low back pain. Occ Med 2011;61:541-8.
  2. McGill SM. Re: Causal assessment of occupational lifting and low back pain: results of a systematic review by Wai et al. Spine J 2011;11:365-6.
  3. Takala EP. Lack of “statistically significant” association does not exclude causality. Spine J 2010;10:944.
  4. Kuijer PP, Frings-Dresen MH, Gouttebarge V, van Dieen JH, van der Beek AJ, Burdorf A. Low back pain: we cannot afford ignoring work. Spine J 2011;11:164.
  5. Kuijer PP, Takala EP, Burdorf A, Gouttebarge V, van Dieen JH, van der Beek AJ, Frings-Dresen MH. Low back pain: doesn’t work matter at all? Occup Med 2012;62:152-3.
  6. van Dieen JH, Kuijer PP, Burdorf A, Marras WS, Adams MA. Non-specific low back pain. Lancet 2012;379:1874.
  7. Hoogendoorn WE, Poppel MNMv, Bongers PM, Koes BW, Bouter LM. Physical load during work and leisure time as risk factors for back pain. Scand J of Work Env Health 1999;25:387-403.
  8. Bakker EW, Verhagen AP, van Trijffel E, Lucas C, Koes BW. Spinal mechanical load as a risk factor for low back pain: a systematic review of prospective cohort studies. Spine 2009;34:E281-93.
  9. Lotters F, Burdorf A, Kuiper J, Miedema H. Model for the work-relatedness of low-back pain. Scand J of Work Env Health 2003;29:431-40.
  10. Kuiper J, Burdorf A, Verbeek JHAM, Frings-Dresen MHW, Beek AJvd, Viikari-Juntura ERA. Epidemiologic evidence on manual materials handling as a risk factor for back disorders: A systematic review. Int J Ind Ergonomics 1999;24:389-404.
  11. da Costa BR, Vieira ER. Risk factors for work-related musculoskeletal disorders: A systematic review of recent longitudinal studies. Am J Ind Med 2010;53:285-323.
  12. Griffith LE, Shannon HS, Wells RP, Walter SD, Cole DC, Cote P, Frank J, Hogg-Johnson S, Langlois LE. Individual participant data meta-analysis of mechanical workplace risk factors and low back pain. Am J Public Health 2012;102:309-18.
  13. Faber GS, Kingma I, van Dieen JH. Effect of initial horizontal object position on peak L5/S1 moments in manual lifting is dependent on task type and familiarity with alternative lifting strategies. Applied Ergonomics 2011;54:72-81.
  14. Hoozemans MJ, Kingma I, de Vries WH, van Dieen JH. Effect of lifting height and load mass on low back loading. Applied Ergonomics 2008;51:1053-63.
  15. Coenen P, Kingma I, Boot CR, Faber GS, Xu X, Bongers PM, van Dieen JH. Estimation of low back moments from video analysis: a validation study. J Biomech 2011;44:2369-75.
  16. Coenen P, Kingma I, Boot CR, Bongers PM, van Dieen JH. Inter-rater reliability of a video-analysis method measuring low-back load in a field situation. Appl Ergon 2013;44:828-34.
  17. Coenen P, Kingma I, Boot CR, Bongers PM, van Dieën JH. Cumulative mechanical low-back load at work is a determinant of low-back pain. Occ Env Med 2014;71 (5), 332-7 PMID: 24676271
  18. Brinckmann P, Biggemann M, Hilweg D. Fatigue fracture of human lumbar vertebrae. Clin Biomech 1988;3:s1-s28.
  19. Brinckmann P, Biggemann M, Hilweg D. Prediction of the compressive strength of human lumbar vertebra. Clin Biomech 1989;4:1-29.
  20. van Dieen JH, Toussaint HM. Evaluation of the probability of spinal damage caused by sustained cyclic compression loading. Hum Factors 1997;39:469-80.
  21. Bisschop A, van Royen BJ, Mullender MG, Paul CP, Kingma I, Jiya TU, van der Veen AJ, van Dieen JH. Which factors prognosticate spinal instability following lumbar laminectomy? Eur Spine J 2012;21:2640-8.
  22. Bisschop A, Kingma I, Bleys RL, Paul CP, van der Veen AJ, van Royen BJ, van Dieen JH. Effects of repetitive movement on range of motion and stiffness around the neutral orientation of the human lumbar spine. J Biomech 2013;46:187-91.
  23. Hoozemans MJ, Koppes LL, Twisk JW, van Dieen JH. Lumbar bone mass predicts low back pain in males. Spine 2012;37:1579-85.
  24. Alexandre C, Vico L. Pathophysiology of bone loss in disuse osteoporosis. Joint Bone Spine 2011;78:572-6.
  25. Bolam KA, van Uffelen JG, Taaffe DR. The effect of physical exercise on bone density in middle-aged and older men: a systematic review. Osteoporos Int 2013;24:2749-62.
  26. Martyn-St James M, Carroll S. Progressive high-intensity resistance training and bone mineral density changes among premenopausal women: evidence of discordant site-specific skeletal effects. Sports Med 2006;36:683-704.
  27. Dahm KT, Brurberg KG, Jamtvedt G, Hagen KB. Advice to rest in bed versus advice to stay active for acute low-back pain and sciatica. Cochrane Database Syst Rev 2010:CD007612.

 

Comments

  1. I’d be interested if you were considering dividing groups according to suspected tissue generators. This seems possible even using simple criteria such as location based upon joint injection provocation studies, age and response to quadrant test for eg

    Thankyou