The influence of biomechanical characteristics on pain and function outcomes from exercise in medial knee osteoarthritis and varus malalignment: exploratory analyses from a randomised controlled trial

Research output: Contribution to journalJournal articleResearchpeer-review

  • Kim L Bennell
  • Fiona Dobson
  • Ewa M. Roos
  • Søren T Skou
  • Paul Hodges
  • Tim V Wrigley
  • Mary Kyriakides
  • Ben Metcalf
  • Michael A Hunt
  • Rana S Hinman

View graph of relations

Objective: To investigate whether selected biomechanical characteristics influence changes in pain and physical function with exercise in people with medial knee osteoarthritis (OA) and varus malalignment. Methods: Post-hoc exploratory analyses from a randomised controlled trial involving 100 people with medial knee OA and varus malalignment who were randomly allocated to one of two 12-week exercise programs; quadriceps strengthening or neuromuscular exercise. Outcome measures were change in overall average knee pain (visual analogue scale) and self-reported physical function (Western Ontario and McMaster Universities Osteoarthritis Index). Candidate biomechanical characteristics measured at baseline included: i) visually-observed varus thrust during walking; ii) obesity determined by body mass index; iii) static varus alignment; and iv) isometric quadriceps strength. Data were analysed with separate two-way analyses of covariance using the interaction term of exercise group by biomechanical characteristic. Results: 92 participants were analysed for each characteristic except varus thrust where 85 participants were included. For change in pain, there was a significant interaction effect between type of exercise and both varus thrust (p=0.001) and obesity (p=0.023). Neuromuscular exercise was more effective for non-obese participants (mean(95%CI) change=29.5(20.5-38.5)mm) and for those with a varus thrust (mean change=28.7(19.4-38.1)mm), whereas quadriceps strengthening was more effective for obese people (mean change=24.7(14.9-34.4)mm) and for those without a varus thrust (mean change=29.4(21.2-37.7)mm). Biomechanical characteristics did not influence the effect of exercise on physical function (p>0.05). Conclusions: These preliminary findings suggest varus thrust and obesity influence the pain-relieving effects of two different types of exercise. Further research is needed to confirm whether exercise that is prescribed according to specific biomechanical characteristics optimises knee OA outcomes. This article is protected by copyright. All rights reserved.

JournalArthritis Care & Research
Issue number9
Pages (from-to)1281–1288
Number of pages8
Publication statusPublished - Sep 2015