Association Between Knee Load and Pain: Within-Patient, Between-Knees, Case–Control Study in Patients With Knee Osteoarthritis

Authors

Trevor B. Birmingham, The University of Western Ontario
Kendal A. Marriott, The University of Western Ontario
Kristyn M. Leitch, The University of Western Ontario
Rebecca F. Moyer, Dalhousie University
Amanda L. Lorbergs, The University of Western Ontario
Dave M Walton, The University of Western Ontario
Kevin Willits, The University of Western Ontario
Robert B. Litchfield, The University of Western Ontario
Alan Getgood, The University of Western Ontario
Peter J. Fowler, The University of Western Ontario
J. Robert Giffin, The University of Western Ontario

Document Type

Article

Publication Date

5-1-2019

Journal

Arthritis Care and Research

Volume

71

Issue

5

First Page

647

Last Page

650

URL with Digital Object Identifier

10.1002/acr.23704

Abstract

© 2018, American College of Rheumatology Objective: The association between knee loading and pain in patients with knee osteoarthritis is reported to be low and of questionable importance, but may be confounded by several factors that differ between patients. We aimed to elucidate the association between dynamic knee load and pain by minimizing confounding using a study design that was within the same patient, with knees discordant for pain. Methods: A total of 265 patients with knees discordant for pain (530 knees) rated the pain in each knee before and after walking for 6 minutes, and then underwent 3-dimensional gait analysis. Results: The peak knee adduction moment and knee adduction impulse (proxies for medial knee loading) were associated with increased pain (odds ratio [OR] 2.43 [95% confidence interval (95% CI) 1.77–3.33] and OR 6.62 [95% CI 3.46–12.7], respectively) and remained significant after controlling for radiographic disease severity. When split into quartiles, ORs indicated knees in the highest loading quartile had greater odds of experiencing increased pain with walking (OR 4.7 95% CI 2.3–9.5] for peak adduction moment; OR 9.0 [95% CI 4.0–20.1] for adduction impulse) compared to knees in the lowest loading quartile. Conclusion: When between-patient confounding is minimized, there is a strong association between medial knee load and increased knee pain during walking.