Statin Use and Knee Osteoarthritis Outcome Measures according to the Presence of Heberden Nodes: Results from the Osteoarthritis Initiative

Radiology. 2019 Nov;293(2):396-404. doi: 10.1148/radiol.2019190557. Epub 2019 Sep 10.


Background The exact contribution of statins to knee osteoarthritis (OA) radiographic outcomes and the characteristics of patients with OA as potential responders to statins remain unclear. Purpose To evaluate the effect of statin use on the incidence of radiographic knee OA (development of Kellgren-Lawrence grade ≥2) and progression of joint space narrowing (JSN) according to the nodal OA status defined according to the presence of Heberden nodes (HNs). Materials and Methods Institutional review boards approved this HIPAA-compliant protocol, and all participants gave informed consent. The Osteoarthritis Initiative (OAI) cohort, which began in 2004 and is ongoing ( identifier, NCT00080171), was used to conduct a longitudinal 1:1 propensity score-matched retrospective analysis of prospectively collected data. Participants were classified as having HN-positive or HN-negative findings according to the presence of HNs at baseline physical examination. In each cohort, per-protocol and new-user design were used to match statin initiators (participants who reported ≤1 year of statin use before enrollment) and nonusers (participants who reported no statin use before enrollment) for variables that potentially contributed to confounding by indication bias. Participants were followed up annually over 8 years. Any associations between statin use and longitudinal knee OA radiographic incidence, JSN progression, or nonacceptable symptomatic state incidence was assessed by using hazard ratios (HRs) of Cox regression. Results In the longitudinal analysis, there were 832 knees of 602 participants (pair-matched knees of statin initiators and nonusers) in the HN-positive cohort (mean age, 64.7 years ± 8.0 [standard deviation]; 377 patients were female [62.6%]) and 386 knees of 285 participants in the HN-negative cohort (mean age, 58.9 years ± 8.2; 144 patients were female [50.5%]). In the HN-positive cohort, statin users had 46% lower risk of JSN progression in comparison with matched nonusers (HR, 0.54; 95% confidence interval [CI]: 0.36, 0.93; P = .02). In contrast, in the HN-negative cohort, statin use had no association with radiographic JSN progression (HR, 1.37; [95% CI: 0.74, 2.53]; P = .32). Conclusion Statin use was associated with reduced risk of radiographic knee osteoarthritis joint space narrowing progression in patients with nodal osteoarthritis. © RSNA, 2019 Online supplemental material is available for this article.

Publication types

  • Clinical Trial
  • Multicenter Study
  • Observational Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Disease Progression
  • Exostoses / diagnostic imaging*
  • Exostoses / drug therapy*
  • Female
  • Humans
  • Hydroxymethylglutaryl-CoA Reductase Inhibitors / therapeutic use*
  • Longitudinal Studies
  • Male
  • Middle Aged
  • Osteoarthritis, Knee / diagnostic imaging*
  • Propensity Score
  • Retrospective Studies


  • Hydroxymethylglutaryl-CoA Reductase Inhibitors

Associated data