Article Text

Download PDFPDF

  1. M. Barguil Macedo1,
  2. V. M. Ostrovski Souza Santos1,
  3. R. M. Pereira1,
  4. R. Fuller1
  1. 1University of Sao Paulo Faculty of Medicine, Rheumatology, Sao Paulo, Brazil


Background: An association between osteoarthritis (OA) and atherosclerosis (AT) has long been suggested, with its plausibility based on several guessed commonalities, from shared risk factors to overlapping molecular pathways, given that both conditions represent a state of chronic low-grade inflammation. However, meta-analyses of the impact of OA on hard cardiovascular (CV) endpoints have displayed conflicting results. Those disparate results probably stemmed from the heterogeneous OA population adjoined, as it has been postulated that OA denotes not a single, but a group of diseases with particular pathogenic mechanisms. Also, CV mortality may capture just the extreme of a continuum of atheromatous burden, and miss earlier subclinical disease.

Objectives: In order to discern a more refined association between OA and AT, a systematic review and a meta-analysis of studies that focused on OA of weight-bearing joints (i.e., hip and knee) were performed, and the results stratified by clinical and subclinical AT.

Methods: A search on PubMed, Web of Science, LILACS, and SciELO from inception until August 2020, was done using the MeSH search terms “osteoarthritis”, “aorta”, “carotid”, “intima-media thickness”, “coronary artery disease”, “atherosclerosis”, “cardiovascular disease”, and “death”. To appraise the quality of the studies, we applied the NewCastle-Ottawa scale. To assess for heterogeneity, I2 was used. A random-fixed effect model was adopted, and outliers were excluded when detected. Publication bias was ascertained qualitatively by funnel plot, and quantitatively by Egger regression test. Analyses were performed on RStudio version 1.3.1093.

Results: The search retrieved 1335 citations, of which 168 were screened for abstract reading. From those, 127 were selected for full-text reading. After exclusion of duplicates and out of scope aricles, a total of 23 studies, comprising 151,187 individuals with OA and 287,444 controls, were included on the narrative synthesis, and 17 on the meta-analysis. All but one study were deemed as of fair or good quality. Fifteen studies found a positive association between OA and AT, the remaining 8 did not reach statistical significance. Four studies subdivided the OA patients by gender, and in half of those a positive association was found exclusively on females. Thirteen studies worked with a subclinical definition of AT, 1 included both subclinical and manifest disease, and the remainder included only manifest disease. Hip and knee OA increased the risk for both subclinical AT (OR 1.15, 95% CI 1.01-1.31, Figure 1A), and CV disease (OR 1.13, 95% CI 1.05-1.22, Figure 1B), but not for CV death (OR 1.08, 95% CI 0.99-1.19).

Conclusion: Hip and knee OA are associated with AT. This may have practical implications on the management of OA patients, as a more intensive control of traditional CV risk factors may be advised.

Figure 1.

Forest plots showcasing the association between hip and knee OA and subclinical AT (A) and CV disease (B).

Acknowledgements: Statistical analysis was performed under the supervision of Danielle Daffre Carvalho.

Disclosure of Interests: None declared.

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.