Article Text

Download PDFPDF

  1. H. Parente1,
  2. F. Guimarães1,
  3. D. Esperança Almeida2,
  4. S. Azevedo1,
  5. J. Rodrigues1,
  6. D. Faria1,
  7. D. Peixoto1,
  8. J. Tavares-Costa1,
  9. C. Afonso1,
  10. F. Teixeira1
  1. 1Unidade Local de Saúde do Alto Minho, Serviço de Reumatologia, Ponte de Lima, Portugal
  2. 2Hospital de Braga, Serviço de Reumatologia, Braga, Portugal


Background: Osteoarthritis’ (OA) pathophysiology is a complex and evolving subject, resulting in different phenotypes1. There is a known association between cardiovascular risk factors (CVRF) and OA, but its subjacent mechanism remains under-documented2. Pursuing this logic of direct systemic effect on joints by CV comorbidities2, new therapeutic approaches could be reached.

Objectives: To determine the impact of CV comorbidities on knee OA, based on radiographic rate of progression and the need for total knee arthroplasty (TKA).

Methods: Cohort retrospective study among non-obese patients with primary knee OA, seen at a Portuguese hospital between 2017 and 2020. Clinical records were reviewed for sociodemographic and clinical data [age at the diagnosis, radiographic scoring through Kellgren Lawrence (KL) classification system at baseline (diagnosis) and 5 years later, KL progression and time to TKA]. CVRF such as Essential Arterial Hypertension (EHA), Dyslipidemia (DLP) and Type 2 Diabetes Mellitus (T2DM) prior to the OA diagnosis, as well as medication history were also recorded. Comparison of both cohorts, with and without the mentioned CVRF, was performed.

Results: A total of 304 patients with knee OA were eligible, 86 (28.3%) men and 218 (71.7%) women. Of those with CVRF (n=169), 121 (71.6%) were women and the mean age was 74.47±0.59 years. Among them, 129 (76.4%) had EAH, 122 (72.2%) had DLP, and 52 (30.8%) had T2DM. Five (3%) were treated with fenofibrate, 114 (67.5%) with statins, 40 (23.7%) with metformin, and 4 (2.4%) with liraglutide. Of those without CV risk factors (n=135), 97 (71.9%) were women and the mean age was 71.36±0.83 years. The presence of CVRF was associated with higher KL score (moderate to severe) at baseline [OR=2.118 (CI 95%: 1.264-3.549)] and 5 years later [OR=14.960 (CI 95%: 3.439-65.079)], faster KL progression, i.e., 2-grade worsening [OR=2.755 (CI 95%: 1.205-6.299)], and greater need for TKA [OR=3.781 (CI 95%:2.257-6.335)]. However, there was no association between the number of CVRF and progression of the KL score nor the rate of this progression, as well as between the number of CVRF and the need for TKA. EAH was associated with higher KL score at baseline [OR=1.819 (CI 95%: 1.108-2.987)] and 5 years later [OR=17.921 (CI 95%: 2.382-134.835)] but not with faster KL progression. It also associated with a greater need for TKA [OR=3.320 (CI 95%:1.197-5.750)]. DLP was associated with higher KL score at baseline [OR=1.843 (CI 95%: 1.120-3.030)] and 5 years later [OR=7.709 (CI 95%: 1.772-33.530)], as well as with faster KL progression [OR=2.560 (CI 95%: 1.249-5.246)], and greater need for TKA [OR=2.493 (CI 95%: 1.457-4.268)]. T2DM was associated with higher KL score 5 years after OA diagnosis [OR=1.103 (CI 95%: 1.059-1.148)], but not with KL score at baseline nor the occurrence of KL score progression during those first 5 years of disease or the need for TKA. Metformin, liraglutide, statins or fenofibrate use (either compounded or isolated) showed no statistically significant association with any of the following: KL score at baseline and 5 years later, occurrence of KL progression, and need for TKA. Time to TKA was not statistically different between patients with or without these medications. The number of medications used was not associated with KL progression or the need for TKA.

Conclusion: CVRF contribute globally to OA progression. Patients with EAH and DLP showed worse radiographic expression of OA features and greater need for TKA, with DLP patients having faster KL progression. The presence of more than one CVRF was not associated with an increased disease progression, and cardiovascular therapeutic medications were not associated with an overall protective role. This study highlights the importance of acting on preventive measures for CVRF, in OA management.

References: [1] Deveza et al. Clinical and experimental rheumatology vol. 37 Suppl 120,5 (2019): 64-72.

[2] Orcha et al. Revue du Rhumatisme vol. 87 Suppl 1 (2020): A244.

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.