Article Text

THU0490 Arterial Stiffness and Ankle-Brachial Index in Patients With Osteoarthrtis-No Evidence of Increased Risk of Cardiovascular Disease
  1. S.A. Provan1,
  2. S. Rollefstad2,
  3. E. Ikdahl2,
  4. I.J. Berg1,
  5. T.K. Kvien1,
  6. I. Haugen1,
  7. A.G. Semb2,
  8. N. Østerås3
  1. 1Rheumatology
  2. 2Preventive Cardio-Rheuma Clinic
  3. 3National resource center for rehabilitation in rheumatology, Diakonhjemmet Hospital, Oslo, Norway


Background Osteoarthritis (OA) may be a risk factor for cardiovascular disease (CVD).

Objectives To investigate whether clinical OA is associated with increased arterial stiffness or a pathological ankle-brachial index (ABI).

Methods Residents of Ullensaker municipality, aged 40- 79 years were mailed a questionnaire concerning musculoskeletal pain. Respondents with self-reported OA were invited to a comprehensive medical examination including assessments of CVD risk. Hip, knee and hand OA was diagnosed according to ACR clinical classification criteria (1). A dichotomized variable; presence vs absence of OA, and an composite variable; number of joints with OA, were constructed. The augmentation index (AIx) and pulse wave velocity (PWV), two validated markers of arterial stiffness and CVD, were recorded by the sphygmocor apparatus. ABI, a marker of atherosclerosis, was estimated and dichotomized according to standardized cut-off levels.

The association of the constructed OA variables to PWV, AIx, systolic BP and ABI were examined in linear and logistical regression models respectively. Education, body mass index (BMI), smoking, presence of diabetes, prior CVD, use of statin/antihypertensive medication, use of nonsteroidal anti-inflammatory drugs, C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), presence of clinical hip, knee and hand were successively entered as possible covariates in age and gender adjusted models. Variables associated with the dependent variable at p≤0.1 were entered into multivariate models. OA variables were re-entered into the final multivariate models.

Results 630 respondents had self-reported OA. Mean age (SD) 63.9 (8.8). 438 (69.5%) were female. 257 were subsequently diagnosed with clinical OA. There were no associations between clinical OA and markers of CVD risk (Table) and presence of OA was not significantly related to the outcomes in the final multivariate models.

In multivariate analyses: PWV was associated to age (β(SE)) (0.10 (0.01)), male sex (-0.09 (0.19)), diabetes (1.37 (0.36)), smoking (-0.53 (0.25)), BMI (0.03 (0.01)), systolic BP (0.05 (0.01)) and CRP (0.07 (0.02)), AIx was associated to age (0.35 (0.04)), male sex (-7.19 (0.71)), smoking (-3.89 (0.97)) and diastolic BP (0.25 (0.04)). Systolic BP was associated to age (0.71 (0.09)), male sex (2.92 (0.17)), education (-0.01 (0.01)) and use of statins/antihypertensive medications (4.48 (1.63)). ABI was associated to age (OR (SE))(1.03 (0.02)), male sex (2.06 (0.36)), CRP (1.10 (0.04)) and diabetes (3.77 (0.58))

Table 1

Conclusions We found no evidence of increased risk of CVD in patients with clinical OA when compared to respondents with self-reported OA.


  1. Østerås N, The MUST protocol BMC Musculoskeletal Disorders 2013 14:201

Disclosure of Interest 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.