Background Obesity is associated with a 2-fold increase of hand osteoarthritis (HOA). However, the association with MetS or its components, independently of the weight, has never been investigated in HOA.
Objectives Since HIV patients display an increased risk of MetS, we addressed whether MetS or its components may increase the risk of HOA in a cohort of HIV patients w/o MetS.
Methods A case-control study included HIV-infected patients with MetS (IDF/AHA criteria) and aged 45-65. They were paired with HIV-infected controls without MetS on age, gender, HIV-RNA level and duration of HIV infection. Hand X-rays were performed and assessed by two readers, after checking for their reproducibility. HOA was defined as a Kellgren-Lawrence score (KL) ≥2 on ≥1 joint. Radiographic severity of HOA was assessed by the sum of KL scores across all joints and by the number of joints having KL≥2. HOA prevalence was compared to the one found in the Framingham study. Logistic and linear regression models were used to determine risk factors of HOA.
Results 301 patients (88% male, mean age±SD: 53.4±5.0 years, mean estimated duration of HIV-infection 18±7) years, mean body mass index (BMI ± SD: 24.9±5.8) were included, of whom 152 were cases (BMI ± SD: 26.4±4.8) and 149 (BMI ± SD: 23.5±6.5) were controls. Overall HOA prevalence was 55.6% and was higher in MetS+ (64.7%) than MetS- patients (46.3%, p=0.002). Thumb base OA (i.e., rhizarthrosis (RZ)) was also more frequently observed in MetS+ patients (26.1 vs 14.1%, p=0.01). Concerning the severity of HOA, mean (SD) sum of KL score was higher in MetS+ (6.8±0.9) than MetS- patients (3.7±0.5), p=0.002. The number of affected joints was higher in MetS+ (3.2±0.4) than MetS- patients (1.8±0.2); p=0.002. In multivariate analysis, MetS independently increased the risk of HOA (OR=2.23, 95%CI: 1.26-3.96; p=0.002) as well as age (OR=1.18, 95%CI:1.12-1.25; p=0.00001) age. A similar tendency was also observed for RZ (MetS: OR=1.86, 95%CI:0.98-3.45; age: OR=1.08, 95%CI: 1.02-1.25). MetS was associated with more severe HOA: in multivariate analysis, it was independently correlated with KL scores sum (β=2.1; p=0.04) and with the number of OA joints (β=1.02; p=0.04). Age was also associated with these 2 features (KL scores sum: β=0.43, p=0.0001; number of OA joints: β=0.21; p=0.0001). Considering each metabolic component separately, insulinoresistance assessed by HOMA-IR tended to be associated with HOA (p=0.06) in the univariate (but not multivariate) analysis and was correlated with the number of OA joints (p=0.05). No association between BMI, dyslipidemia and HOA was observed. Finally, prevalence of HOA was higher in the HIV-infected population than in the general population in the same age group (56% for men in the present study vs 38% in the Framingham study). No association was found between HOA and previous/current exposure to protease inhibitors or HIV features (viral load, CD4 count, T4/T8 rate, duration of HIV infection).
Conclusions We show for the 1st time the link between HIV and OA, through the accelerated aging and MetS. Thanks to this unique cohort worldwide, we demonstrate an association of HOA with MetS independently of obesity (i.e. BMI).
Haugen H Ann Rheum Dis 2011
Disclosure of Interest None declared
Statistics from Altmetric.com
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.