Purpose: Osteoarthritis (OA) is a degenerative musculoskeletal disorder, characterized by cartilage degradation, synovitis and sclerosis of subchondral bone. Results from observational studies have suggested the existence of a distinct form of OA linked to obesity related factors such as type 2 diabetes (T2D) and dyslipidaemia, also known as metabolic OA. However, conclusive evidence for the existence of such a separate form of OA is still lacking. Recently there has been an increase of interest in the relation between metabolic factors and OA, in particular the metabolic syndrome and its components. In this study, we aimed to investigate the relation between several metabolic factors and conditions (diabetes mellitus, metabolic syndrome, and its components) and progression of radiographic osteoarthritis (OA) in knee, hip and hand joints. Methods: We used data from the Rotterdam Study, a large population-based longitudinal cohort aimed at identifying risk factors for cardiovascular, neurological, endocrine and musculoskeletal diseases in the elderly. The present study included 6,191 individuals with up to 12 years of follow-up data for OA (knee, hip and hand) and data on metabolic factors. Progression of radiographic OA in knee and hip joints was defined as a combination of incident OA (i.e. Kellgren/Lawrence (KL) grade <2 at baseline and ≥2 at follow-up) and progression (increase in KL grade starting from 2 or above). An increase of ≥2 in KL sum score in distal Interphalangeal (DIP)/proximal interphalangeal (PIP) joints was defined as progression in hand OA. We analysed the associations of (components of) the metabolic syndrome with progression of radiographic knee and hip OA stratified for sex using multivariate regression models with generalised estimating equations. In the first model we adjusted for age, smoking, alcohol use, education, subcohort, baseline K/L grade and months between radiographs, while in model 2 we additionally adjusted for BMI. In addition, quartiles of each metabolic factor were analysed for risk of OA progression. Results: Metabolic syndrome and two of its components, abdominal obesity and elevated blood pressure, were associated with higher odds of knee OA progression. However, after additional adjustment for BMI (models 2), the association with metabolic syndrome showed a protective and significant association only in women and the ORs for the elevated blood pressure were attenuated and no longer significant (see Table 1), while the association of abdominal obesity disappeared suggesting that it can be explained by BMI. Diabetes and high fasting glucose, another metabolic component, showed significant protective associations only after adjusting for BMI as shown in Table 1. In additional analyses, the upper 3 quartiles of systolic blood pressure had an increased odds of radiographic knee OA progression before (OR 1.44, 95% CI 1.20 - 1.73) and after BMI adjustment (OR 1.27, 95% CI 1.06 - 1.53) when women and men were analysed together. Figure 1 shows the sex stratified analysis, where the effect sizes attenuated slightly after adjustment for BMI and results did not reach significance level anymore. No associations with hip and hand OA progression were found for the metabolic syndrome components that were tested. Furthermore, secondary survival analyses suggested that the protective effects seen for the metabolic syndrome and diabetes were due to competing risks. Conclusions: Metabolic syndrome and its component were not associated with higher risk of progression in hip, knee or hand OA independently from BMI, suggesting that the associations are mainly driven by obesity and high BMI. For diabetes, high fasting glucose and metabolic syndrome (only in women) an apparent protection against knee OA progression was observed in both non-stratified and stratified analysis. However, the apparent protective effect of diabetes and metabolic syndrome against knee OA progression is most probably due to competing risks. The only factor that seems associated with an increased risk for knee OA progression is higher systolic blood pressure.View Large Image Figure ViewerDownload Hi-res image Download (PPT)
Read full abstract