Serum carboxymethyllysine concentration is associated with erosive hand osteoarthritis.

OBJECTIVE: Carboxymethyllysine (CML) and homocitrulline (HCit) are the products of two non-enzymatic post-translational modifications of protein, a process related to age. We investigated whether serum CML and HCit concentrations were associated with hand osteoarthritis (HOA), especially erosive HOA. DESIGN: Serum CML and HCit were measured by using liquid chromatography coupled with tandem mass spectrometry at inclusion in 386 patients included in the DIGItal Cohort Design (DIGICOD) cohort. We investigated whether serum CML and/or HCit concentrations were associated with erosive HOA or with HOA clinical and radiological features. Moreover, we compared the tissular concentrations of CML and HCit in OA and non-OA cartilage from proximal interphalangeal and metacarpo-phalangeal (MCP) joints from human cadaveric donors. RESULTS: Median (IQR) serum CML concentration was lower in patients with erosive HOA than those with non-erosive HOA (178.7 [157.1-208.8] vs 194.7 [168.9-217.1] µmol/mol Lys, P = 0.002), but median HCit concentration did not differ between the groups (193.9 [162.9-232.0] vs 193.9 [155.9-224.6] µmol/mol Lys). Cartilage HCit and CML concentrations were not correlated with clinical features. Serum CML concentration was higher in OA than non-OA MCPs (7.0 vs 4.0 mmol/mol Lys, P = 0.01). CONCLUSIONS: Serum CML concentration was lower in erosive HOA than non-erosive HOA, and cartilage CML concentration was higher in OA than non-OA cartilage. These results encourage further studies to test whether serum CML could be a new prognostic biomarker in HOA.

Diabetes is a risk factor for knee osteoarthritis progression.

PURPOSE: Recent studies have suggested that metabolic factors (obesity, diabetes, hypertension and dyslipidemia) and their clustering in metabolic syndrome (MetS) might be involved in the pathophysiology of knee osteoarthritis (OA). We investigated their impact on radiographic progression by an annualised measure of the joint space narrowing (JSN) of the medial tibiofemoral compartment. METHODS: 559 patients older than 50 years with symptomatic knee OA were recruited for the placebo arm of the SEKOIA trial. The presence of diabetes, hypertension and dyslipidemia was determined at baseline interview. Body mass index (BMI) was calculated, obesity was considered >30 kg/m(2). MetS was defined by the sum of metabolic factors ≥ 3. Minimal medial tibiofemoral joint space on plain radiographs was measured by an automated method at baseline and then annually for up to 3 years. RESULTS: The mean age of patients was 62.8 [62.2-63.4] years; 392 were women. A total of 43.8% was obese, 6.6% had type 2 diabetes, 45.1% hypertension, 27.6% dyslipidemia and 13.6% MetS. Mean annualised JSN was greater for patients with type 2 diabetes than without diabetes (0.26 [-0.35 to -0.17] vs 0.14 [-0.16 to -0.12] mm; P = 0.001). This association remained significant after adjustment for sex, age, BMI, hypertension and dyslipidemia (P = 0.018). In subgroup analysis, type 2 diabetes was a significant predictor of JSN in males but not females. The other metabolic factors and MetS were not associated with annualised JSN. CONCLUSION: Type 2 diabetes was a predictor of joint space reduction in men with established knee OA. No relationships were found between MetS or other metabolic factors and radiographic progression.