Background Different anthropometric variables such as obesity with its different ways of measure (body fat index, waist circumference, waist/hip ratio) as well as the metabolic syndrome and his individual components have been related to knee osteoarthritis (OA). However, no specific studies analyzing the influence of these factors on clinical severity have been carried out in patients with synovial effusion.
Objectives To assess the relationship between different anthropometric measures, the metabolic syndrome and its individual components with clinical severity as measured by the Lequesne algofunctional questionnaire in women with knee OA and significant synovial effusion.
Methods Transversal study including consecutively women aged 50 to 80 years with symptomatic knee OA, Kellgren-Lawrence I-IV with significant synovial effusion (>4mm at mid-patellar line on US examination). Demographic data and history of hypertension (HT), dyslipidemia, obesity or diabetes mellitus (DM) were collected. Blood samples were taken to determine basal glycemia, total, HDL- and LDL-cholesterol, insulinemia and glycated hemoglobin. Anthropometric variables assessed included weight, height, body mass index (BMI), waist circumference, hip circumference and percentage of body fat as measured by bioelectrical impedance analysis). Patients were evaluated for the presence of metabolic syndrome by strict NCEP-ATP-III criteria. All patients underwent joint aspiration to confirm non-inflammatory synovial fluid without microcrystals. To evaluate knee OA clinical severity patients completed the Lequesne algofunctional index. Univariant and multivariant analyses were performed to evaluate which variables had a greater influence on the Lequesne index.
Results One hundred and fifteen women were analysed, average age 68.8 ±11.1 years, average knee OA symptoms duration 50 ±23 months. Percentage of HT was 54.58%, DM 10.4%, obesity 57.4% and dyslipidemia 47.8%, all above the expected by age and gender (40%, 10%, 40% and 30%, respectively). Metabolic syndrome was present in 40.9% of patients, also above 30% expected. In the univariant analysis all the anthropometric variables showed a significant association with the Lequesne index with weak to moderate correlation (body fat index p=0.008/r =0.243, BMI p=0.002/r =0.277, weight p=0.007/r =0.249, waist circumference p<0.001/r =0.404, waist circumference, p=0.006/r =0.252 and hip circumference and waist/hip ratio p=0.01/r =0.221). Neither the metabolic syndrome nor any of its individual components achieved statistical significance. After performing a forward stepwise multivariate analysis including OA symptom duration, age and Kellgren-Lawrence, waist circumference emerged as the only significant variable (p<0.001/r =0.338), observing a magnitude effect of a 0.9 increase in the Lequesne index for every increase of waist circumference of 5 cm (beta coefficient 0.180, 0.083–0.276). The relationship between BMI and Lequesne index was confirmed but when the waist circumference was included in the model only the waist circumference remained significant.
Conclusions In this study waist circumference was the anthropometric variable more related to clinical severity measured by the Lequesne index in women with knee OA and synovial effusion.
Disclosure of Interest None declared