RESUMO
Osteoarthritis of the knee is strongly associated with central obesity, and the prevalence of obesity in patients undergoing knee arthroplasty procedures is expected to increase. Central obesity is associated with multiple adverse outcomes after knee arthroplasty, but the impact of obesity specifically around the operated knee is poorly understood. One reason for this is the absence of a standardized measure, analogous to BMI for central obesity, to facilitate research. The aim of this study was to develop a radiological limb obesity scoring system which reliably corresponds with anthropometric measurements. Detailed anthropometric and radiological measurements were taken from patients undergoing total and unicompartmental knee replacement. Anthropometric measurements were combined into a composite score using principal components analysis. Different radiological scoring methods were evaluated against the anthropometric measure using correlation and regression analyses. Intraclass correlation coefficients were used to evaluate intra- and inter-observer reliability of the radiographic measurements. Measurements were obtained from fifty patients. There was a significant correlation between the composite anthropometric score and unadjusted radiological soft tissue width at the level of the femoral condyles (p < 0.001). There was also a significant correlation with the ratio of tissue to bone widths at the same level (p < 0.001), and fit was improved by including variables coding for soft tissues overlapping the medial or lateral boundaries of the radiograph. We have demonstrated a methodology for producing standardized radiological limb obesity scores which are strongly correlated with anthropometric measurements, and which can be easily applied in diverse clinical and research settings.
