Independent Predictors of Knee Osteoarthritis in an Elderly Japanese Population : A Multivariate Analysis

Knee osteoarthritis (OA) is common in elderly populations, and independent predictors of knee OA in Japanese have not been thoroughly investigated. The aim of this study was to identify independent factors associated with radiographic knee OA in an elderly Japanese population by using multiple logistic regression analysis in order to use them to predict knee OA. Subjects of this cross-sectional study were 1,256 participants (548 men and 708 women) in the fourth Matsudai Knee Osteoarthritis Survey conducted in 2000. Knee OA was diagnosed based on X-ray findings, according to the Kellgren and Lawrence scale (grade 0-IV). Demographic and physical characteristics, past disease history, the femorotibial angle (FTA, an index of varus alignment of the knee joint), thrust, history of knee effusion, round back, and history of knee injury were obtained. The average age of the subjects was 70.4 years (SD 6.2). The prevalences of grade≥II knee OA were 223/548 (40.7%) in men and 465/708 (65.7%) in women. Stepwise multiple logistic regression analysis showed female sex odds ratio ((OR)=2.46), age (OR=1.06), body mass index ((BMI), Q1-4 v.s. Q5) (OR=2.09), history of knee injury (OR=1.76), FTA (OR=1.15), thrust (OR=1.81) history of knee effusion (OR=2.08), and round back (OR=1.47) to be significant factors associated with grade≥II knee OA. In conclusion, FTA, thrust, knee effusion, and round back are possible predictors of knee OA in elderly Japanese, independent of already known predictors, such as sex, age, BMI, and history of knee injury. Causal relationships should be further investigated.