AimTo evaluate whether knee alignment explains the higher prevalence of lateral compartment tibiofemoral radiographic osteoarthritis (TFROA) among rural Chinese compared with that among Whites.MethodsThe Wuchuan OA Study is a population-based longitudinal study of risk factors for knee OA. At baseline 1030 participants had home interviews, clinical examinations and weight-bearing posteroanterior semi-flexed radiographs of the tibiofemoral joints. Anatomic knee alignment was measured using an e-film workstation and divided into three categories: normal (182°-184°), valgus (> 184°), and varus (< 182°) alignment. A knee was defined as having medial or lateral compartmental ROA if its Kellgren and Lawrence grade was ≥ 2 and joint space narrowing ≥ 1 in the medial or lateral compartment, respectively. We examined the association between knee alignment with prevalent medial or lateral knee ROA separately using multiple logistic regression.ResultsAmong 1030 participants, the proportions of knees with normal, valgus and varus alignment were 29.9%, 56.5% and 13.7%, respectively. The prevalence of medial and lateral ROA was 16.0% and 4.3%, respectively. Valgus alignment was associated with prevalence of lateral compartment ROA (odds ratio [OR] = 5.0, 95% CI: 2.4-10.5), while varus alignment was associated with medial compartment ROA (OR = 6.1, 95% CI: 4.4-8.6). The ratio of prevalence of lateral versus medial compartment TFROA was greater in Wuchuan than that in the Framingham OA Study and valgus malalignment was more common in Wuchuan than in the Rotterdam study.ConclusionsThe prevalence of compartment-specific TFROA differs between rural Chinese and Whites. This difference is likely due to relatively high prevalence of valgus malalignment in rural Chinese compared with that in Whites.