Although knee injury has been implicated as a risk factor for the development of knee osteoarthritis (OA), there is great disparity in the magnitude of quantifiable risk. Our aim was to systematically review the relationship between history of knee injuries and knee OA.

Six electronic databases were searched between August and October 2010. Relative risk estimates or odds ratio (OR) and 95% confidence intervals (95% CI) were extracted or calculated from observational studies meeting the inclusion criteria. Publication bias was determined using funnel plot and the Egger’s test. Heterogeneity was examined using Cochran Q test and I² statistic. Random effects model was used to pool the heterogeneous results and OR was used to present the results. Subgroup analyses were performed to examine potential causes of heterogeneity.

Twenty-four observational studies (20,997 subjects) were included in the meta-analysis of which there were seven cohort, five cross-sectional and 12 case-control studies. The overall pooled OR was 4.20 (95% CI 3.11–5.66, I² = 81.0%). Association between history of knee injuries and knee OA was significantly different for specified injuries such as ligament or tendon injuries; meniscus damage or meniscectomy; and fracture of femur, knee or lower part of the leg (OR = 5.95, 95% 4.57–7.75), compared to unspecified injuries (OR = 3.12, 95% 2.17–4.50).

History of knee injury is a major risk factor for the development of knee OA irrespective of study design and definition of knee injury. As one of the few modifiable/preventable risk factors, knee injury should be part of the future prevention programme in reducing the risk of knee OA.