OBJECTIVES:Knee injury is strongly associated with the development of knee osteoarthritis. While there is preliminary evidence for an increased risk of knee replacement (KR) surgery after sports injury, no studies have investigated this at a population level. This population-level study aimed to quantify the likelihood of KR surgery and direct healthcare costs 10-15 years after sports injury. DESIGN:Statewide population-based cohort study. METHODS:The cohort was established by linking two key administrative datasets capturing all hospital admissions and emergency department (ED) presentations in Victoria, Australia. Sports injury presentations from 2000-2005 and KR admissions from 2000-2015 were identified using ICD-10-AM codes. A Cox proportional hazards model estimated likelihood of KR using time to surgery admission data, adjusting for potential confounders. KR costs for the sports-injured cohort were estimated from the health system perspective using diagnosis codes and national hospital cost weights. RESULTS:Over the study period there were 64,038 sports injuries (including 7205 knee injuries) resulting in ED presentation or hospitalisation, and 326 KR procedures. Multivariate analysis showed that having a knee injury more than doubled the hazard of subsequent KR (hazard ratio 2.41, 95%CI 1.73-3.37), compared to all other sports injuries. Direct healthcare costs for KR totaled $AUD7.93 million for the cohort, with 21% of costs attributable to the knee injury group. CONCLUSIONS:Sports-related knee injury manifests in a significantly greater likelihood of KR, at considerable cost to society. Targeted health policy and effective interventions are needed to prevent sports-related knee injuries and contain this substantial burden.