<b>Background and objectives.</b> Knee osteoarthritis (KOA) is a leading cause of disability in the developed world. Additionally, it is possibly linked with premature mortality. Low-grade inflammation is associated with a high risk of non-traumatic KOA incidence but also with metabolic syndrome. Knee injury is a major risk factor for KOA and is associated with an inflammatory response. Heterogeneity of both the symptoms and progression makes early identification of individuals at risk difficult. The aim of this thesis was to examine the diagnostic value of selected biomarkers to identify microscopic synovitis, examine their predictive value for KOA development and any association with excess premature mortality. <b>Methods.</b> Four serum biomarkers, COMP, resistin, C3M and CRPM, all linked with both knee pain and synovial or systemic inflammation, were selected and an additional sonographic marker was tested. Data from two cohorts has been utilised for the purpose of this thesis: the Oxford Knee Injury Cohort, a prospective study of NHS patients with recent ACL and/or meniscal injuries, and the Chingford women's study, a community-based prospective cohort with over 23 years of follow-up. <b>Results.</b> Cross-sectionally, sonographic markers correspond well with microscopically defined post-traumatic synovitis. Two biomarkers, COMP and resistin, were associated with development of incident radiographic KOA (RKOA), while two others, C3M and CRPM, predicted development of painful RKOA independently to age and BMI. High C3M levels and presence of painful RKOA were associated with premature mortality. Knee pain alone, especially in presence of RKOA was an independent predictor of mortality, but RKOA without pain was not. <b>Conclusion.</b> These results support the use of the studied biomarkers in complex predictive models for development of KOA and associated premature mortality. Taking the competing risk of death into account is an important consideration in this field.