OBJECTIVES:   To estimate and compare the lifetime costs and health consequences associated with unicompartmental knee replacement (UKR) and total knee replacement (TKR). METHODS:   Various sources of routinely-collected data from the English National Health System were used to inform the analysis. Propensity score matching was used to identify comparable patients who received a UKR or TKR. Health outcomes, measured in Quality Adjusted Life Years (QALYs), as well as hospital and primary care costs to the health system were estimated and compared for subgroups based on age (<60, 60-75 and 75+) and gender, and for two subgroups based on surgeon usage of UKR (whether <10% or ≥10% of knee replacements performed by a surgeon were UKRs). These estimates and their distributions were used to populate a lifetime Markov model. RESULTS:   For all age and gender subgroups the provision of UKR is expected to lead to a gain in QALYs compared to TKR (Male <60: 0.11, 60-75: 0.27, 75+: 0.19, Female <60: 0.06, 60-75: 0.26, 75+: 0.39) and a reduction in costs (Male <60: -£1,070, 60-75: -£1,676, 75+: -£1,903, Female <60: -£804, 60-75: -£1,249, 75+: -£1,215 per patient over the lifetime). UKR is expected to lead to a reduction in QALYs compared to TKR when performed by surgeons with low UKR utilisation but an increase among those with high utilisation (<10%: -0.19, ≥10%: 0.17). Regardless of surgeon usage, costs associated with UKR are expected to be lower than those of TKR (<10%: -£886, ≥10%: -£1,258). CONCLUSIONS:   Based on its current provision in the UK, UKR can be expected to generate better health outcomes and lower lifetime costs than TKR. However, health outcomes are worse for UKRs when performed by surgeons who use UKR for <10% of their knee replacements and these surgeons should consider broadening their indications.