OBJECTIVE: To evaluate the cost and effectiveness associated with total knee replacement (TKR) and unicompartmental knee arthroplasty (UKA) in patients with knee osteoarthritis of medial compartment. METHODS: This was a two-year non-randomized prospective observational cohort study in unicompartmental knee osteoarthritis patients scheduled for TKR (n=431) or UKA (n=102). Costs were identified using administrative database and health outcome was measured using the SF-36 and the Oxford knee score (OKS) one week before, six months after, and two years after surgery. Incremental cost-effectiveness ratio (ICER) for TKR versus UKA was calculated and its 95% confidence interval was estimated using bootstrapping method. Cost-effectiveness acceptability curves were constructed from different perspectives. RESULTS: UKA achieved similar health benefit at significantly lower costs compared to TKR. The mean ICER for TKR versus UKA was CAN$153,000 per QALY from the society perspective, CAN$143,000 per QALY from the patient’s perspective, and CAN$11,000 per QALY from the government’s perspective. The 95% confidence interval of ICER for TKR versus UKA cannot be defined because more than 5% bootstrapping ICERs fell into the upper left quadrant of cost-effectiveness plane from all three perspectives. CONCLUSIONS: UKA is a cost-saving alternative treatment to TKR for OA patients in short-term. A long-term prospective study is necessary to determine the cost-effectiveness of TKR and UKA.