One hundred fifty patients undergoing elective unilateral TKA under spinal anesthesia were randomized to a self-warming blanket (SW; n=75) or an upper-body forced-air warming blanket (FAW; n=75). Patients were pre-warmed for 30 min; intra-operatively the assigned blanket was used (FAW at 38 °C), with rescue FAW at 43 °C if core temperature fell <36.0 °C (NICE threshold). The primary outcome was core temperature on RR admission; secondary outcomes included peri-operative skin/core temperature trajectories, hypothermia incidence and duration, usability, costs, and complications, with continuous monitoring up to 1 h post-op. Overall, the results showed no between-group difference in core temperature on RR admission. However, intra-operative hypothermia was more frequent with SW (61% vs 49%), prompting more rescue warming. In conclusion, SW met statistical non-inferiority for the primary endpoint, but FAW yielded fewer episodes of intra-operative hypothermia under this protocol