This study records the length of hospital stay of 50 total knee arthroplasty patients involved in an accelerated postoperative rehabilitation protocol, and a control group of patients undergoing routine rehabilitation. This protocol involved modifications to normal knee replacement procedure, including infiltration of bupivacaine and adrenaline to the divided tissue layers at the time of surgery, spinal anaesthesia, and mobilisation on the day of surgery. These modifications were combined with an organised multidisciplinary approach anticipating issues that may delay discharge. In addition, patients and hospital staff were encouraged to expect an earlier discharge from the hospital. The mean length of hospital stay after surgery was reduced to 3.6 (S.D. 1.0) days, from a previous departmental average of 10.5 days. The control group inpatient stay was 6.6 (S.D. 2.6) days. Plasma bupivacaine levels were found to be well within safe levels, and pain records indicated that the protocol did not cause increased levels of discomfort. American Knee Society and Oxford knee scores demonstrated good levels of knee function at 6 weeks post surgery. In addition, it was noted that no postoperative blood transfusions were required. This is regarded as a significant further benefit.