Materials and Methods: Sixty-four osteoarthritic knee patients undergoing TKA were randomly assigned to one of two groups. In one group, the tourniquet was released before a polyethylene insert was put in place (Before Group) and in the other, release occurred after wound closure (After Group). Tourniquet pressure was set at patient systolic blood pressure plus 150 mmHg. A suction drain was placed and retained for 48 hours postoperatively. Measured blood loss, blood drainage, 24-hour postoperative fall in hematocrit, change of hematocrit, and rate of blood transfusion were measured. Wound complications, deep vein thrombosis, and knee flexion were also recorded. All patients were followed for at least three months.

Results: Mean total blood loss was similar between the two groups (377±155 in the Before Group and 450±172 mL in the After Group, p = 0.09). Patients in the Before Group had less blood loss in the drain but they had an average additional intraoperative blood loss of approximately 72±53 mL. Change in hematocrit was not different between the groups (7.8% in the Before Group and 8.4% in the After Group, p = 0.45). Fewer patients in the Before Group required blood transfusion (30% versus 48%, p = 0.001). There were no wound complications or deep-venous thrombosis in either group. Postoperative knee flexion was also similar in both groups.

Conclusion: Releasing the tourniquet before wound closure does not change postoperative total blood loss compared with after wound closure, but it does reduce postoperative blood loss collected in the drain.

Keywords: Blood loss, Suction drain, Tourniquet, Total knee arthroplasty