Tourniquet release for hemostasis increases bleeding. A randomized study of 77 knee replacements.

We conducted a prospective, randomized study of 77 primary knee replacement operations on 75 patients (52 women), with a mean age of 71 years, to evaluate the effect of tourniquet release for hemostasis on blood loss and transfusion requirements. The operations were all done with spinal anesthesia and the use of a midline skin incision and medial parapatellar approach. In group I, the tourniquet was released for hemostasis before the wound was closed. In group 2, the tourniquet was first released after the wound was closed and a compressive dressing had been applied. The total intra- and postoperative blood losses were, on average, 858 mL (SD 443) in group I and 589 mL (347) in group 2 (p = 0.01). The median units of blood given and the postoperative decreases in hemoglobin values were similar in both groups. In a subgroup of 45 cementless prostheses, the 25 patients with prostheses allocated to group 1 lost 1022 mL (397) blood, compared to 646 mL (333) by the 20 patients with prostheses in group 2 (p = 0.01). Our findings speak against the efficacy of tourniquet release for hemostasis in knee replacement surgery.