The purpose of this study is to examine the effect of a postoperative blood salvage drain and timing of tourniquet release on the maximal hematocrit drop after total knee arthroplasty. Thirty-seven total knees were prospectively randomized into either an early or late tourniquet release group. Hematocrit drop and drainage amounts were recorded. We found no significant difference in maximal hematocrit drop, drainage amounts, or total surgical time between the groups. We conclude that the use of a blood salvage drain should not influence the surgeon's preference on timing of tourniquet release in total knee arthroplasty.