Methods

Six centers prospectively randomized 155 revisions to one of four regimens: 1g of intravenous (IV) TXA prior to incision, a double dose regimen of 1g IV TXA prior to incision and 1g IV TXA during wound closure, a combination of 1g IV TXA prior to incision and 1g intraoperative topical TXA, or three doses of 1950mg oral TXA administered 2 hours preoperatively, 6 hours postoperatively, and on the morning of postoperative day one. Randomization was based upon revision subgroups to ensure equivalent group distribution, including: femur only, acetabulum only, both component, explant/spacer, and second stage reimplantation. Patients undergoing an isolated modular exchange were excluded. An a priori power analysis (alpha = 0.05; beta = 0.80) determined 40 patients per group were required to identify a 1g/dL difference in postoperative hemoglobin reduction between groups. Per-protocol analysis involved an analysis of variance, Fisher's exact tests, and two one-sided t-tests for equivalence.