Topical intra-articular compared with intravenous tranexamic acid to reduce blood loss in primary total knee replacement: a double-blind, randomized, controlled, noninferiority clinical trial.

Abstract

Abundant literature regarding the use of intravenous tranexamic acid (TXA) in primary total knee replacement is available. Randomized controlled trials have confirmed the efficacy of topical TXA compared with placebo, but the comparison between topical and intravenous TXA is unclear. The present study was designed to verify noninferior efficacy and safety of topical intra-articular TXA compared with intravenous TXA in primary total knee replacement with cemented implants. A Phase-III, single-center, double-blind, randomized, controlled clinical trial was performed to compare topical intra-articular TXA (3 g of TXA in 100 mL of physiological saline solution) with two intravenous doses of TXA (15 mg/kg in 100 mL of physiological saline solution, one dose before tourniquet release and another three hours after surgery) in a multimodal protocol for blood loss prevention. The primary outcome was the blood transfusion rate, and the secondary outcomes included visible blood loss (as measured in the drain) at twenty-four hours postoperatively and invisible blood loss (as estimated from the Nadler formula) at forty-eight hours postoperatively. The sample size of seventy-eight patients was calculated to give a statistical power of 99% for demonstrating noninferiority. Thirty-nine patients each were allocated to receive topical intra-articular TXA (the experimental group) and intravenous TXA (the control group); there were no significant differences in demographics or preoperative laboratory values between the groups. Noninferiority was estimated by comparing the confidence intervals with a delta of 10%. Student t and Mann-Whitney tests were used to assess the significance of any differences. The transfusion rate was zero in both groups; thus, noninferiority was demonstrated for the primary efficacy end point, suggesting equivalence. Noninferiority was also demonstrated for the secondary efficacy end points. Drain blood loss at twenty-four hours was 315.6 mL (95% confidence interval [CI], 248.5 to 382.7 mL) in the experimental group and 308.1 mL (95% CI, 247.6 to 368.5 mL) in the control group (p = 0.948, Mann-Whitney). Also, estimated blood loss at forty-eight hours was 1259.0 mL (95% CI, 1115.6 to 1402.3 mL) in the experimental group and 1317.9 mL (95% CI, 1175.4 to 1460.4 mL) in the control group (p = 0.837, Mann-Whitney). No significant safety differences were seen between groups. Topical administration of TXA according to the described protocol demonstrated noninferiority compared with intravenous TXA, with no safety concerns. This randomized controlled trial supports the topical intra-articular administration of TXA in primary total knee replacement with cemented implants. Therapeutic Level I. See Instructions for Authors for a complete description of levels of evidence.