The Effect of Topical Tranexamic Acid on Postoperative Bleeding From Superficial Wounds

Overview

After surgical procedures, interventions to reduce postoperative bleeding are of great importance. In this study, the effect will be investigated of administering tranexamic acid, which is designed for injection, directly onto the raw wound surface (topical application) created during surgery. In this way only a small amount of drug is to reach a large wound area. There will be a higher drug concentration of it in the exposed wound surface than after injection, but only a very low concentration in the body, and no risk of injury from needles. The researchers have recently shown that topically applicated tranexamic acid reduces bleeding in women who had two-sided breast reduction surgery. Now it will be studied whether topically applicated tranexamic acid reduces bleeding from superficial wounds, using as a study model the homogenous wounds created by tangential skin excision when harvesting split skin grafts for skin transplants. Two identical wound surfaces in the same patient will serve as case and control.

One or more paired donor sites will be created for each patient using the Zimmer dermatome. Paired wounds withcomparable size, depth, and location and will be labeled "A" and "B." A study nurse not involved in the operation will prepare two identical vials of 20 ml saline, marking them "A" and "B." A sealed numbered study envelope will state which vial should receive TXA. In the study drug vial, 5 ml will be replaced with 5 ml of 100 mg/ml TXA, yielding a solution of 25 mg/ml TXA. Both vials will receive 0,1 ml epinephrine 1 mg/ml, yielding an epinephrine concentration of 5 µg/ml to match the hemostatic saline-epinephrine solution applied topically to donor wounds at the burn center. The sets of vials and the corresponding study ID numbers will be delivered to the operating theatre. Prior to bandaging, the respective study wounds will be moistened with the corresponding study fluid. Gloves will be changed between each application to avoid cross-contamination. The study wounds will then be covered with an innermost non-absorbent layer of Vaseline gauze and five dry surgical gauzes to absorb the blood and exudate. The weight of a single dry gauze is consistent (27.5 g). The dressings will be marked "A" and "B" according to the drug vials used. On the first postoperative day, the dressings will be removed except for the innermost Vaseline gauze. The wound surface area and the area of the blood stain on the innermost gauze will be measured. Dressing weight gain will be calculated by weighing the five dry gauzes and subtracting the dry weight (137.5 g). The paired gauzes will be visually compared for bleeding.and photo documented for later direct comparison. During the first 24 h and over the following days and weeks, the participants will be monitored for possible adverse events, postoperative complications, and time to re-epithelialization. Randomization Computer-generated randomization, production of corresponding sealed study envelopes, and organization of electronic case report forms will be provided by the Clinical Research Unit of St. Olav's University Hospital, Trondheim, Norway. Randomization instructions will be executed by a study nurse who was otherwise not connected to surgical procedures or patient follow-ups. All participants and personnel involved in surgery, follow-up, data collection, and statistical analysis will be blinded to the randomization. Study end points The primary endpoint will be postoperative bleeding, defined as the net weight gain of the dressings per wound area. The secondary endpoints will be blood stain to wound area ratio and visual comparison of the amount of blood between paired dressings. All variables will be recorded on the first postoperative day. Additional secondary outcomes will be time to re-epithelialization, defined as no oozing in the dressings, and the occurrence of complications, such as wound infections and thromboembolic events. Statistical analysis A ≥ 25 % reduction in bleeding in TXA donor wounds will be considered clinically significant. A delay in healing time or an increase in infection rate of ≥ 25 % will be considered clinically significant. The standard deviation (SD) was uncertain, as few similar studies exist but were estimated to be 0.4, based on previous effect studies22,23. As each patient will be his or her own control, using a paired samples t-test to detect a difference of 0.25, and a standard deviation of 0.4, α of 0.05, and power of 0.80, a sample size of 23 wound pairs is needed (power calculation. http://www.biomath.info/power/prt.htm25). We choose to include a total of 36 wound pairs for additional power in case of technical difficulties, since previous effect studies do not use the surrogate variables for bleeding used in this study. Continuous data will be analyzed using the paired samples t-test for normally distributed data and Wilcoxon signed rank test for non-normally distributed data. Categorical data will be analyzed using the chi-squared test.