Blood loss during a surgical procedure is inevitable, its reduction is a key factor for surgical success. Also, to avoid progress to severe complications like hemodynamic decompensation, cardiac arrest or the need to blood transfusions. According to the Center for Disease Control (CDC), there is a usage of more than 17 million transfused blood products units per year. Blood transfusion is a convenient technique for reposition of blood during major bleed, but it involves several probable complications like anaphylaxis, bloodborne infections and others. Consequently, meticulous hemostasis during surgery is crucial to diminish blood loss.

Hemostatic agents play a pivotal role during surgical time. Amongst them topical, energy-based and systemic agents are the vast majority. Tranexamic acid (TXA) is a lysine synthetic derivate that inhibits fibrinolysis by blocking the 5 lysine-binding sites to plasminogen. It has been used in clinical practice since 1962 and become very popular after 2010 when the CRASH-2 study showed a decreased risk of death in trauma patients.

Tranexamic acid use is widely extended among diverse surgical fields: orthopedics, cardiac surgery and obstetrics. In plastic surgery it uses is limited mainly to craniofacial surgery as a local agent. The aim of this trial is to show the efficacy of tranexamic acid as hemostatic agent in liposuction and to compare its efficacy among different administration routes.

In 3 plastic surgery centers, the investigators randomized patients in a double-blind, randomized, controlled trial comparing the clinical outcomes of different administration routes of TXA, intravenous versus subcutaneous versus placebo between January 2019 to February 2020. The study was approved by the Ethics committee at the central coordinating center and at each of the participating sites. The centers of the study were, Dhara clinic, Bogota, Colombia, Centro Medico de las Americas clinic, in Mérida, Mexico and Innovare hospital in Jalisco, Mexico. Once a month, the investigators performed only meetings between the researcher coordinator and delegates from each of the centers to audit the data collection and information registration.

During the preoperative consultation, the study interventions were explained clearly to each participants. Participants were given adequate time to ponder the information, had any questions answered and gave their free and voluntary consent. A total of 141 participants were recollected. The primary outcome was the difference regarding to hemoglobin loss points between groups. All patients had hemoglobin and hematocrit tests taken prior to surgery, 24 hours postoperatively and 5 days postoperatively. The investigators determined the hemoglobin loss points in each group by deduct the postoperative hemoglobin at day one to the preoperative hemoglobin (For hemoglobin loss at day 1) and the postoperative hemoglobin at day five to the preoperative hemoglobin (For hemoglobin loss at day 5). As a secondary outcome, the investigators performed a subgroup analysis to determine the difference in hematological parameters (hemoglobin and hematocrit) between the groups.

Sample size and Statistical Analysis Sample size was calculated using the online app from sealed envelope where 96 patients were required to have a 90% of chance of detecting, as significant at the 5% level, a difference between the mean postoperative Hb value from 10.5 g/dl in the control group to 11.5 g/dl in the experimental group and an estimated standard deviation (SD) of 1.5 g/dl. Since, the investigators planned two experimental approaches: Subcutaneous (SC) TA and Intravenous (IV) TA; 141 patients were randomized into 3 groups: 1. IV-TA group (n=47), 2. SC-TA group (n=47) and NS group. Demographic and clinical characteristics were described, hematological parameters were analyzed per group before surgery, at the immediate post-operative period, and at the fifth post-operative day. The descriptive analysis was made through means and medians according to the distribution of variables.

For categorical variables, counts and percentages were used. The quantitative variables with parametric distribution were compared using ANOVA, and for variables with non-parametric distribution using Kruskal-Wallis test, while the frequencies of the qualitative variables were compared using Chi2 test. We used a boxplot to illustrate the differences between the means of the hemoglobin by group. Bilateral hypotheses were formulated at two tails with significance level less than 0.05. The data were collected in Microsoft excel and the statistical analysis were performed in the statistical program STATA version 15.0.

Inclusion Criteria:

• Patients that underwent liposculpture as unique procedure, surgical time between 2 and 5 hours, age between 20 and 45 years old and BMI between 20 and 30 m2/kg.

Exclusion Criteria:

• Patients that were operated of liposculpture and any additional procedure (Mammoplasty, tummy tuck, etc.), patients with history of thromboembolic disease or in Treatment with aspirin within 14 days before surgery or anticoagulants within 5 days prior to surgery, patients with history of epilepsy, patients with allergy to tranexamic acid, comorbidity grade III or IV according to ASA (American society of Anesthesiologists) and patients with prothrombin time (PT) and/or activated partial thromboplastin time (aPTT) >1.5 times the baseline.