Tranexamic Acid for Bleeding in Breast Surgery

Background:

Aberrant bleeding is an important consideration in breast surgery, which can involve large resections of soft tissues in the breasts. Hematomas and/or seromas are common complications with these procedures, and can lead to further complications, such as flap necrosis. Seromas, perhaps the commonest complication, are documented to occur at rates as high as 85% in certain breast procedures. Hematoma rates are typically lower than that of seroma, ranging from 1% to 2% in mastectomy procedures and 5% to 8% in reduction mammaplasty. These bleeding complications pose an obvious risk to patient safety, and their management can result in significant costs to patients and healthcare alike.

Tranexamic acid (TXA), also known by the brand name Cyklokapron® from Pfizer™, is a lysine analogue used to prevent and/or treat bleeding by competitively blocking the activation of plasminogen to plasmin, thereby preventing plasmin from binding to blood clots. Tranexamic acid is used safely in several medical and surgical settings, such as cone biopsies, epistaxis and hyphaema control, and dental surgery. This antifibrinolytic medication is commonly used in many surgical procedures, namely cardiac and orthopedic surgery, for the purposes of preventing blood loss. TXA has been demonstrated to effectively reduce perioperative blood loss and transfusion rates in many different surgical procedures, such as cardiac, urology, gynecology, and orthopedic surgeries, without any significant safety concerns.

Despite these findings, there is a clear paucity of studies on TXA in breast surgery. There are currently no studies exploring perioperative TXA use in breast reduction, and only one study in breast surgery overall. This study by Oertli et al. looked at the use of TXA in mastectomy or lumpectomy with axial dissection, in a randomized placebo-controlled trial (RCT). Seroma rate was shown to be reduced by 10% with a reduction in TXA versus placebo. However, due to a low power of the study, Oertli et al. were unable to determine statistical significance.

Purpose:

The purpose of this study is to determine if perioperative administration of tranexamic acid (TXA) is superior to placebo in minimizing bleeding complications (as determined by: hematoma/seroma incidence in the breast, re-operation and/or additional procedures, blood transfusion volume, drainage volume (Jackson-Pratt drain or percutaneous drainage), number of days spent in hospital) in breast surgery (Reduction mammaplasty, mastectomy with and without immediate tissue expander and implant-based reconstruction, oncoplastic breast surgery). Conterminous with this objective, the investigators also intend to study the safety of tranexamic acid use perioperatively in these procedures by recording any adverse effects that occur at the time of surgery and throughout the follow-up period. The study will randomize and prospectively observe two patient populations undergoing any one of these procedures: one group receiving preoperative and subsequent post-operative 1g doses of tranexamic acid, and the other receiving placebo (normal saline). These two groups will be observed based on specific properties and complications from surgery. Finally, to determine if perioperative TXA use is more cost-effective than placebo in the setting of breast surgery. Information on health care utilization will be collected prospectively in both groups to provide a cost analysis of tranexamic acid.

The following hypotheses will be tested:

1. Perioperative TXA use is superior to placebo in reducing bleeding complications (ie. hematoma, seroma, and excessive blood loss) in the surgical or donor site.
2. Perioperative TXA is safe for use in this patient population.

Study Design

The present study will be a randomized, double-blind, placebo-controlled trial. This RCT will follow the CONSORT statement. Conterminous to this RCT, data on treatment costs (hospital and patient cost) will be recorded. The study will last approximately two years. Case Report Forms will be filled out at each follow-up period. Follow-up will occur at 2 weeks, 6 weeks, and 12 weeks postoperatively. Cost-analysis of the intervention will also be performed.

Sample Size:

Based on published data, the incidence of the combined hematoma/seroma endpoint is expected to range between 6 and 10% in the reductions patients and 16-87% in the mastectomy patients. Since the investigators expect to have equal numbers of each type of patient, the investigators anticipate the overall event rate in the placebo arm to range between 11-47%. Taking the midpoint, the investigators anticipate a placebo arm event rate of 29%. The investigators would consider a 30% relative reduction (i.e., an absolute difference of 9%) to be the minimal clinically important difference that would, if demonstrated, change practice. To detect this difference with 80% power, the investigators require 385 patients per arm using a two-sided test at the 5% level of significance To account for attrition of 15 patients per arm (<5%), the investigators will aim to 400 patients per group.

Data Analysis:

The unit of analysis will be each individual patient for all outcomes. Baseline characteristics will be compared between study arms using descriptive statistics (frequencies and proportions for categorical variables, mean and standard deviation for continuous variables with a normal distribution and median and inter-quartile range for variables with a skewed distribution). The investigators' primary outcome, the incidence of hematoma and/or seroma, will be analyzed using a chi-square test with continuity correction. The difference between the arms will be described using a relative risk with 95% confidence interval. The overall rate in each group will also be reported. Secondary outcomes will be analyzed using chi-square tests (dichotomous outcomes) or Wilcoxon tests (drainage volume, transfusion volume, and days in hospital). Drainage and transfusion volume, as well as the number of days in hospital, will be summarized as median and interquartile range. All data analysis will be carried out by a statistician from the Ottawa Hospital Research Institute.