Tranexamic Acid Mechanisms and Pharmacokinetics in Traumatic Injury

Phase 2

Completed

Conditions: Hemorrhage
Shock
Wounds and Injuries

Interventions: Other: Placebo
Drug: Tranexamic Acid

Registration Number: NCT02535949

Lead Sponsor: Washington University School of Medicine

Brief Summary: The purpose of this study is to evaluate the effects of TXA on the immune system, its pharmacokinetics, as well as safety and efficacy in severely injured trauma patients.

Detailed Description: Trauma is the leading cause of death in persons younger than 40 years. Hemorrhage is the etiology in 30% of these deaths, and remains the leading cause of potentially preventable mortality (66-80%) on the battlefield. Death secondary to hemorrhagic shock occurs from both surgical bleeding and coagulopathy. Due to the knowledge of increased fibrinolysis promoting a hypocoagulable state in severe trauma, trials have been performed to determine if antifibrinolytics such as tranexamic acid (TXA) could reduce morbidity and mortality by reducing death from hemorrhage. TXA is an antifibrinolytic that inhibits both plasminogen activation and plasmin activity, thus preventing clot break-down rather than promoting new clot formation. Despite the extensive use of TXA in many surgical populations and an increasing use in severe trauma patients, TXA does not have an FDA approved indication for patients with traumatic injuries. The effect of TXA on immune function has not been thoroughly examined, especially in patients with severe traumatic injury. The study of the effects of TXA use on endothelial activation and injury is also important due to the inter-relationship between coagulation and endothelial function. Endothelial injury secondary to local hypoperfusion causes acute traumatic coagulopathy with fibrinolysis. Therefore a thorough and comprehensive evaluation of the effects of TXA on immune, coagulation, and endothelial parameters is important to allow for a better understanding of the mechanisms of action of this agent.

This is a randomized placebo controlled trial to obtain mechanism of action data, pharmacokinetic information, and efficacy and safety data for the use of TXA in severely injured trauma patients. Participants will be randomized into 1 of 3 treatment arms (1:1:1): TXA 2 gram IV bolus, TXA 4 gram IV bolus, or placebo. The study period is from time of enrollment to hospital discharge or transfer. The study intervention will occur only once upon enrollment in the trial. Participants will receive study drug within two hours from their initial injury. Blood samples will be drawn at multiple time points for immune parameters, Pharmacodynamics, and repository samples.

Immune parameter samples will be drawn at at approximately 0, 6, 24 and 72 hours after study drug/placebo administration.

Pharmacokinetic and pharmacodynamic samples will be drawn according to two schedules. Even number sampling times, blood will be drawn at the approximate time points: 0, 20 min, 1 hr, 2 hr, 4 hr, 6 hr, 8 hr, and 12 hr. A patient sampled on odd number sampling times will have samples drawn at the approximate time points: 0, 10 min, 40 min, 1.5 hr, 3 hr, 6 hr, 10 hr and 24 hr.

Repository samples will be drawn at approximate time points: 0, 1, 6, 24, and 72 hours.

Recruitment & Eligibility

Status: COMPLETED

Sex: All

Target Recruitment: 150

Inclusion Criteria

Patients with traumatic injury that are ordered to receive at least 1 blood product and/or
Patients admitted to the Emergency Department with a traumatic injury and require immediate transfer to the operating room to control the bleeding
Able to receive the study drug within 2 hours from estimated time of injury **Please note that in circumstances where the patient initially met inclusion/exclusion criteria (i.e. received blood products in the ED before a full evaluation of their injuries is complete) but is later found to only have a soft tissue involved injury or does not have a traumatic bleeding source), the Investigator may determine that the patient should not be randomized into the trial and the patient should be considered a screen failure

Exclusion Criteria

Patients known to be < 18 years of age
Suspected Acute MI or stroke(thromboembolic and/or hemorrhagic) on admission
Known inherited coagulation disorders
Known history of thromboembolic events (DVT, PE, MI, Stroke)

• Please note that past medical history of hemorrhagic stroke is permitted, but not current admission with hemorrhagic stroke
Known history of seizures and/or seizure after injury/on admission related to this hospitalization
Suspected or known pregnancy
Known to be lactating
Suspected or known prisoners
Futile care
Known current state of immunosuppression (i.e. on high dose steroids, chemotherapeutics, etc.)
Unknown estimated time of injury 12). Patients wearing an "Opt Out" TAMPITI Study bracelet 13). Known presence of subarachnoid hemorrhage.

14.) Isolated injuries to hands and/or feet (distal) 15.) Administration of antifibrinolytics pre-hospital and/or during this ED admission prior to enrollment

Study & Design

Study Type: INTERVENTIONAL

Study Design: PARALLEL

Arm && Interventions

Group	Intervention	Description
Placebo	Placebo	Matching Volume Normal Saline Placebo given IV over 10 minutes within 2 hours of initial injury
Tranexamic Acid 2 Gram	Tranexamic Acid	One time dose IV TXA 2 Grams given over 10 minutes within 2 hours of initial injury
Tranexamic Acid 4 Gram	Tranexamic Acid	One time dose IV TXA 4 Grams given over 10 minutes within 2 hours of initial injury

Primary Outcome Measures

Name	Time	Method
Change in HLA-DR Expression on Monocytes 72 Hours After Drug or Placebo Administration in Patient Groups (0g TXA (Placebo); 2g TXA; 4g TXA)."	Samples Drawn through 72 hours after study initiation	Blood was drawn from patients at baseline (0 h, just before placebo or drug administration) and at 72 hours post placebo or drug administration. Leukocytes in these blood samples were stained with fluroescent antibodies specific for CD45, CD14, and HLA-DR, analyzed by flow cytometry, and the median fluorescen intensity (MFI) of HLA-DR signal was recorded for monocytes (CD45+CD14+). The fold change in HLA-DR expression from prior to placebo/drug administration to 72 h after placebo/drug administration ("0 h : 72 h") was calculated as HLA-DR MFI72hours ÷ HLA-DR CD14 MFI0hours. Non-paramteric one-way ANOVA (Kruskal-Wallis test) was performed between each treatment group at the given time pont, and the p-value reported.

Secondary Outcome Measures

Name	Time	Method
Differences in Cytokine Profiles Between the Three Study Groups	Samples Drawn through 72 hours after study initiation	To evaluate the effects of TXA on immune function parameters we will, in a RCT, analyze samples from 150 patients (50 in each study group), at multiple time points. Parameters are: a. Cytokines measured from time 0 to 72 hours.
Differences in Leukocyte Function Parameters Between the Three Study Groups	Samples Drawn through 72 hours after study initiation	To evaluate the effects of TXA on immune function parameters we will, in a RCT, analyze samples from 150 patients (50 in each study group), at multiple time points. Parameters are: a. Flow cytometric analyses on leukocytes measured from time 0 to 72 hours.
Determine the Incidence of Seizures at 24 Hours in All Three Study Groups.	24 hours following TXA	The incidence of seizures at 24 hours in all three study groups. Number of participants with seizures are reported
Total Transfusion Volume CL	24 hours	Pharmacokinetic data was analyzed with NONMEM, using both the first-order and conditional non-Laplacian (with centering) estimation techniques. We considered two- and three-compartment models, parameterized in terms of both compartment volumes and clearances (distribution and elimination). We compared a basic model (in which pharmacokinetic parameters were independent of weight) to a model in which the pharmacokinetic parameters were assumed to be proportional to weight. The optimal model was selected on the basis of the objective function logarithm of the likelihood of the results) using standard criteria (NONMEM guide). Equations from optimal model: CL=109\((WT/70)\\0.75) \ (SCRint\^-0.084) \* ((NIRSInt)/96)\^ -0.27 ) \* ((PLTint)/130)\^0.45) V1=1,160\(WT/70) \ (TxTot)\^0.03) Q=174\((WT/70)\\0.75) V2=1080 \(WT/70) "Total Transfusion Volume CL" equals clearance (CL) affected by the covariate of Total Transfusion Volume (TxTot). This value is unitless per NONMEM reporting.
Determine the Incidence of Thromboembolic Events (DVT, MI, PE, Stroke) in All Three Study Groups.	Hospital Discharge (average 10 days)	The number of events per group for the incidence of thromboembolic events (DVT, MI, PE, Stroke) in all three study groups.
Q- Intercompartmental Clearance (L/70kg)	24 hours	Pharmacokinetic data was analyzed with NONMEM, using both the first-order and conditional non-Laplacian (with centering) estimation techniques. We considered two- and three-compartment models, parameterized in terms of both compartment volumes and clearances (distribution and elimination). We compared a basic model (in which pharmacokinetic parameters were independent of weight) to a model in which the pharmacokinetic parameters were assumed to be proportional to weight. The optimal model was selected on the basis of the objective function logarithm of the likelihood of the results) using standard criteria (NONMEM guide). Equations from optimal model: CL=109\((WT/70)\\0.75) \ (SCRint\^-0.084) \* ((NIRSInt)/96)\^ -0.27 ) \* ((PLTint)/130)\^0.45) V1=1,160\(WT/70) \ (TxTot)\^0.03) Q=174\((WT/70)\\0.75) V2=1080 \(WT/70) "Q" equals intercompartmental clearance in L/70kg.
Determine the Incidence of All Adverse Events in All Three Study Groups	Hospital Discharge (average 10 days)	All adverse events were totaled for each of the three study groups based on the number of incidents.
Platelet Count CL	24 hours	Pharmacokinetic data was analyzed with NONMEM, using both the first-order and conditional non-Laplacian (with centering) estimation techniques. We considered two- and three-compartment models, parameterized in terms of both compartment volumes and clearances (distribution and elimination). We compared a basic model (in which pharmacokinetic parameters were independent of weight) to a model in which the pharmacokinetic parameters were assumed to be proportional to weight. The optimal model was selected on the basis of the objective function logarithm of the likelihood of the results) using standard criteria (NONMEM guide). Equations from optimal model: CL=109\((WT/70)\\0.75) \ (SCRint\^-0.084) \* ((NIRSInt)/96)\^ -0.27 ) \* ((PLTint)/130)\^0.45) V1=1,160\(WT/70) \ (TxTot)\^0.03) Q=174\((WT/70)\\0.75) V2=1080 \(WT/70) "Platelet Count CL" equals clearance (CL) affected by the covariate of Platelet Count (PLTint). This value is unitless per NONMEM reporting.
Near Infrared Spectroscopy CL	24 hours	Pharmacokinetic data was analyzed with NONMEM, using both the first-order and conditional non-Laplacian (with centering) estimation techniques. We considered two- and three-compartment models, parameterized in terms of both compartment volumes and clearances (distribution and elimination). We compared a basic model (in which pharmacokinetic parameters were independent of weight) to a model in which the pharmacokinetic parameters were assumed to be proportional to weight. The optimal model was selected on the basis of the objective function logarithm of the likelihood of the results) using standard criteria (NONMEM guide). Equations from optimal model: CL=109\((WT/70)\\0.75) \ (SCRint\^-0.084) \* ((NIRSInt)/96)\^ -0.27 ) \* ((PLTint)/130)\^0.45) V1=1,160\(WT/70) \ (TxTot)\^0.03) Q=174\((WT/70)\\0.75) V2=1080 \(WT/70) "Near Infrared Spectroscopy CL" equals clearance (CL) affected by the covariate of Near Infrared Spectroscopy (NIRSint). This value is unitless per NONMEM reporting.
Creatinine Count CL	24 hours	Pharmacokinetic data was analyzed with NONMEM, using both the first-order and conditional non-Laplacian (with centering) estimation techniques. We considered two- and three-compartment models, parameterized in terms of both compartment volumes and clearances (distribution and elimination). We compared a basic model (in which pharmacokinetic parameters were independent of weight) to a model in which the pharmacokinetic parameters were assumed to be proportional to weight. The optimal model was selected on the basis of the objective function logarithm of the likelihood of the results) using standard criteria (NONMEM guide). Equations from optimal model: CL=109\((WT/70)\\0.75) \ (SCRint\^-0.084) \* ((NIRSInt)/96)\^ -0.27 ) \* ((PLTint)/130)\^0.45) V1=1,160\(WT/70) \ (TxTot)\^0.03) Q=174\((WT/70)\\0.75) V2=1080 \(WT/70) "Creatinine Count CL" equals clearance (CL) affected by the covariate of Creatinine levels (SCRint). This value is unitless per NONMEM reporting.
V2- Peripheral Volume (L/70kg)	24 hours	Pharmacokinetic data was analyzed with NONMEM, using both the first-order and conditional non-Laplacian (with centering) estimation techniques. We considered two- and three-compartment models, parameterized in terms of both compartment volumes and clearances (distribution and elimination). We compared a basic model (in which pharmacokinetic parameters were independent of weight) to a model in which the pharmacokinetic parameters were assumed to be proportional to weight. The optimal model was selected on the basis of the objective function logarithm of the likelihood of the results) using standard criteria (NONMEM guide). Equations from optimal model: CL=109\((WT/70)\\0.75) \ (SCRint\^-0.084) \* ((NIRSInt)/96)\^ -0.27 ) \* ((PLTint)/130)\^0.45) V1=1,160\(WT/70) \ (TxTot)\^0.03) Q=174\((WT/70)\\0.75) V2=1080 \(WT/70) "V2" equals Peripheral Volume in L/70kg.
V1- Central Volume (L/70kg)	24 hours	Pharmacokinetic data was analyzed with NONMEM, using both the first-order and conditional non-Laplacian (with centering) estimation techniques. We considered two- and three-compartment models, parameterized in terms of both compartment volumes and clearances (distribution and elimination). We compared a basic model (in which pharmacokinetic parameters were independent of weight) to a model in which the pharmacokinetic parameters were assumed to be proportional to weight. The optimal model was selected on the basis of the objective function logarithm of the likelihood of the results) using standard criteria (NONMEM guide). Equations from optimal model: CL=109\((WT/70)\\0.75) \ (SCRint\^-0.084) \* ((NIRSInt)/96)\^ -0.27 ) \* ((PLTint)/130)\^0.45) V1=1,160\(WT/70) \ (TxTot)\^0.03) Q=174\((WT/70)\\0.75) V2=1080 \(WT/70) "V1" equals central volume in L/70kg.
CL- Clearance of TXA (mL/(Min*70kg))	24 hours	Pharmacokinetic data was analyzed with NONMEM, using both the first-order and conditional non-Laplacian (with centering) estimation techniques. We considered two- and three-compartment models, parameterized in terms of both compartment volumes and clearances (distribution and elimination). We compared a basic model (in which pharmacokinetic parameters were independent of weight) to a model in which the pharmacokinetic parameters were assumed to be proportional to weight. The optimal model was selected on the basis of the objective function logarithm of the likelihood of the results) using standard criteria (NONMEM guide). Equations from optimal model: CL=109\((WT/70)\\0.75) \ (SCRint\^-0.084) \* ((NIRSInt)/96)\^ -0.27 ) \* ((PLTint)/130)\^0.45) V1=1,160\(WT/70) \ (TxTot)\^0.03) Q=174\((WT/70)\\0.75) V2=1080 \(WT/70) "CL" equals clearance of TXA in mL/(min\*70kg).