Introduction
Severe injury causes coagulopathy in up to 25% to 33% of human trauma victims, and this complication, known as acute traumatic coagulopathy (ATC), may worsen hemorrhage and increase the risk of death compared to people with similar injuries but without ATC.1,2 In people, ATC is characterized by either impaired clot formation alone or impaired clot formation concurrent with excessive clot lysis, known as hyperfibrinolysis (HF). Coagulation dysfunction with HF also occurs in severely injured dogs, dogs with spontaneous hemoabdomen from ruptured neoplastic masses, and dogs with neoplasia.3–11 A prospective study6 identified ATC in 15% of severely injured dogs, and its development was correlated with severity of trauma, hypotension, and hyperlactatemia. Nonsurviving dogs in that study had HF, with a mean percent of clot lysis at 30 minutes (LY30%) measured with kaolin-activated thromboelastography that was 3X the upper end of their reference interval. The population incidence of HF as a component of ATC in dogs is unknown, but another recent study12 identified HF in a similar proportion of dogs diagnosed with ATC (3 of 11 [27%]) and 9% of all trauma cases.12
Hyperfibrinolysis is a maladaptive overactivation of the physiologic process of clot breakdown (fibrinolysis) and vessel reperfusion. Although clot lysis is part of the normal and balanced pro- and anticlotting processes invoked by injury, acceleration of lysis occurs in some cases with severe hemorrhagic shock and/or high injury severity. It increases the likelihood that injured humans will require massive transfusion, and it is independently associated with increased risk of death.13–15 The mechanisms driving HF in trauma are not fully understood, but potential contributors include overwhelming release of tissue plasminogen activator (tPA) from activated endothelium,16,17 decreased hepatic clearance of tPA due to decreased perfusion or receptor saturation,18 activated protein C–mediated inhibition of endogenous antifibrinolytics,19 and activation of an alternative pathway of fibrinolysis mediated by neutrophil elastase.20 Regardless of the mechanism, multiple studies21–23 indicate that early (within the first 3 hours) treatment of bleeding trauma patients with lysine-analog antifibrinolytic drugs such as tranexamic acid (TXA) can be lifesaving.21–23 Tranexamic acid blocks the tPA-mediated conversion of plasminogen to plasmin, preventing the degradation of fibrin and allowing clot persistence. The good efficacy, favorable safety data, low cost, and high portability of this hemostatic drug have led to its routine use in military and prehospital settings as well as placement of the drug on the World Health Organization’s List of Essential Medicines.24
If canine ATC responds to TXA in a similar manner, there may be a role for use of the drug to inhibit HF in dogs with severe injury or uncontrolled hemorrhage from neoplasia. This is of particular interest for treatment of combat-wounded working dogs, in which high mortality rates have been documented and 93% of deaths occur in the prehospital setting, indicating a critical need for portable hemostatic agents. 25,26 Large-scale veterinary studies27–29 have demonstrated a 5.7% to 12% overall mortality for civilian-owned dogs presenting for trauma care, but the majority of these cases had only minor injuries. A recent study30 evaluating dogs with high injury severity and a mean animal trauma triage (ATT) score of 7.4 reported a 44% mortality rate. Injury severity directly impacts mortality, with the risk of death increasing sharply as ATT scores rise above 4.31
Hyperfibrinolysis can be detected with viscoelastic testing, a technique that provides comprehensive characterization of whole blood coagulation dynamics and is performed through several different methodologies such as thromboelastography, thromboelastometry, and glass surface activation of whole blood. In these assays, the kinetics of clot initiation, propagation, maturation, and breakdown (lysis) over time are assessed with a device that measures changes in viscosity of a blood sample that has been activated to form a clot. Clot lysis is measured as the percent reduction from maximum clot viscosity 30 and 60 minutes after maximum clot strength is reached. Patients with clinical bleeding disorders that also demonstrate lysis in excess of a normal reference range are assumed to have HF. In dogs, the reference ranges for LY30% and percent of clot lysis at 60 minutes (LY60%) are extremely low, often close to 0. Therefore, to assess the antifibrinolytic effect of TXA administered to healthy dogs, investigators must modify the preanalyzed sample with the addition of recombinant tPA to induce HF. This technique, called tPA-thromboelastography, has been used to determine pharmacokinetics and viscoelastic effects of TXA in healthy dogs.32,33 Those studies demonstrated that TXA doses ranging from 10 to 20 mg/kg, IV, can impact clot dissolution for several hours, with the higher dose yielding lower LY60% values for up to 6 hours after administration. The pharmacokinetic and pharmacodynamic effects of TXA may be altered in dogs with hemorrhagic shock, and the dose required to reverse naturally occurring HF has not been prospectively evaluated in that setting.
Our study was designed to determine the antifibrinolytic effect of a continuous TXA infusion protocol in dogs with naturally occurring HF and hemorrhagic shock secondary to trauma or spontaneous hemoabdomen due to rupture of hepatosplenic neoplasia. We elected to administer a continuous infusion immediately following a loading dose to conform to contemporary treatment protocols in humans, to increase the likelihood that there was a continuous drug effect during the initial stabilization period, and to allow us to confirm that HF was controlled at the conclusion of continuous exposure at T8.23,34 Our hypothesis was that the dosing regimen used in the study protocol will reverse HF detected by thromboelastography. To increase the sensitivity of the assay to detect HF we used both thromboelastography and tPA-thromboelastography to detect HF and observe the effect of treatment with TXA. Secondary aims include assessment of the impact of TXA administration on blood product usage and adverse events in dogs when compared to placebo.
Methods
Determination of normal ranges for healthy dogs and the research protocol for subject dogs were approved by the IACUCs at both institutions and by the study sponsor.
Establishment of normal values for clot lysis
Healthy pet dogs from 2 participating institutions were used to establish normal ranges for rapid thromboelastography (rTEG; RapidTEG; Haemonetics) and tPA-stressed rTEG (tPA-stressed rTEG) parameters following the protocol described below. Control dogs, owned by faculty, staff, or students, were 1 to 10 years old with no history of recent illness. They were assessed as normal based on physical examination at the time of blood collection and normal clinical pathology (CBC and biochemistry profile) within the previous 3 months. Initial normal ranges for the thromboelastography parameters reaction time, clot kinetic time, angle, maximum amplitude, and LY30% were established using samples from 21 dogs run in triplicate. Values that were outliers using the Tukey fence method were censored. Values for reaction time and LY30% in both the stressed and unstressed assays were not normally distributed, and CIs for the median values were determined; LY30% values of 0.05 were substituted for measured values of 0. The initial approach to enrollment was to include only dogs whose LY30% values were greater than the 95% upper limit of control LY30% determined (MedCalc; Ostend) using a 90% CI with a robust nonparametric percentile method.35 Thus, our original cutoffs for normal LY30% for rTEG and tPA-stressed rTEG were set at 3.8% and 62%, respectively; values equal to or greater than those defined HF and permitted enrollment. The cutoff for tPA-stressed rTEG was later revised downward as described later in this report.
Thromboelastography assay procedure
A 6.1-mL venous blood sample was collected from either a winged needle catheter atraumatically inserted for peripheral venipuncture or a central venous catheter inserted immediately prior to blood collection. For either technique, a purge of 1.2 mL (winged needle) or 3 mL (central venous catheter) of blood was collected before diagnostic sample collection. The sample was divided into four 1.4-mL citrate tubes for the rTEG assays and one 0.5-mL EDTA tube for a CBC and PCV, total protein, and blood lactate measurement. The citrate tubes were mixed by inversion 5 times, then allowed to rest at room temperature for 30 minutes. During the rest period, the investigator retrieved a vial of prediluted tPA solution from freezer storage. The tPA solution was made in advance by serial dilutions of stock product (Cathflo Activase [Alteplase]; Genentech; 2 mg/mL, in 2% bovine serum albumin-HEPES buffer solution to a final concentration of 25 IU/10 μL). The tPA solution was made fresh every 6 months and stored at –80 °C prior to use. The tubes were numbered to ensure a standardized sequence of use, and each tube was assigned to either the left or the right channel of the TEG analyzer to ensure an equal distribution of channel assignment for the tPA-stressed rTEG assays. The rTEG vials containing tissue factor activator and kaolin were reconstituted with 20 μL distilled water and gently swirled, then allowed to rest. Immediately before assay, 10 μL of rTEG mixture and 20 μL of calcium chloride solution were added to each TEG cup. For the rTEG assays, the citrated tube was gently inverted 5 times, and 340 μL of blood was pipetted directly into the test cup and gently mixed by pipetting the reagent-blood mixture in and out of the cup 3 times, and the analyzer carriage was raised to start the test. For the tPA-stressed rTEG assay, a 490-μL aliquot of citrated blood was pipetted into the tPA vial and was also gently mixed, yielding a final concentration of tPA in blood of 50 IU/mL. Then, 340 μL of the blood-tPA mixture were added to the test cup assigned to the tPA-stressed rTEG. This mixture also was pipetted out of and into the cup 3 times to mix all components before the assay was begun. Once the assays began, the remaining citrated blood was spun and the plasma frozen at –80 °F (–62 °C) for other studies.
Study subjects
Study subjects were initially restricted to client-owned dogs weighing 4 to 75 kg with traumatic shock that were injured within the previous 3 hours; these underwent screening assessment at the 2 contributing university emergency departments. Dogs that had been treated with antithrombotic drugs within the previous 2 weeks or that had received IV fluid therapy with synthetic colloid solutions after the injury were not eligible for further screening. Protocol triage assessment included vital signs, point-of-care ultrasound examination of the pleural space and abdomen, venous PCV, total protein, blood gas and lactate assays, and calculation of shock index. At least 1 identifiable source of hemorrhage was required; this could include body cavity hemorrhage, hemorrhage at fracture sites, or hemorrhage from wounds. Enrollment also required that the dog had bled enough to benefit, in the opinion of the attending clinician, from transfusion with at least 10 mL/kg of plasma and 10 mL/kg of packed RBCs or 20 mL/kg of whole blood.
Dogs meeting triage physical examination assessment criteria were considered for rTEG screening if they met 1 or more of 3 additional criteria: (1) a shock index > 1, (2) a blood lactate concentration > 2 mmol/L, or (3) a base excess < –6 mmol/L.36 After obtaining informed consent from the owner, dogs underwent a second point-of-care ultrasound examination of the chest and abdomen to recheck any effusion and score its severity. Effusion samples were collected to confirm that the effusion was due to hemorrhage, and ATT and modified Glasgow coma scores were calculated. Dogs that qualified based on the above findings were then further screened for the presence of HF via the use of thromboelastography. All subject and treatment data were entered into an interactive spreadsheet (Excel; Microsoft Corp) that provided protocol instructions in sequence as information was added.
By the end of study year 1, only 2 dogs had qualified for enrollment, prompting us to increase the scope of enrollment criteria in 2 ways. First, the latency from injury to enrollment was increased from 3 hours to 24 hours, based on observations during year 1 that some otherwise eligible animals with HF had presented to the emergency departments in hemorrhagic shock that had progressed for over 24 hours following injury. Second, given the paucity of qualified traumatic shock victims and the similar HF coagulopathy associated with both traumatic and spontaneous hemoperitoneum,10,11 we enrolled some dogs with shock from severe hemorrhage caused by spontaneous rupture of a liver or splenic mass and meeting the same criteria for shock index, blood lactate concentration, and base excess.
During study year 2 we obtained comprehensive coagulation testing on stored plasma collected from 4 dogs (3 with trauma, 1 with spontaneous hemorrhage) that had met the original LY30% criteria and 7 dogs (4 with trauma, 3 with abdominal hemorrhage) that had been excluded because they did not. We compared the results of multiple tests (activated partial thromboplastin time, prothrombin time, fibrinogen, antithrombin, d-dimer, protein C activity, antiplasmin activity, plasminogen activity, and thrombin-antithrombin concentration) with the thromboelastography findings. Five of these 11 dogs, including 4 with 1 or both T0 LY30% values too low for enrollment, had screening tPA-stressed rTEG LY30% values of ≥ 20% but < 62%. All 5 had abnormal coagulation panels, including low protein C activity (n = 4), high thrombin-antithrombin concentration (4), high d-dimer concentration (3), prolonged activated partial thromboplastin time and prothrombin time and low fibrinogen concentration (2), and low antithrombin and/or antiplasmin (2), consistent with severe acquired coagulopathy. Based on those findings we thereafter reduced the minimum tPA-stressed rTEG LY30% threshold for enrollment from 62% to 20% (Table 1). If HF was identified by using either rTEG method (rTEG or tPA-stressed rTEG), the dog was enrolled into the treatment protocol.
Summary of initial and revised enrollment criteria for all study dogs (n = 25) from March 16, 2018, to May 20, 2022.
Initial criteria
|
Revised criteria
|
In response to low enrollment and based on the results of comprehensive coagulation testing for enrolled and rejected dogs, the initial criteria were revised to more liberal criteria in 2019.
rTEG = Rapid thromboelastography. tPA-stressed rTEG = Tissue plasminogen activator–stressed rapid thromboelastography. LY30% = Percent reduction of clot amplitude 30 minutes after achieving maximum clot strength.
Treatment protocol
All dogs were pretreated with maropitant (Cerenia; Zoetis), 1 mg/kg, IV, 5 minutes to 3 hours prior to injection of the study drug. Treatment kits containing three 10-mL vials of test solution (either TXA solution [Tranexamic acid injectable solution 100 mg/mL; AuroMedics Pharma] or saline [SAL]) were prepared in advance by pharmacy staff unaffiliated with the study in blocks of 4 (2 TXA and 2 SAL), and the order of use was randomized by a single pharmacist. The interactive spreadsheet provided customized weight-based dilution and administration instructions to minimize the risk of dosing or other administration errors. The loading dose of test solution was diluted to 30 mg/mL according to weight-specific instructions provided by the data entry template and administered over 20 minutes. Once the loading dose was complete (T0), a continuous infusion of test solution (20 mg/kg of TXA or an equivalent volume of saline) was administered over the next 8 hours.
Resuscitation was accomplished by hemorrhage control (provided by tamponade, surgical correction, or surgical removal of a bleeding mass, as needed), transfusion with blood products, administration of crystalloid fluids, and care of injuries for the traumatized dogs. All dogs received transfusions of at least 20 mL/kg of whole blood or a combination of packed RBCs and plasma after enrollment and before the end of the 8-hour test solution infusion. Additional blood products were administered, and routine care of injuries was provided at the attending veterinarians’ discretion. Blood samples were collected for PCV/total protein, lactate, rTEG, and tPA-stressed rTEG assays 8, 12, and 24 hours after administration of the TXA loading dose (T8, T12, and T24, respectively).
All investigators were blinded to the test solution identity, but the protocol allowed unblinding in the event that uncontrolled hemorrhage persisted after the loading dose of study drug was administered if, in the attending clinician’s judgment, the dog needed to be switched from the SAL to TXA group.
Statistical analysis
The primary outcome measure of interest was treatment effect on LY30% in both the rTEG and tPA-stressed rTEG assays. Secondary objectives included determining any effect of treatment on survival to 24 hours, survival to discharge, blood product usage in the first 24 hours, and frequency of adverse effects. For analysis purposes, whole blood volume was converted to an estimate of packed RBC and fresh frozen plasma volumes by assuming a PCV of 50%; therefore, we divided the whole blood volume by 2 to estimate the volume of each component.
Comparisons of changes in TEG parameters and other clinical pathology data values between treatment groups and between time points were done with the Wilcoxon rank sum test and Wilcoxon matched-pairs signed rank test, respectively, and examinations of the association between shock/ATT scores and LY30% were done via the Spearman correlation. The Fisher exact test was used to compare types of injuries at T0 and survival between groups at T24 and at discharge. Associations between time to treatment and survival in each treatment group were done with a Wilcoxon rank sum test within each treatment group. The relationships between LY30% and each of the regular and tPA-stressed rTEG parameters reaction time, clot kinetic time, angle, and maximum amplitude at T0 were examined with the Fisher exact test. The association between treatment group and the total volume of blood products administered was examined with the Wilcoxon rank sum test. All analyses were conducted in R, version 4.2.1.37
Results
Enrolled dogs presented with hemorrhagic shock from trauma (n = 21) or spontaneous rupture of an intra-abdominal mass (4). Causes of trauma included motor vehicle accidents (n = 17), fight with another dog (2), penetrating injury (1), and fall from a cliff ledge (1). Seven dogs were enrolled at NCSU College of Veterinary Medicine and 18 were enrolled at the Cummings School of Veterinary Medicine. Dogs in the TXA group (n = 14) include 1 female, 1 male, 6 spayed females, and 6 castrated males. Dogs in the SAL group (n = 11) included 1 female, 4 male, 3 spayed females, and 3 castrated males. There were no differences between treatment groups for the T0 enrollment criteria of shock index, base excess, and blood lactate concentration. Similarly, there were no T0 differences between treatment groups for ATT, modified Glasgow coma, point-of-care ultrasound body cavity effusion scores, PCV, total protein, platelet count, volume of crystalloids administered during initial stabilization efforts, or time from injury to treatment (Table 2). The number of dogs with confirmed pleural or peritoneal hemorrhage, external wounds, and fractures did not differ between groups (Table 3). The group identity of 1 dog was revealed due to evidence of persistent hemorrhage between T0 and T8; no protocol changes were made for that subject. Eighteen dogs (9 in each group) had elevated rTEG LY30% values at T0. All 14 TXA group dogs and 7 of 11 SAL group dogs had elevated stressed rTEG LY30% values at T0 (Table 4).
Baseline characteristics of dogs treated with tranexamic acid (TXA; n = 14) or saline placebo (SAL; 11)
| Parameter | SAL median (IQR) | TXA median (IQR) | P value |
|---|---|---|---|
| Weight (kg) | 24.2 (18.8 to 27.7) | 18.7 (9.2 to 26.3) | .298 |
| Age (y) | 1.5 (1.2 to 4.2) | 4.6 (0.8 to 8) | .805 |
| Shock index | 1.9 (1.7 to 2.4) | 1.7 (1.3 to 2) | .094 |
| Lactate (mmol/L) | 5.2 (3.4 to 6.7) | 4.3 (3.6 to 5.4) | .528 |
| BE (mEq/L) | –7.7 (–9.6 to –5.3) | –9 (–10.6 to –7.1) | .66 |
| ATT | 5 (4 to 5) | 4 (3.3 to 5) | .196 |
| Modified GCS | 18 (18 to 18) | 18 (17 to 17.5) | > .99 |
| Abdominal effusion score | 2 (0.5 to 2) | 2 (1.3 to 3.8) | .409 |
| Crystalloid volume (mL/kg) | 13.3 (7.2 to 31) | 20.9 (18.8 to 35.6) | .41 |
| Time to treatment (min) | 330 (291 to 367) | 267 (192 to 379) | .35 |
| Platelet count (X 103/μL) | 180 (115 to 218) | 167 (127 to 209) | > .99 |
| PCV (%) | 48 (42 to 50.5) | 42.5 (38 to 51.8) | .641 |
| TP (gm/dL) | 5.5 (5.2 to 6) | 5.5 (5.2 to 6.2) | .826 |
| rTEG R (s) | 1 (0.6 to 1.4) | 0.8 (0.7 to 1.6) | .621 |
| Stressed rTEG R (s) | 0.9 (0.8 to 1.2) | 0.8 (0.6 to 1.7) | .378 |
| rTEG K (s) | 3.5 (2.7 to 4.1) | 2.8 (2.1 to 3.3) | .241 |
| Stressed rTEG K (s) | 4.2 (3 to 4.9) | 2.8 (2.4 to 4.2) | .417 |
| rTEG angle (slope) | 54.3 (47.5 to 60.7) | 57.6 (42.9 to 64.5) | .913 |
| Stressed rTEG angle (slope) | 53.7 (45 to 59.9) | 59.3 (37 to 63) | .809 |
| rTEG MA (mm) | 40.5 (36.7 to 43.5) | 43.3 (28.9 to 49.4) | .913 |
| Stressed rTEG MA (mm) | 38.4 (35.9 to 41.3) | 43.6 (31.2 to 48.8) | .529 |
| rTEG LY30% | 18.5 (6.9 to 42) | 13.7 (2 to 24.7) | .297 |
| Stressed rTEG LY30% | 42.3 (16 to 56.9) | 41.5 (28.6 to 60.3) | .366 |
Shock index is heart rate/systolic blood pressure. Abdominal effusion score is the number of quadrants with effusion found on ultrasound examination. Time to treatment is the minutes elapsed between injury or onset of signs and treatment. Angle is the slope of the rate of increase in clot strength, in degrees.
ATT = Animal trauma triage score. BE = Base excess. GCS = Glasgow coma score. K = Kinetics time, or time for clot strength to reach 20% of maximum amplitude. MA = Maximum amplitude of clot strength, in millimeters. R = Reaction time. TP = Plasma total protein.
Distribution of hemorrhage sources between treatment groups.
| Finding | SAL (n = 11) | TXA (n = 14) | P value |
|---|---|---|---|
| Peritoneal hemorrhage | 8 | 10 | .689 |
| Pleural hemorrhage | 2 | 2 | > .99 |
| Torso/neck wounds | 2 | 2 | > .99 |
| Other lacerations | 3 | 1 | .288 |
| Femur fracture | 1 | 1 | > .99 |
| Pelvic fracture | 1 | 0 | .44 |
| Other fractures | 2 | 1 | .565 |
T0 thromboelastogram parameters for TXA and SAL group dogs.
| rTEG R (min) | rTEG K (min) | rTEG angle (°) | rTEG MA (mm) | rTEG LY30% | Stressed rTEG R (min) | Stressed rTEG K (min) | Stressed rTEG angle (°) | Stressed rTEG MA (mm) | Stressed rTEG LY30 (%) | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Group | Assay reference interval: | 0.5–0.8 | 2.4–2.9 | 59.2–64.3 | 50.2–53.7 | 0–3.8 | 0.5–0.6 | 2.4–2.9 | 59.1–63.8 | 48.5–52 | 0–20 |
| TXA (n = 14) | No. above RR (%) | 6 (42.9) | 7 (50) | 4 (28.6) | 2 (14.3) | 9 (64.3) | 9 (64.3) | 7 (50) | 2 (14.3) | 2 (14.3) | 14 (100) |
| No. within RR (%) | 8 (57.1) | 1 (7) | 2 (14.3) | 1 (7.1) | 5 (35.7) | 4 (28.6) | 4 (28.5) | 5 (35.7) | 3 (21.4) | 0 (0) | |
| No. below RR (%) | 0 (0) | 6 (43) | 8 (57.1) | 11 (78.6) | 0 (0) | 1 (7.1) | 3 (21.4) | 7 (50) | 9 (64.3) | 0 (0) | |
| SAL (n = 11) | No. above RR (%) | 6 (54.5) | 8 (72.7) | 2 (18.2) | 2 (18.2) | 9 (81.8) | 10 (90.9) | 8 (72.7) | 2 (18.2) | 1 (9.1) | 7 (63.6) |
| No. within RR (%) | 3 (27.3) | 0 (0) | 1 (9.1) | 0 (0) | 2 (18.2) | 0 (0) | 1 (9.1) | 2 (18.2) | 0 (0) | 4 (36.4) | |
| No. below RR (%) | 2 (18.2) | 3 (27.3) | 8 (72.7) | 9 (81.8) | 0 (0) | 1 (9.1) | 2 (18.2) | 7 (63.6) | 10 (90.9) | 0 (0) |
Whole numbers indicate the number of dogs within each treatment group with rTEG and tPA-stressed rTEG values above, within, or below the reference ranges (RRs) established using healthy dogs, and parenthetical values are the percent of dogs within that treatment group. The ranges given at the top of each column represent the extremes of a 95% CI around the normal dog mean K, angle, and MA, or the median R. The range for rTEG LY30% was established according to the Clinical and Laboratory Standards Institute method,35 and the range for the tPA-stressed LY30% values was established empirically as described in the text. Bold values in the table body indicate abnormal TEG parameters that are associated with hypocoagulability. All dogs had an abnormality of LY30% and/or tPA-stressed LY30%, as required for enrollment.
Primary outcome measure
The rTEG LY30% values were 0 for all but 3 dogs at T8 and for all but 2 dogs at T12 and T24; there was no difference between treatment groups at T0 (P = .279) or at T8, T12, or T24 (P = .663, .43, and .189 respectively). In 1 nonsurviving TXA group dog this value increased from 0 to 6.9% at T8 and fell to 0% at T12 while its tPA-stressed rTEG LY30% fell from 31.1% to 0.2% (T8) and 0% (T12).
The T0-T8 difference in tPA-stressed rTEG LY30% values was significant within both the SAL (P = .01) and TXA (P = .0001) groups as assessed with the Wilcoxon matched-pairs test, and remained so at T12. The TXA group median tPA-stressed LY30% values at T0, T8, and T12 were 41.5, 0, and 0, respectively, and the SAL group median tPA-stressed LY30% values were 42.3, 10.8, and 6.75, respectively, at T0, T8, and T12 (Figure 1). Median tPA-stressed rTEG values were lower in the TXA group than the SAL group at both T8 (P = .001) and T12 (P = .02), and no TXA group dogs had values above 6.9% from T8 forward.
Box-and-whisker plot of the percent of clot lysis at 30 minutes, with tissue plasminogen activator–stressed rapid thromboelastography (rTEG), at T0, T8, and T12 for the saline (SAL) and tranexamic acid (TXA) groups. Dotted pattern boxes represent SAL group dogs, and boxes with diagonal lines represent TXA group dogs. The bottom and top of each box represent the upper boundaries of the 1st and 3rd quartiles, respectively, and the lower and upper whiskers represent the lower and upper boundaries of the 1st and 4th quartiles, respectively. Median values (denoted by a horizontal line) for TXA dogs were significantly lower than those of SAL dogs at T8 (P = .001) and T12 (P = .02). Mean values are denoted by an x.
Citation: Journal of the American Veterinary Medical Association 2024; 10.2460/javma.24.03.0216
One or more of the rTEG and tPA-stressed rTEG parameters reaction time, clot kinetics time, angle, and maximum amplitude were abnormal in all dogs at T0 (Table 4). No dogs achieved normalization of all rTEG parameters during the first 24 hours, and there was no significant treatment effect on any rTEG parameter except the tPA-stressed rTEG LY30% at T8 and T12.
Secondary outcome measures
Four dogs with traumatic injury (2 in each treatment group) were euthanized prior to completion of the study due to complications (including suspected continued hemorrhage) that warranted a poor prognosis, and a fifth dog (in the TXA group) died of its traumatic injuries. The 3 SAL group dogs with the highest tPA-stressed rTEG LY30% values at T8-T24 all survived to discharge. There was no statistical difference between groups in survival to discharge (P > .99) or survival at 24 hours (P = .604). No adverse effects were attributed to treatment with TXA.
All dogs received at least 20 mL/kg of blood products; additional blood products were administered based on clinical need as perceived by the attending clinicians (most of whom were not associated with the study) and included stored whole blood, stored packed RBCs, and fresh frozen plasma. The administered volumes were converted to a milliliter-per-kilogram basis. There was no difference in the volume of blood products administered to each group. The median (IQR) administered volume of packed RBCs in the TXA and SAL groups was 10.8 mL/kg (8.88 to 12.95 mL/kg) and 11.3 mL/kg (9.5 to 13.1 mL/kg), respectively (P = .701), and the volume of fresh frozen plasma was 12.6 mL/kg (10.85 to 18.7 mL/kg) and 11.6 mL/kg (8.8 to 14 mL/kg), respectively (P = .494).
Discussion
Our results show that resuscitation of dogs in hemorrhagic shock, with or without treatment with TXA, almost completely abolished HF (as measured with rTEG) within 8 hours. Either treatment also significantly reduced HF as measured by the more sensitive tPA-stressed rTEG assay, but the addition of treatment with TXA yielded more complete suppression of HF as measured by that method. Although the drug appeared to be safe, there was no effect on blood product use or survival. The TXA dose we used was chosen based on previous work in healthy dogs, and it is not known whether a lower dose might provide a similar effect in dogs with HF associated with ATC or it is necessary to provide the drug as a continuous infusion.33
We modeled our TXA treatment method (loading dose followed by an infusion) on a recommended protocol for humans.23,34 Direct extrapolation of human TXA dosing protocols may not be appropriate for dogs based on evidence that, even in health, dogs have higher natural fibrinolytic activity than people.38–40 Canine plasma contains more plasminogen activator than human plasma,40 and 1 in vitro study41 demonstrated that canine plasma concentrations of TXA consistent with effective dosing in humans were not sufficient to normalize tPA-modified thromboelastography results. While some studies42–44 have failed to document clinical or thromboelastography effects following TXA administration to dogs, other authors9,45,46 report significant treatment effects on clinical findings or coagulation parameters in bleeding dogs. However, those reports suggesting positive effects include a retrospective study that presented only a description of its use in a referral hospital or case reports/series describing its use for neoplasia or trauma.
We used LY30% on rTEG and tPA-stressed rTEG as the diagnostic indicator of HF for both enrollment and treatment response. The tPA-stressed rTEG method increased the sensitivity of the viscoelastic test for assessing fibrinolysis, a parameter in which derangements are usually difficult to detect in the dog due to very low LY30% reference ranges. However, 18 of 21 (86%) of the dogs with trauma-induced hemorrhage qualified for the trial on the basis of their rTEG alone, and the tPA-stressed method was not necessary to detect HF in those dogs. This finding contrasts to that of the dogs with nontraumatic hemorrhage, in which 4 of 4 dogs failed to qualify on rTEG and qualified only by the tPA-stressed method, a finding similar to that previously reported.10 This suggests that unmodified rapid viscoelastic testing with rTEG to identify HF in traumatically injured, bleeding dogs may be more sensitive than previously thought, and it may be a useful tool for detecting dogs that are more likely to benefit from therapy. A physiological basis for the increase in rTEG LY30% from 0 (T0) to 6.9% (T8) in 1 nonsurvivor TXA group dog is difficult to explain, as the more sensitive tPA-stressed rTEG LY30% values fell from 31.1 to 0.2 at the same time, suggesting that the T8 rTEG value was a measurement error.
In injured people, TXA confers the greatest survival benefit when the drug is administered within an hour following injury,23 then diminishes by 10% for every 15 minutes of treatment delay until 3 hours, after which the benefit is much reduced.47 If the response is similar in dogs, the relatively long time from bleeding onset to TXA administration (median, 267 minutes) may have impacted our ability to detect differences in clinical outcome. The elapsed time to treatment in the 3 dogs in the TXA group that had suspected ongoing hemorrhage at the time of death or euthanasia was 3 to 10 hours.
Relatively low ATT scores and associated mortality probably also blunted any treatment effect. The average ATT score was 4.2 and 4.7 in the TXA and SAL groups, respectively, values that are associated with a baseline mortality of approximately 20%.48 Even if TXA had conferred a 100% survival treatment effect, this study was not adequately powered to detect a mortality difference between groups. A post hoc calculation showed that 58 dogs (29/group) would have been required to achieve 80% power using a 23% control mortality rate. Alternatively, our sample size would have been sufficient if mortality rate in the control group were 44%, more consistent with ATT scores in the range of 6 to 8.48
Resuscitation with blood products and crystalloids combined with surgical control of hemorrhage in dogs with spontaneous hemoabdomen has been associated with resolution of HF in that condition.11 All 4 dogs in our study with a bleeding abdominal mass had normal T0 rTEGs and elevated T0 tPA-stressed rTEGs that normalized by T8. Three of these dogs were in the TXA group, and the relative role of TXA in resolution of HF, if any, is unknown.
In the SAL group, the T0 rTEG LY30% was elevated in 9 of 11 dogs, and all of these normalized by T8, suggesting that resuscitation with hemorrhage control combined with administration of blood products and crystalloid fluids resolved much of the HF, even without antifibrinolytic therapy. Similarly, the tPA-stressed rTEG LY30% fell between T0 and T8 in 9 of 11 SAL group dogs and normalized in 8, though the magnitude of reduction was greater in the TXA group. This observation is in alignment with laboratory animal evidence that plasma resuscitation decreases LY30% during experimentally induced hemorrhagic shock.49 It is also possible that resolution of shock alone, independent of blood product administration, can restore the balance of clot degradation. However, clinical14 and laboratory49,50 evidence suggests that HF may be exacerbated when high-volume crystalloid fluid administration is used to resolve shock.
Tranexamic acid administration carries the potential for side effects and is not indicated unless severe trauma or shock has induced a state of HF with clinical bleeding. The most obvious side effects are nausea and vomiting when it is administered at higher doses without pretreatment with an antiemetic. In 1 report,32 1 of 10 normal Beagles vomited following administration of 20 mg/kg, IV, and 2 of 2 dogs vomited following administration of 50 mg/kg. A retrospective study45 of 266 dogs that were treated with a median dosage of 10 mg/kg every 6 hours for 4 treatments reported no vomiting, and only 4 dogs had hypersalivation, but most of those dogs were pretreated with maropitant. The same report described 1 dog that received 280 mg/kg over 27 hours and developed seizures that were attributed to the drug. None of the dogs in our study vomited despite the higher loading dose that we used, a finding we attribute to the pretreatment with maropitant. Vomiting is a potentially catastrophic complication in a dog that can’t guard its airway, and TXA should not be administered to obtunded dogs that are not pretreated with an antiemetic. Excessive clot formation is another potential complication of TXA administration that has not been adequately studied in dogs. Similar to previous observations, HF in the SAL group dogs largely resolved on its own with control of bleeding and volume expansion alone, suggesting that antifibrinolytic therapy is not necessary in dogs with low ATT scores and where source control is possible.11
Given the apparent safety of our TXA dosing protocol and its efficacy in reversing laboratory-measured HF, it is reasonable to consider its use in dogs with severe hemorrhagic shock (total ATT score > 5, and with scores of 2 to 3 for the perfusion and cardiac categories), particularly in dogs with ongoing hemorrhage that are treated in resource-limited or prehospital settings where blood-based resuscitation is not possible. Larger prospective studies, with more severely injured dogs, will be needed to properly assess the impact of TXA administration on outcome measures such as transfusion volumes required and/or mortality.
Acknowledgments
None reported.
Disclosures
The authors have nothing to disclose. No AI-assisted technologies were used in the generation of this manuscript.
Funding
This work is supported by the US Army Medical Research and Materiel Command under Contract No. W81XWH-17-C-0054. The views, opinions and/or findings contained in this report are those of the authors and should not be construed as an official Department of the Army position, policy, or decision unless so designated by other documentation. In conducting research using animals, the investigators adhered to the Animal Welfare Act Regulations and other Federal statutes relating to animals and experiments involving animals and the principles set forth in the current version of the Guide for Care and Use of Laboratory Animals, National Research Council.
References
- 1.↑
Brohi K, Cohen MJ, Davenport RA. Acute coagulopathy of trauma: mechanism, identification and effect. Curr Opin Crit Care. 2007;13(6):680-685. doi:10.1097/MCC.0b013e3282f1e78f
- 2.↑
MacLeod JB, Lynn M, McKenney MG, Cohn SM, Murtha M. Early coagulopathy predicts mortality in trauma. J Trauma. 2003;55(1):39-44. doi:10.1097/01.TA.0000075338.21177.EF
- 3.↑
Abelson AL, O’Toole TE, Johnston A, Respess M, de Laforcade AM. Hypoperfusion and acute traumatic coagulopathy in severely traumatized canine patients. J Vet Emerg Crit Care (San Antonio). 2013;23(4):395-401. doi:10.1111/vec.12073
- 4.
Gottlieb DL, Prittie J, Buriko Y, Lamb KE. Evaluation of acute traumatic coagulopathy in dogs and cats following blunt force trauma. J Vet Emerg Crit Care (San Antonio). 2017;27(1):35-43. doi:10.1111/vec.12525
- 5.
Herrero Y, Jud Schefer R, Muri BM, Sigrist NE. Prevalence of acute traumatic coagulopathy in acutely traumatized dogs and association with clinical and laboratory parameters at presentation. Vet Comp Orthop Traumatol. 2021;34(3):214-222. doi:10.1055/s-0040-1721707
- 6.↑
Holowaychuk MK, Hanel RM, Wood RD, Rogers L, O’Keefe K, Monteith G. Prospective multicenter evaluation of coagulation abnormalities in dogs following severe acute trauma. J Vet Emerg Crit Care (San Antonio). 2014;24(1):93-104. doi:10.1111/vec.12141
- 7.
Mischke R. Alterations of the global haemostatic function test ‘resonance thrombography’ in spontaneously traumatised dogs. Pathophysiol Haemost Thromb. 2003;33(4):214-220. doi:10.1159/000081511
- 8.
Mischke R. Acute haemostatic changes in accidentally traumatised dogs. Vet J. 2005;169(1):60-64. doi:10.1016/j.tvjl.2004.01.008
- 9.↑
Langhorn R, Poulsen MK, Heidemann PL, et al. Diagnosis of primary hyperfibrinolysis and in vitro investigation of the inhibitory effects of tranexamic acid in a group of dogs with sarcomas—a pilot study. Res Vet Sci. 2021;136:472-477. doi:10.1016/j.rvsc.2021.03.028
- 10.↑
Fletcher DJ, Rozanski EA, Brainard BM, de Laforcade AM, Brooks MB. Assessment of the relationships among coagulopathy, hyperfibrinolysis, plasma lactate, and protein C in dogs with spontaneous hemoperitoneum. J Vet Emerg Crit Care (San Antonio). 2016;26(1):41-51. doi:10.1111/vec.12346
- 11.↑
DeStefano IM, Rozanski EA, Koethe B, de Laforcade AM. Rapid resolution of hyperfibrinolysis in dogs with spontaneous hemoperitoneum following surgical control of hemorrhage. J Vet Emerg Crit Care (San Antonio). 2023;33(3):305-314. doi:10.1111/vec.13300
- 12.↑
Herrero Y, Schefer RJ, Muri BM, Sigrist NE. Serial evaluation of haemostasis following acute trauma using rotational thromboelastometry in dogs. Vet Comp Orthop Traumatol. 2021;34(3):206-213. doi:10.1055/s-0040-1719167
- 13.↑
Chapman MP, Moore EE, Ramos CR, et al. Fibrinolysis greater than 3% is the critical value for initiation of antifibrinolytic therapy. J Trauma Acute Care Surg. 2013;75(6):961-967. doi:10.1097/TA.0b013e3182aa9c9f
- 14.↑
Cotton BA, Harvin JA, Kostousouv V, et al. Hyperfibrinolysis at admission is an uncommon but highly lethal event associated with shock and prehospital fluid administration. J Trauma Acute Care Surg. 2012;73(2):365-370. doi:10.1097/TA.0b013e31825c1234
- 15.↑
Schöchl H, Frietsch T, Pavelka M, Jámbor C. Hyperfibrinolysis after major trauma: differential diagnosis of lysis patterns and prognostic value of thrombelastometry. J Trauma. 2009;67(1):125-131. doi:10.1097/TA.0b013e31818b2483
- 16.↑
Chapman MP, Moore EE, Moore HB, et al. Overwhelming tPA release, not PAI-1 degradation, is responsible for hyperfibrinolysis in severely injured trauma patients. J Trauma Acute Care Surg. 2016;80(1):16-23. doi:10.1097/TA.0000000000000885
- 17.↑
Ostrowski SR, Henriksen HH, Stensballe J, et al. Sympathoadrenal activation and endotheliopathy are drivers of hypocoagulability and hyperfibrinolysis in trauma: a prospective observational study of 404 severely injured patients. J Trauma Acute Care Surg. 2017;82(2):293-301. doi:10.1097/TA.0000000000001304
- 18.↑
Moore HB, Moore EE. Fibrinolysis dysregulation following trauma. In: Moore HB, Neal MD, Moore EE, eds. Trauma Induced Coagulopathy. Springer International Publishing; 2021:157-174. doi:10.1007/978-3-030-53606-0_11
- 19.↑
Johansson PI, Sørensen AM, Perner A, et al. Disseminated intravascular coagulation or acute coagulopathy of trauma shock early after trauma? An observational study. Crit Care. 2011;15(6):R272. doi:10.1186/cc10553
- 20.↑
Hayakawa M, Sawamura A, Gando S, et al. Disseminated intravascular coagulation at an early phase of trauma is associated with consumption coagulopathy and excessive fibrinolysis both by plasmin and neutrophil elastase. Surgery. 2011;149(2):221-230. doi:10.1016/j.surg.2010.06.010
- 21.↑
CRASH-3 trial collaborators. Effects of tranexamic acid on death, disability, vascular occlusive events and other morbidities in patients with acute traumatic brain injury (CRASH-3): a randomised, placebo-controlled trial. Lancet. 2019;394(10210):1713-1723. doi:10.1016/S0140-6736(19)32233-0
- 22.
Morrison JJ, Dubose JJ, Rasmussen TE, Midwinter MJ. Military application of tranexamic acid in trauma emergency resuscitation (MATTERs) study. Arch Surg. 2012;147(2):113-119. doi:10.1001/archsurg.2011.287
- 23.↑
Shakur H, Roberts I, Bautista R, et al.; CRASH-2 trial collaborators. Effects of tranexamic acid on death, vascular occlusive events, and blood transfusion in trauma patients with significant haemorrhage (CRASH-2): a randomised, placebo-controlled trial. Lancet. 2010;376(9734):23-32. doi:10.1016/S0140-6736(10)60835-5
- 24.↑
Web-Annex A. World Health Organization Model List of Essential Medicines - 23rd List, 2023. In: The Selection and Use of Essential Medicines 2023: Executive Summary of the Report of the 24th WHO Expert Committee on the Selection and Use of Essential Medicines, 24-28 April 2023. World Health Organization; 2023.
- 25.↑
Baker JL, Havas KA, Miller LA, Lacy WA, Schlanser J. Gunshot wounds in military working dogs in Operation Enduring Freedom and Operation Iraqi Freedom: 29 cases (2003-2009). J Vet Emerg Crit Care (San Antonio). 2013;23(1):47-52. doi:10.1111/j.1476-4431.2012.00823.x
- 26.↑
Edwards TH, Scott LLF, Gonyeau KE, Howard EH, Parker JS, Hall K. Comparison of trauma sustained by civilian dogs and deployed military working dogs. J Vet Emerg Crit Care (San Antonio). 2021;31(4):498-507. doi:10.1111/vec.13064
- 27.↑
Davros AM, Gregory CW, Cockrell DM, Hall KE. Comparison of clinical outcomes in cases of blunt, penetrating, and combination trauma in dogs: a VetCOT registry study. J Vet Emerg Crit Care (San Antonio). 2023;33(1):74-80. doi:10.1111/vec.13253
- 28.
Hall KE, Holowaychuk MK, Sharp CR, Reineke E. Multicenter prospective evaluation of dogs with trauma. J Am Vet Med Assoc. 2014;244(3):300-308. doi:10.2460/javma.244.3.300
- 29.↑
Kolata RJ, Kraut NH, Johnston DE. Patterns of trauma in urban dogs and cats: a study of 1,000 cases. J Am Vet Med Assoc. 1974;164(5):499-502.
- 30.↑
Fontes GS, McCarthy RJ, Kutzler MA, Zitek-Morrison E. The effects of sex and neuter status on trauma survival in dogs: a Veterinary Committee on Trauma registry study. J Vet Emerg Crit Care (San Antonio). 2022;32(6):756-763. doi:10.1111/vec.13210
- 31.↑
Ash K, Hayes GM, Goggs R, Sumner JP. Performance evaluation and validation of the animal trauma triage score and modified Glasgow Coma Scale with suggested category adjustment in dogs: a VetCOT registry study. J Vet Emerg Crit Care (San Antonio). 2018;28(3):192-200. doi:10.1111/vec.12717
- 32.↑
Kakiuchi H, Kawarai-Shimamura A, Fujii Y, et al. Efficacy and safety of tranexamic acid as an emetic in dogs. Am J Vet Res. 2014;75(12):1099-1103. doi:10.2460/ajvr.75.12.1099
- 33.↑
Osekavage KE, Brainard BM, Lane SL, Almoslem M, Arnold RD, Koenig A. Pharmacokinetics of tranexamic acid in healthy dogs and assessment of its antifibrinolytic properties in canine blood. Am J Vet Res. 2018;79(10):1057-1063. doi:10.2460/ajvr.79.10.1057
- 34.↑
TCCC updates. Tactic combat casualty care guidelines for medical personnel: 3 June 2015. J Spec Oper Med. 2015;15(3):129-147. doi:10.55460/QQQM-C7QH
- 35.↑
Clinical and Laboratory Standards Institute. Defining, Establishing, and Verifying Reference Intervals in the Clinical Laboratory: Approved Guideline. 3rd ed. CLSI document EP-28 A3c ed. Wayne, PA: Clinical and Laboratory Standards Institute, 2008.
- 36.↑
Peterson KL, Hardy BT, Hall K. Assessment of shock index in healthy dogs and dogs in hemorrhagic shock. J Vet Emerg Crit Care (San Antonio). 2013;23(5):545-550. doi:10.1111/vec.12090
- 37.↑
R Development Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing; 2013.
- 38.↑
Cade JF, Robinson TF. Coagulation and fibrinolysis in the dog. Can J Comp Med. 1975;39(3):296-298.
- 39.
Lang IM, Marsh JJ, Konopka RG, et al. Factors contributing to increased vascular fibrinolytic activity in mongrel dogs. Circulation. 1993;87(6):1990-2000. doi:10.1161/01.CIR.87.6.1990
- 40.↑
Moser KM, Cantor JP, Olman M, et al. Chronic pulmonary thromboembolism in dogs treated with tranexamic acid. Circulation. 1991;83(4):1371-1379. doi:10.1161/01.CIR.83.4.1371
- 41.↑
Fletcher DJ, Blackstock KJ, Epstein K, Brainard BM. Evaluation of tranexamic acid and ε-aminocaproic acid concentrations required to inhibit fibrinolysis in plasma of dogs and humans. Am J Vet Res. 2014;75(8):731-738. doi:10.2460/ajvr.75.8.731
- 42.↑
Kelmer E, Marer K, Bruchim Y, et al. Retrospective evaluation of the safety and efficacy of tranexamic acid (Hexakapron) for the treatment of bleeding disorders in dogs. Isr J Vet Med. 2013;68(2):94-100.
- 43.
Kelmer E, Segev G, Papashvilli V, et al. Effects of intravenous administration of tranexamic acid on hematological, hemostatic, and thromboelastographic analytes in healthy adult dogs. J Vet Emerg Crit Care (San Antonio). 2015;25(4):495-501. doi:10.1111/vec.12319
- 44.↑
Marsh JJ, Konopka RG, Lang IM, et al. Suppression of thrombolysis in a canine model of pulmonary embolism. Circulation. 1994;90(6):3091-3097. doi:10.1161/01.CIR.90.6.3091
- 45.↑
Kelley M, Sinnott-Stutzman V, Whelan M. Retrospective analysis of the use of tranexamic acid in critically ill dogs and cats (2018-2019): 266 dogs and 28 cats. J Vet Emerg Crit Care (San Antonio). 2022;32(6):791-799. doi:10.1111/vec.13237
- 46.↑
Muri B, Schmierer P, Schwarz A, Sigrist N. Hyperfibrinolysis diagnosed with rotational thromboelastometry and treated with tranexamic acid in a dog with acute traumatic coagulopathy. Schweiz Arch Tierheilkd. 2018;160(4):227-233. doi:10.17236/sat00155
- 47.↑
Gayet-Ageron A, Prieto-Merino D, Ker K, et al.; Antifibrinolytic Trials Collaboration. Effect of treatment delay on the effectiveness and safety of antifibrinolytics in acute severe haemorrhage: a meta-analysis of individual patient-level data from 40 138 bleeding patients. Lancet. 2018;391(10116):125-132. doi:10.1016/S0140-6736(17)32455-8
- 48.↑
Chik C, Hayes GM, Menard J. Development of a veterinary trauma score (VetCOT) in canine trauma patients with performance evaluation and comparison to the animal trauma triage score: a VetCOT registry study. J Vet Emerg Crit Care (San Antonio). 2021;31(6):708-717. doi:10.1111/vec.13135
- 49.↑
Moore HB, Moore EE, Morton AP, et al. Shock-induced systemic hyperfibrinolysis is attenuated by plasma-first resuscitation. J Trauma Acute Care Surg. 2015;79(6):897-903. doi:10.1097/TA.0000000000000792
- 50.↑
Kostousov V, Wang YW, Cotton BA, Wade CE, Holcomb JB, Matijevic N. Influence of resuscitation fluids, fresh frozen plasma and antifibrinolytics on fibrinolysis in a thrombelastography-based, in-vitro, whole-blood model. Blood Coagul Fibrinolysis. 2013;24(5):489-497. doi:10.1097/MBC.0b013e32835e4246
