Page 83 - JSOM Winter 2019

reflect the complex changes in perfusion and hemodynamics a syringe pump. In the IO group, TXA was injected as a bo-
associated with hemorrhagic shock. To clarify these effects on lus through the previously placed IO needle. In the IM group,
drug absorption, we conducted a PK analysis of TXA adminis- TXA was administered with a 22-gauge needle as a bolus in
tered via these routes in a swine model of hemorrhagic shock. the semitendonosis/semimembranosis muscle complex, 4cm
We hypothesized that IO administration would be comparable distal to the ischial tuberosity on the contralateral side of the
to the IV route and that IM dosing would demonstrate de- arterial line. This site was chosen because it is the largest mus-
creased bioavailability due to blood shunting away from skel- cle complex and it has well-defined anatomic landmarks, thus
etal muscle during shock. allowing consistency across animals.

Methods Data Collection
Blood samples were collected before both hemorrhage and
This study was approved by the Institutional Animal Care and TXA injection for arterial blood gas, lactate, and hematocrit
Use Committee at David Grant USAF Medical Center, Travis analysis. Whole blood was collected in serum separator tubes,
Air Force Base, CA. All animal care and use were in compli- allowed to clot for 15 minutes, and centrifuged at 3,000g for
ance with the Guide for the Care and Use of Laboratory Ani- 10 minutes. Blood samples were collected at 5, 10, 15, 30,
mals in a facility accredited by the Association for Assessment 60, 120, 240, and 300 minutes after TXA injection and spun
and Accreditation of Laboratory Animal Care International. in a centrifuge, and serum aliquots were stored at –80°C for
later determination of TXA concentrations by liquid chroma-
Experimental Preparation tography–mass spectrometry (LC-MS). Heart rate, MAP, Spo ,
2
Fifteen castrated male Yorkshire-cross pigs (Sus scrofa; S & S and peripheral perfusion were recorded at each of these time
Farms, Ramona, CA) weighing 67.7 ± 1.3 kg were acclimated points as well.
for at least 10 days in conventional housing. After an 8- to
12-hour fast with free access to water, they were anesthetized Recovery
with an IM injection of 6.6mg/kg tiletamine/zolazepam (TEL- Five hours after TXA administration, all shed blood was
AZOL; Fort Dodge Animal Health, New York, NY) followed returned along with 4.83g of calcium gluconate IV over 10
by isoflurane mask induction. After endotracheal intubation, minutes. Animals then recovered from anesthesia and were
animals were maintained under anesthesia with isoflurane monitored for adverse events and skin changes for at least
mixed in 100% oxygen. Mechanical ventilation was regulated 7 days. To reduce animal use, animals were monitored for
to maintain end-tidal CO between 35 and 45mmHg. Body changes until they were used for other purposes such as model
2
temperature was maintained between 35ºC and 37ºC using development and educational activities. Care was taken not to
warming blankets. A 16-gauge precaval central venous line inflict tissue damage to the injection sites. After euthanasia,
was placed percutaneously to administer intravenous 0.9% animals underwent necropsy. Specifically, the injection site and
saline (5mL/kg/h). A 22-gauge peripheral intravenous (PIV) adjacent tissues were sectioned with either a knife or a saw
catheter was placed in an ear vein. A peripheral capillary ox- and visually inspected for gross evidence of necrosis.
ygen saturation (Spo ) monitor was attached to the opposite
2
ear. An ultrasound-guided 7Fr femoral arterial line was placed LC-MS Specimen Processing
for controlled hemorrhage and mean arterial pressure (MAP) Briefly, the banked serum samples were prepared by thawing
monitoring. A laser Doppler peripheral tissue perfusion sensor and mixing 200μL of sample with 200μL of 2.5% PCA and
(Model BLF21DC; Transonic Systems Inc, Ithaca, NY) was 3μL of 0.5 μg/mL CIS-4 internal standard. Samples were vor-
placed on the limb opposite the arterial line at the level of the texed and centrifuged for 10 minutes at 14,000 rpm. Then,
proximal tibia. 100μL of supernatant was added to labeled LC-MS sample
vials containing 150μL of 0.1mol/L NaOH and mixed. Cal-
Hemorrhagic Shock ibration standards and controls were prepared by spiking
Thirty-five percent of the blood volume (calculated as weight blank pig serum with known amounts of TXA. Next, 0.04μL
[kg] × 66 [mL/kg] × 0.35] was removed from the femoral ar- was injected on the LC-MS system consisting of an Agilent
tery catheter over 15 minutes. Blood was collected in citrate Technologies 1290 Infinity HPLC coupled to an Agilent Tech-
phosphate dextrose blood collection bags (Fenwal Inc., Lake nologies 6550 ifunnel Q-TOF mass spectrometer (Santa Clara,
Zurich, IL) and placed in a 38°C water bath for later resuscita- CA). The mobile phase consisted of 99% of 0.1% formic acid
tion. Animals remained in hemorrhagic shock for 15 minutes in water (solvent A) and 1% methanol (solvent B). Separation
to simulate the time between injury and treatment by a mili- was achieved using a ZORBAX Eclipse Plus C18 Rapid Reso-
tary medic. During this equilibration phase, all animals were lution HD (Agilent Technologies) column (2.1mm × 50mm ×
randomized to one of three groups; IV, IO, and IM (n = 5 per 1.8μm) maintained at 28ºC. The mass spectrometer was oper-
group). If the animal was randomized to receive IO injection ated in the positive ion mode and monitored for m/z 158.12
of TXA, a 25mm IO needle was inserted with a power drill for TXA and 144.10 for CIS-4.
(Arrow EZ-IO ; Teleflex Incorporated, Wayne, PA) in the
®
®
proximomedial tibia, approximately 1cm distal to the tibial Data Analysis
tuberosity on the contralateral side to the arterial line. The number of animals per arm was determined using the
resource equation (E = total number of animals–total num-
Intervention ber of groups). In our experiment, E = 15–3 = 12, which lies
After 15 minutes of hemorrhagic shock (T0), TXA was admin- between 10 and 20, indicating an adequate sample size for
istered to all animals. In all three groups, 1g of TXA (X-Gen animal research. The primary outcome measured was serum
Pharmaceuticals, Inc., Horseheads, NY) was diluted in 10mL TXA concentration over time. This was used to determine PK
of 0.9% sodium chloride for injection. In the IV group, TXA parameters using the STATA (Stata version 15, Stata Corp,
®
was infused through the ear vein PIV over 10 minutes using College Station, TX) suite of PK commands. Baseline vital
TXA via IV, IO, and IM Routes in a Porcine Model | 81