Evaluation of lyophilized platelets as an infusible hemostatic agent in experimental non-compressible hemorrhage in swine



    1. Combat Casualty Care, Regenerative Medicine Department, Naval Medical Research Center, Silver Spring, MD
    2. Department of Surgery, Walter Reed Army Medical Center, Washington, DC
    Search for more papers by this author
  • E. A. ELSTER,

    1. Combat Casualty Care, Regenerative Medicine Department, Naval Medical Research Center, Silver Spring, MD
    2. Department of Surgery, National Naval Medical Center, Bethesda, MD
    3. Uniformed Services University of Health Sciences, Bethesda, MD
    Search for more papers by this author
  • D. FRYER,

    1. Combat Casualty Care, Regenerative Medicine Department, Naval Medical Research Center, Silver Spring, MD
    Search for more papers by this author

    1. Department of Surgery, National Naval Medical Center, Bethesda, MD
    2. Uniformed Services University of Health Sciences, Bethesda, MD
    Search for more papers by this author

    1. Department of Comparative Pathology, Walter Reed Army Institute of Research, Silver Spring, MD, USA
    Search for more papers by this author

    1. Uniformed Services University of Health Sciences, Bethesda, MD
    Search for more papers by this author
  • T. TOMORI,

    1. Combat Casualty Care, Regenerative Medicine Department, Naval Medical Research Center, Silver Spring, MD
    2. Department of Neurosurgery, Tokyo Medical and Dental University, Tokyo, Japan
    Search for more papers by this author
  • T. S. BROWN,

    1. Combat Casualty Care, Regenerative Medicine Department, Naval Medical Research Center, Silver Spring, MD
    Search for more papers by this author
  • D. K. TADAKI

    1. Combat Casualty Care, Regenerative Medicine Department, Naval Medical Research Center, Silver Spring, MD
    2. Uniformed Services University of Health Sciences, Bethesda, MD
    Search for more papers by this author

Douglas Tadaki, Regenerative Medicine Department, Combat Casualty Care, Naval Medical Research Center, Silver Spring, MD 20910, USA.
Tel.: +1 301 319 3223; fax: +1 301 319 7210.
E-mail: doug.tadaki@med.navy.mil


Summary. Introduction: Human lyophilized platelets hold promise as a novel hemostatic infusion agent for the control of traumatic hemorrhage. Rehydrated, lyophilized platelets (Stasix) were investigated as an infusible hemostatic agent in experimental non-compressible hemorrhage, using a porcine liver injury model. Methods: Yorkshire swine underwent a grade III liver injury and uncontrolled bleeding. After 15 min, animals were infused with Stasix (n = 10) or normal saline vehicle (n = 10). At 2 h, the liver was repaired, and the animals were monitored for another4 h. Resuscitation, including blood transfusion, was administered during the hospital phase. Laboratory data, including arterial blood gas, complete blood count, thromboelastography (TEG), and coagulation parameters, were collected. All animals underwent necropsy with complete histopathologic examination. Results: Overall survival in the Stasix group [8/10 (80%)] was significantly higher than in the control group [2/10 (20%)] (P = 0.023). Mean total blood loss index (g kg−1) was lower in Stasix-treated animals (22.2 ± 3.5) than in control animals (34.7 ± 3.4) (P = 0.019). Hemodynamic parameters were improved in the Stasix group, and a trend towards higher hemoglobin and lower lactate was observed. Coagulation and TEG parameters were not different between the groups. One surviving animal in the Stasix group had evidence of thrombi on necropsy. Conclusions: This is the first reported study to evaluate rehydrated, lyophilized platelets as an infusible hemostatic agent for non-compressible hemorrhage. Stasix improved survival and reduced blood loss in a liver injury porcine model. However, evidence of thrombotic complications warrants further investigation prior to human use in the setting of traumatic hemorrhage.


Despite significant advances in trauma resuscitation and casualty evacuation systems, hemorrhage remains the most common cause of preventable death after injury, in both civilian and military populations [1–3]. Examination of urban trauma mortality revealed that 50–70% of deaths occur in the prehospital setting or early after arrival at the trauma center, and hemorrhage was responsible for 50% of these deaths [1]. In military trauma, recent analyses of fatalities sustained in Operation Iraqi Freedom and Operation Enduring Freedom identified hemorrhage as the leading cause of potentially survivable deaths [4,5]. Notably, non-compressible thoracoabdominal hemorrhage accounted for approximately one-half of these deaths. The development of an immediately available advanced hemostatic agent targeting internal hemorrhage would provide a significant decrease in early battlefield mortality and be widely applicable in civilian trauma.

The application of an effective infusible hemostatic agent for traumatic hemorrhage has obvious advantages, particularly in the care of combat casualties. Military operations often take place in austere environments, which may result in significant delay before hospital care can be provided. In addition, logistical constraints limit the use of conventional component blood therapy. Accordingly, one of the primary goals of combat casualty research is the development of infusible hemostatics to reduce bleeding in the prehospital setting.

The therapeutic potential of infusible hemostatics to reduce or stop bleeding related to injury has received considerable attention recently, in light of the increasing use of recombinant factor VIIa (rFVIIa) for the treatment of hemorrhage. However, although it is efficacious for the treatment of hemophilic patients with inhibitors to FVIII or FIX, the use of rFVIIa in the setting of trauma remains equivocal [6,7]. The use of antifibrinolytic drugs has also been extensively investigated for the treatment of surgical hemorrhage [8]; however, at present, there is insufficient evidence to support the use of these agents following traumatic injury [9].

Lyophilized, or freeze-dried, human platelets have been under development for nearly 50 years [10,11]. With use of a mild aldehyde stabilization technique, lyophilized platelet products have been engineered with preserved morphology and functional properties [12,13]. This method is based on the covalent cross-linking of surface membrane proteins and lipids to stabilize platelets for freezing, lyophilization, and rehydration. In vitro and in vivo studies have demonstrated that rehydrated, lyophilized platelets retain a near normal ultrastructure by electron microscopy, adhere to denuded subendothelium via glycoprotein (GP)Ib–von Willebrand factor (VWF) interactions, are capable of intracellular stimulus–response signaling to increase surface membrane thrombogenicity, and degranulate in an activation-dependent manner [12,14–16]. In addition, lyophilized platelets acquire a partially activated state as a result of the preparation process, thereby enhancing their immediate hemostatic properties upon infusion [16].

Lyophilized platelets retain morphologic and functional characteristics, and offer significant logistic and therapeutic advantages as a hemostatic agent for the control of traumatic hemorrhage. The current study was designed to evaluate rehydrated, lyophilized platelets (Stasix) as an infusible hemostatic agent for non-compressible hemorrhage in a liver injury porcine model.

Materials and methods

The experiments reported herein were conducted according to the principles set forth in the Guide for the Care and Use of Laboratory Animals, Institute of Laboratory Animals Resources, National Research Council, National Academy Press, 1996. The study was approved by the National Medical Research Center Institutional Animal Care and Use Committee (IACUC), and all procedures were performed in animal facilities approved by the Association for Assessment and Accreditation for Laboratory Animal Care International (AAALAC).

Rehydrated, lyophilized platelet preparation

Lyophilized human platelets (Stasix) were provided by Entegrion (Research Triangle Park, NC, USA). Stasix was received as a lyophilized quantity of platelets (6.25 × 1010 cells) in a single vial. Each vial was divided into equal aliquots on the basis of weight, for storage at − 80 °C in order to prevent repeat freeze–thaw damage. Immediately prior to use, the platelets were reconstituted in a 50-mL tube with normal saline (NS).

Animal preparation

Male and female Yorkshire swine (Sus scrofa domestica) were used. Feed was withheld 12 h before surgery. Animals were sedated and anesthesia was induced with intramuscular ketamine hydrochloride (33 mg kg−1) and mask ventilation with isoflurane (5.0%), to facilitate endotracheal intubation. Pigs were ventilated (Ohmeda 7800 series ventilator; Datex, Madison, WI, USA) at 12–15 breaths per min and a tidal volume of 10 mL kg−1. Anesthesia was maintained with isoflurane (1.5–2.5%) in 21–25% O2.

Following adequate anesthesia, the right external jugular vein and carotid artery were dissected and isolated. A pulmonary artery catheter (Edwards Life Sciences, Irvine, CA, USA) was inserted in the external jugular vein for continuous hemodynamic monitoring. An angiocath was placed in the carotid artery, and mean arterial pressure was continuously transduced. A midline laparotomy was performed to expose the liver and isolate the left lateral lobe. Rectal temperature was monitored continuously. Animals were not actively warmed during the prehospital phase, to replicate combat casualty evacuation (Fig. 1).

Figure 1.

 Experimental design. Liver injury was initiated at 0 min. Stasix or normal saline (NS) vehicle was infused at 15 min. Fluid resuscitation (NS) was initiated at 30 min. Uncontrolled hemorrhage occurred until 60 min, at which time the abdomen was packed and temporarily closed. The animal was observed until 120 min, at which time hospital care was initiated. The liver was repaired and the abdomen was definitively closed. Blood transfusion was administered as indicated. The pig was actively warmed. The animal was observed for a total of 360 min. Necropsy was performed when the animal died or at 360 min following euthanasia. Standard invasive monitoring was performed throughout the experiment.

Liver injury and resuscitation

A standardized liver injury was created by placing a ring clamp over the left lower lobe, 50% in width and 3.5–5 mm from the apex, adjusting for the relative size of the animal [17]. The clamp was closed and a #10 scalpel blade was used to lacerate the lobe from the top of the clamp through the remaining width. The liver injury denoted the start of the prehospital phase (time 0). After 1 min, the clamp was removed and the remaining tissue was excised, resulting in a 25% lobectomy, consistent with a grade III liver injury. Bleeding was spontaneous, and blood was removed continually via suction and quantified by weight. Blood loss weight (g) was divided by animal weight (kg) and reported as the blood loss index (g kg−1). After 60 min, blood collection was discontinued, and the abdomen was packed and then closed with towel clips.

At 15 min of the prehospital phase, animals were randomly allocated to treatment with Stasix or NS vehicle. Stasix was infused in 30 mL of NS over 15 min with a syringe pump (BS-8000; Braintree Scientific, Inc., Braintree, MA, USA). In the control group, a 30-mL NS vehicle was infused at an identical rate. After 30 min, animals in both groups were infused with 1 L of NS over the next 90 min. All fluids were administered at room temperature.

After a total of 120 min, the prehospital phase was completed and hospital arrival was simulated. The abdomen was reopened, residual blood was suctioned, sponges were collected, and blood loss was quantified by weight. The liver injury was repaired with sutures, and the abdomen was definitively closed. One unit of allogeneic whole blood (10 mL kg−1) (Thomas Morris, Reisterstown, MD, USA) was administered for hemoglobin < 7 g dL−1; otherwise, animals received additional NS at 10 mL kg−1. Animals were actively rewarmed with a warming device during the hospital phase to normothermia (37 °C). Monitoring was continued for an additional 240 min, at which time the surviving animals were killed. The total experiment lasted for 360 min for surviving animals.

Laboratory data

All functional laboratory assays were performed at 37 °C, consistent with recorded normothermic animal temperatures (37.5 ± 0.9 °C). Complete blood count (CBC) and arterial blood gas (ABG) blood samples were collected at 0, 15, 30, 60, 75, 90, 120, 180, 240, 300 and 360 min. CBC with differentiation was performed with a cell counter (NDvia120 Hematology System; Siemens, Deerfield, IL, USA). ABG was measured with a blood gas instrument (ABL 750; Radiometer, Copenhagen, Denmark).

Thromboelastography (TEG) and coagulation parameters were collected at 0, 30 and 360 min or at the time of animal death. The TEG reaction time (TEG-R) corresponding to fibrin formation, kinetics of clot formation, ie. rate of fibrin deposition (TEG-K and TEG-α), maximum amplitude, fibrinolysis and coagulation index were measured with a hemostasis analyzer (TEG 5000; Haemoscope Corp., Niles, IL, USA). The test was initiated with 340 μL of whole blood recalcified with 20 μL of CaCl2. Coagulation parameters, including prothrombin time, partial thromboplastin time (PTT), fibrinogen, thrombin–antithrombin, and antithrombin-III, were measured with a coagulation workstation (STA Compact, Diagnostica Stago, Parsippany, NJ, USA). Phlebotomy blood volumes (totaling 175 mL for animals surviving 360 min) were not included in reported hemorrhage volumes.

Pathologic examination

Full necropsy for pathologic examination was immediately performed on all animals at time of death or euthanasia by a dedicated veterinarian pathologist. Tissue sections of heart (transversely through papillary muscles), lung (left caudal lobe), liver (at site of injury and distal to injury), kidney and small bowel were obtained for histological analysis. Tissue specimens were fixed by immersion in 10% neutral buffered formalin. Tissues were embedded in paraffin wax, sectioned at 5-μm thickness onto glass slides, and stained using hematoxylin and eosin.

Phosphotungstic acid hematoxylin (PTAH) staining was performed for fibrin deposition. Lung sections were deparaffinized, and fixed with Zenker’s solution (mercuric chloride, potassium dichromate, and sodium sulfate). The sections were then treated with Lugol’s iodine and stained in PTAH overnight at room temperature.

Statistical analysis

Animals were randomly assigned to treatment with Stasix or NS vehicle. Survival analysis between groups was performed with the Fisher exact test and by Kaplan–Meier survival plots with the log rank test. Blood loss and laboratory parameters were compared with the Student t-test or non-parametric tests as appropriate. One-way anova with a repeated measures design was used for continuous time-dependent comparisons. Statistical analysis was performed using spss (SPSS Inc., Chicago, IL, USA). A two-tailed P-value < 0.05 was considered to be statistically significant. All data are represented as means ± standard error of the mean.


Stasix dose titration

In order to determine the appropriate quantity of infusible Stasix, we performed the following dose titration experiment. Initially, Stasix was infused at a full dose of 6.25 × 1010 cells (n = 4), which resulted in rapid hemostasis at the site of injury. However, this was followed by pathologic clot propagation, resulting in acute hepatic outflow obstruction and death at 30 ± 4 min following injury in all animals. Next, Stasix was diluted 30-fold, or to 2.08 × 109 cells (n = 3), with a moderate decrease in blood loss (data not shown) when compared to controls, although the difference was not statistically significant (P = 0.38). Finally, Stasix was diluted 15-fold, or to 4.17 × 109 cells, for the remaining experiments (n = 10). All subsequent data analysis includes only animals given Stasix at this dose. Notably, the amount of Stasix infused was approximately 1–2% of the total circulating platelets, and there was no difference in platelet number after Stasix infusion (data not shown).

Baseline characteristics

There were no significant differences in baseline characteristics between Stasix and control animals, including weight (28.0 ± 0.8 kg vs. 25.6 ± 0.8 kg), hepatectomy weight index (0.49 ± 0.05 g kg−1 vs. 0.53 ± 0.8 g kg−1), baseline pH (7.44 ± 0.04 vs. 7.44 ± 0.02), and hemoglobin (8.8 ± 0.3 g dL−1 vs. 8.7 ± 0.3 g dL−1).

Survival analysis

Overall survival in the Stasix group [8/10 (80%)] was significantly higher than in the control group [2/10 (20%)], as demonstrated by the Kaplan–Meier survival plot (P = 0.007) (Fig. 2).

Figure 2.

 Kaplan–Meier survival curve for Stasix vs. control experiments with log-rank test for statistical comparison.

Blood loss

During the uncontrolled blood loss period, the Stasix-treated animals demonstrated significantly decreased blood loss (P = 0.001) (Fig. 3A). Notably, the blood loss rate decreased immediately during Stasix infusion at 15 min. The mean total blood loss index (g kg−1) was significantly lower in the Stasix-treated animals (22.2 ± 3.5) than in the control animals (34.7 ± 3.4) (P = 0.019) (Fig. 3B). On inspection of the injured liver edge at the hospital phase (2 h), there was complete hemostasis in 6/10 (60%) of the Stasix-treated animals as compared with 0/10 of the control animals (P = 0.011) (Fig. 3C). Additionally, in 2/10 (20%) of the Stasix-treated animals, complete hemostasis was achieved within the uncontrolled blood loss period (data not shown).

Figure 3.

 (A) Blood loss index (g kg−1) during uncontrolled hemorrhage period in Stasix vs. control experiments. (B) Total blood loss index (g kg−1). (C) Hemostasis (%) in hospital phase period. NS, normal saline.

Hemodynamic parameters

The heart rate was significantly lower (P < 0.001) and the mean arterial pressure was higher (P < 0.001) in the Stasix group during the prehospital phase (Fig. 4). During the hospital phase, the heart rate remained lower in the Stasix group (P = 0.001), whereas blood pressure was not significantly different between the groups (P = 0.46). The mean pulmonary artery pressure was not significantly different during the prehospital phase (P = 0.26) and, inexplicably, was higher in the Stasix group during the hospital phase (P < 0.001).

Figure 4.

 Hemodynamic changes with Stasix vs. control experiments. (A) Heart rate. (B) Mean arterial pressure. (C) Mean pulmonary artery pressure.

Laboratory data

The mean hemoglobin level was generally higher in the Stasix group during the prehospital phase, although the difference was not statistically significant (Table 1). There was no change in platelet numbers following the administration of the Stasix as compared with the control [171 ± 31.7 vs. 188 ± 31.8 (× 103 mL−1)]. In addition, transfusion requirements in the hospital phase were not different between the Stasix group [6/10 (60%)] and the control group [6/10 (60%] (P-value not significant). The mean serum lactate level was significantly lower in the Stasix group at 60 min (P < 0.05) (Fig. 5). After 120 min (hospital phase), the mean lactate level normalized in the surviving control animals. The mean serum base excess was not significantly different between the two groups (data not shown).

Table 1.   Hemoglobin, mean ± standard error of the mean (g dL−1)
TreatmentT = 0 minT = 15 minT = 30 minT = 60 minT = 120 minT = 240 minT = 360 min
  1. No statistical differences observed.

Control (n = 10)8.8 ± 0.38.2 ± 0.37.5 ± 0.36.7 ± 0.36.4 ± 0.3
(n = 7)
7.8 ± 0.5
(n = 3)
7.8 ± 1.0
(n = 2)
Stasix (n = 10)8.7 ± 0.38.4 ± 0.38.6 ± 0.67.1 ± 0.36.6 ± 0.4
(n = 9)
7.4 ± 0.3
(n = 9)
7.9 ± 0.3
(n = 8)
Figure 5.

 Mean serum lactate (mmol L−1). *P < 0.05.

There were no statistically significant differences in TEG and coagulation parameters between the Stasix-treated and control animals (Tables 2 and 3). However, in the Stasix group, the TEG-R and TEG-K values were significantly decreased from baseline at the final time point (P < 0.05). This represents a mixed hypercoagulable (decreased TEG-R) and hypocoagulable (decreased TEG-K) state in this group. In both the Stasix group and the control group, the PTT values were significantly decreased from baseline at the final time point (P < 0.05).

Table 2.   Thromboelastography
VariableTreatmentT = 0 minT = 30 minFinal
  1. *P < 0.05 from baseline (paired Student t-test). There were no statistical differences observed between the Stasix group and the control group (unpaired Student t-test).

Reaction time (s)Control460 ± 23373 ± 83296 ± 46
Stasix466 ± 23385 ± 76291 ± 27*
Rate, K (s)Control111 ± 8111 ± 29104 ± 29
Stasix108 ± 898 ± 1577 ± 8*
Angle (°)Control65.7 ± 1.566.2 ± 4.567.4 ± 5.2
Stasix66.2 ± 1.568.4 ± 2.672.4 ± 1.8
Maximum amplitude (mm)Control70.9 ± 1.171.6 ± 2.265.1 ± 5.8
Stasix71.1 ± 0.871.9 ± 1.371.5 ± 1.8
30 minute lysis, LY30 (%)Control2.2 ± 0.42.2 ± 0.71.7 ± 0.8
Stasix2.7 ± 0.32.6 ± 0.52.3 ± 0.4
Coagulation indexControl3.3 ± 0.23.8 ± 0.43.0 ± 0.8
Stasix3.3 ± 0.13.7 ± 0.43.9 ± 0.3
Table 3.   Coagulation parameters
VariableTreatmentT = 0 minT = 30 minFinal
  1. AT-III, antithrombin-III; PT, prothrombin time; PTT, partial thromboplastin time; TAT, thrombin–antithrombin. *P < 0.05 from baseline (paired Student t-test). There were no statistical differences observed between the Stasix group and the control group (unpaired Student t-test).

PT (s)Control14.4 ± 0.214.7 ± 0.315.7 ± 0.7
Stasix14.3 ± 0.214.3 ± 0.315 ± 0.8
PTT (s)Control32.6 ± 2.235.2 ± 2.826.9 ± 3.9*
Stasix32.9 ± 2.331.6 ± 4.526.2 ± 4.0*
Fibrinogen (mg dL−1)Control114 ± 4115 ± 14114 ± 29
Stasix118 ± 4137 ± 25118 ± 9
AT-III (mg mL−1)Control81 ± 376 ± 567 ± 7
Stasix83 ± 376 ± 575 ± 6
TAT (μg L−1)Control17.5 ± 0.616.8 ± 0.416.2 ± 0.8
Stasix17.8 ± 0.620.2 ± 3.517.4 ± 0.8

Histopathologic evaluation

On necropsy, one animal that received the therapeutic Stasix dose and survived the entire observation period was found to have an endocardial thrombus adherent to the right ventricular free wall. This was confirmed on histologic analysis, and additional pulmonary artery thrombi were visualized; these findings possibly represent a thromboembolic phenomenon (Fig. 6). No other animals had identifiable thrombi in the tissues examined.

Figure 6.

 Hematoxylin and eosin stains. (A) Right ventricular free wall with adherent thrombus (arrow) and adjacent cardiomyocyte degeneration and necrosis. (B) Pulmonary artery organized thrombus (arrow).

Phosphotungstic acid staining for fibrin deposition in the lungs revealed minimal to mild diffuse staining in the medium to large vessels. There was no evidence of microthrombi.


This is the first report of rehydrated, lyophilized platelets as an infusible hemostatic agent for experimental non-compressible hemorrhage in a large animal study. Stasix demonstrated efficacy, with decreased blood loss, improved hemodynamics, diminished lactic acid levels and markedly increased survival in a porcine liver injury model. However, significant thrombotic events were observed, particularly with undiluted Stasix doses, possibly related to xenographic interactions between human platelets and porcine coagulation factors.

Freeze-dried platelets were first developed for the purpose of in vitro diagnostic testing of VWF [18]. Long-term stabilization of platelets with paraformaldehyde maintained intact VWF receptors for years of frozen storage; however, the fixation process rendered the platelets inert to activation stimuli. The minimal stabilization conditions necessary to ensure structural and functional integrity after freeze-drying and rehydration of platelets have been defined [12]. With a refined stabilization technique, rehydrated, lyophilized platelets retain near-normal ultrastructure by electron microscopy, and maintain important functional characteristics.

The advantages of an infusible lyophilized platelet product are numerous. The storage life of fresh platelets is limited to 5 days; lyophilized platelet particles have a shelf-life of years, alleviating the logistic constraints of component blood therapy. Microbial contamination is also a concern with fresh platelet storage [19], but is eliminated during the freezing and lyophilization process, yielding a sterile transfusion product [20]. Furthermore, fresh platelets do not retain normal function on reinfusion, owing to a gradual impairment of platelet integrity during storage, a process termed the platelet storage lesion [21]. In contrast, rehydrated, lyophilized platelets retain functional properties that are critical for hemostasis [10].

To promote effective hemostasis, platelets must rapidly respond to changes in normal blood flow or vessel injury. Following vascular injury, platelets adhere to exposed subendothelium, aggregate, and form a primary platelet plug, and this is followed by fibrin formation. The initiation of a thrombus is mediated primarily through platelet membrane receptors that are essential for effective hemostasis [22,23]. Platelet adhesion to a damaged vessel wall and its extracellular matrix at high shear is mediated through the interactions between the GPIb–IX complex on the platelet membrane and endothelium-derived VWF [24]. These interactions are flow-dependent and overlapping, highlighting the complexity of hemostasis physiology [25–27]. Importantly, intact platelet receptors are fundamental for normal platelet function, and must be preserved during platelet storage to ensure hemostatic capacity.

Numerous in vivo and in vitro studies have investigated the functional properties of rehydrated, lyophilized platelets. Studies using flow cytometry with fluorescent monoclonal antibodies to GPIbα or the GPIb–IX complex demonstrated that the distribution of GPIb on rehydrated, freeze-dried platelets was nearly equivalent to that of fresh fixed platelets [12]. In a Baumgartner perfusion system experiment, rehydrated, lyophilized platelets were adherent to denuded arterial vessels under high shear, demonstrating intact GPIb integrin receptor function [15]. The platelets did not adhere to normal vessels, suggesting that the rehydrated platelets would form an intravascular plug only at sites of endothelial injury, rather than thrombose spontaneously in normal vasculature. Flow cytometry and binding studies have demonstrated that, unlike thatof GPIb, the distribution of GPIIb–IIIa receptors on rehydrated platelets is limited, indicating that the freeze-drying and rehydration process does not permit full reconstitution of these fibrinogen receptors. However, transmission electron micrographs of clots formed in recalcified plasma containing the rehydrated platelets revealed fibrin strands associated with the surface of the freeze-dried platelets, suggesting that some GPIIb–IIIa receptor function is preserved [20,28].

In addition to binding fibrin on their surface membranes, lyophilized platelets accelerate the conversion of prothrombin at a rate higher than that for fresh platelets, indicating some degree of procoagulant activation resulting from the preparation process [16]. Despite this evidence of a partially activated state, lyophilized platelets remain capable of stimulus–response activation, as their incorporation into fibrin gels results in an irreversible activation response involving shape change, pseudopodia extension, granule centralization and secretion, and cytoskeletal rearrangements [29]. These events result from preserved intracellular signaling pathways, including the protein kinase C and myosin light chain kinase systems [14]. Altogether, these investigational studies have exhaustively measured the functional properties of rehydrated, lyophilized platelets.

Although the morphologic and functional properties of rehydrated, lyophilized platelets have been well characterized by various in vitro and in vivo experiments, we sought to evaluate these platelets as a hemostatic agent because the ability to bind fibrin and to accelerate clot formation. The current model employs a standard, non-compressible intra-abdominal (liver) injury with a simulated prehospital phase prior to definitive care, and has been previously reported in large animal hemorrhage studies [17]. In this study, Stasix-treated animals demonstrated decreased blood loss immediately on Stasix infusion, resulting in significantly lower total blood loss than in control animals. On reaching the hospital phase, 60% of the liver injuries in the Stasix group were hemostatic, and, impressively, 20% achieved complete hemostasis during the uncontrolled blood loss period. Hemodynamics also improved and lactic acid levels decreased in the Stasix group. Thus, Stasix demonstrated efficacy under conditions specific to the current study design.

TEG and coagulation parameters did not differ between Stasix-treated and control animals, despite apparently improved hemostasis in the Stasix group. The measured TEG and coagulation parameters largely reflect secondary hemostatic mechanisms, whereas rehydrated, lyophilized platelets predominantly function via primary hemostatic adhesion mechanisms (GPIb–VWF interactions) [30]. Thus, Stasix treatment would not be expected to have a measurable effect on these secondary hemostatic metrics. At the final blood draw, the TEG-R and PTT values were shortened as compared with baseline in both Stasix-treated and control animals (although the TEG-R value did not reach statistical significance in the control group), representing a relatively hypercoagulable condition in both groups. This is consistent with prior reports of a hypercoagulable state manifesting as the result of hemorrhage in large animal studies [31] and in trauma patients early after hospital admission [32,33]. The TEG-K value was also decreased from baseline in the Stasix group, indicating some degree of mixed hypocoagulability, possibly related to the longer survival time in these animals.

An important concern in the development of infusion therapeutics for hemostasis is that the product is both effective at controlling hemorrhage and safe with respect to inducing pathologic intravascular coagulation. Although Stasix is extremely effective as a hemostatic agent, significant thrombotic complications were identified with the use of Stasix in this study. The therapeutic dose in the current study is 100 times lower than what would be used for the treatment coagulopathies resulting from hemorrhage in humans. The need for a lower dose of the lyophilized product may be due to the fact that fresh platelets are mostly unactivated and naïve, whereas the lyophilized products are partially activated, and thus will home more efficiently to the site of hemorrhage. Another possibility is a xenographic response that heightens the activity of the lyophilized product. At a full dose of Stasix, or 6.25 × 1010 platelets, all animals suffered profound thrombotic complications resulting in rapid death. At what we determined to be a therapeutic dose, or 4.17 × 109 platelets, there was evidence of endocardial and pulmonary artery thrombi in one animal, although it survived the entire observation period. Rehydrated, lyophilized platelets have previously been infused into rabbits [34], dogs [12] and baboons [35] without thrombotic complications being reported, and our findings were therefore unexpected. Interestingly, the pathologic coagulation events may result from a xenographic reaction related to known molecular incompatibilities between coagulation factors in pig and human platelets [36]. Specifically, porcine VWF is capable of inducing aggregation of human platelets in vitro independently of any agonist, the consequence of a conformational difference in VWF across species and enhanced binding to GPIb [37,38]. Therefore, porcine VWF would facilitate pathologic intravascular coagulation via non-specific interactions with GPIb receptors on the human lyophilized platelets [39]. A study by Gaca et al. [40] showed that blocking of porcine VWF–GPIb with an anti-GPIb monoclonal antibody prevented agglutination of human and baboon platelets in vitro and prevented platelet deposition in a pig-to-baboon lung transplant model. Alternatively, it is also possible that the thrombotic events represent amplified procoagulant Stasix activity in the setting of traumatic hemorrhage. Another intriguing prospect is that the xenographic interactions predominantly mediated the potent hemostatic effects of Stasix, an observation that calls for further exploration. In any case, these thrombotic events mandate additional investigations in the preclinical setting prior to human application of Stasix for traumatic hemorrhage.

Several important limitations of this study must be addressed. As discussed, potential xenographic interactions between human lyophilized platelets and porcine VWF probably confound the study results. The use of porcine lyophilized platelets may be more appropriate for evaluation of these preparations as a hemostatic agent within the current model. In the study of human lyophilized platelets, the use non-human primates may minimize species-related incompatibilities and serve as a more suitable large animal model. Also, it is possible that the Stasix dose that we identified as therapeutic may be excessive, and that with further dose titration a safer dose could be established. Infusible hemostatics represent a class of pharmaceuticals with a narrow therapeutic window, where benefits (effective hemostasis) and risks (thrombotic events) need to be balanced within the context of the drug’s indication (life-threatening hemorrhage). Finally, as the TEG and coagulation parameters were largely unrevealing, the addition of abciximab (Reopro, Centocor Inc., Philadelphia, PA, USA), a GPIIb–IIIa receptor inhibitor, to the TEG assay could be used to isolate the platelet contribution to these coagulation assays and may be useful in future studies to definitively quantify the effect of the lyophilized platelets on these metrics [41].

In summary, lyophilized platelets offer significant logistic and therapeutic advantages as a hemostatic agent for the control of traumatic hemorrhage. In the current study, rehydrated, lyophilized platelets (Stasix) demonstrated efficacy with significantly decreased blood loss, improved hemodynamics, decreased lactic acid levels and markedly increased survival in a porcine liver injury non-compressible hemorrhage model. However, significant thrombotic events were observed, possibly related to xenographic interactions between human platelets and porcine coagulation factors. Although further preclinical investigation is certainly warranted, this study represents an important step towards the clinical feasibility of a rehydrated, lyophilized infusible hemostatic platelet product.


Conception and design: J. S. Hawksworth, E. A. Elster, and D. K. Tadaki. Acquisition of data: J. S. Hawksworth, D. Fryer, F. Sheppard, V. Morthole, G. Krishnamurthy, and T. Tomori. Analysis and interpretation of data: J. S. Hawksworth, E. A. Elster, T. S. Brown, and D. K. Tadaki. Drafting of manuscript: J. S. Hawksworth, F. Sheppard, and D. K. Tadaki. Critical revision: E. A. Elster, T. S. Brown, and D. K. Tadaki. Obtaining funding: D. K. Tadaki. Supervision: E. A. Elster and D. K. Tadaki.


We gratefully acknowledge C. Morrisette for his statistical advice and expertise. We also thank T. Bristol and M. G. Delima for their skillful technical assistance with the animal experiments. This research was funded by the BUMED Advanced Medical Technology Development Program.

Disclosure of Conflict of Interests

The authors state that they have no conflict of interest.


  1. The views expressed in this presentation are those of the authors and do not reflect the official policy of the Department of the Army, Department of the Navy, the Department of Defense or the United States government.