Autotransfusion of Hemothorax Blood Is Prothrombotic but Inhibits Platelet Aggregation: A Potentiator of Disseminated Intravascular Coagulation

Introduction

Hemostatic resuscitation in combat casualty care presents unique challenges not typically encountered in metropolitan areas. Fresh whole blood and blood components are considered preferable to crystalloid or colloid solutions, but availability may be limited in austere wartime environments. Autologous shed blood from a traumatic hemothorax or hemoperitoneum, in theory, is an attractive alternative. This position is supported by the American Academy of Blood Banking, Eastern Association for the Surgery of Trauma, and Society of Critical Care Medicine, despite the fact that the safety and efficacy of shed blood has never been evaluated in randomized trials. Recent reports suggest that shed blood has poor hemostatic properties. This study was performed to evaluate the effects of shed hemothorax plasma and microparticles on platelet function and global tests of coagulation.

Methods

Shed traumatic hemothorax blood was obtained from 17 adult trauma patients requiring tube thoracostomy, every hour up to 4 hours after placement, according to IRB approved protocol. The hemothorax plasma was then isolated and frozen for later analysis. The effects of shed frozen plasma at 1 and 4 hours (HFP1; HFP4) was examined by performing mixing studies with fresh platelet rich plasma from normal donors at pre-specified serial dilutions of 1:8, 1:4, and 1:2 to mimic intravascular resuscitation . This was then analyzed by several assays. Thromboelastometry (ROTEM) of HFP was done to evaluate clotting time (CT), clot formation time (CFT), alpha angle, and lysis index at 45 minutes. Platelet aggregometry was done to assess the effect of shed HFP on platelet function in response to adenosine diphosphate (ADP), arachidonic acid (ASPI) and collagen. Fibrin degradation products (FDP), von Willebrand factor (vWF) and tissue factor (TF) levels in HFP were determined by enzyme-linked immunosorbent assays (ELISA ). vWF multimers in HFP were further separated by size using gel electrophoresis and detected by immunostaining. Microparticle analysis was performed using isolated microparticles from ultra-centrifuged HFP1 and HFP4 (HMP1 and HMP4) and compared to control isolated microparticles from fresh frozen plasma (FMP). HMP and FMP were then diluted with platelet rich plasma to a predetermined platelet count after which, ROTEM and platelet aggregometry were performed as described above. Non-diluted HFP and resuspended HMP1, HMP4 and FMP were then analyzed by flow cytometry for a variety of microparticle (MP) derived antigens including APC-H7-CD41a (GPIIb/IIIa derived MP), and APC-CD235a (Glycophorin A from red cell derived MP), and FITC-Lactadherin phosphatidylserine (platelet derived MP).

Results

ROTEM of HFP demonstrated a CT of 55.8% and 60.6% at the 1:8 dilution and 69.3% and 71.4% at the 1:2 dilution, for HFP1 and HFP4 respectively, compared to platelet rich plasma (PRP) and FFP (p<0.05) controls. Similar findings were seen in the CFT at the 1:8 dilution. HMP1 and HMP4 demonstrated decreased CT when compared to PRP and FMP (p<0.05).

HMP1 and HMP4 inhibited platelet aggregation in response to ADP, TXA2, and collagen with data expressed as area under the curve (U), when compared to FMP and PRP, all statistically significant (p<0.05). For example, PRP and FMP response to ADP was greater than 100U and HMP less than 50U (p<0.05). ELISA of HFP demonstrated significantly higher levels of FDP (>700ng/mL vs 0ng/mL p<0.05) and TF (>500 pg/mL vs <200 pg/mL p<0.05) when compared to FFP controls.

Flow cytometry showed increased microparticle concentrations in HFP1, HFP4, HMP1, and HMP4 samples with greater than 10 fold increase (MP/uL) in several markers (p<0.05) when compared to FFP and FMP. vWF multimer ratio demonstrated a loss in high molecular weight multimers and a gain in low molecular weight multimers in HFP1 and HFP4 (ratio high molecular weight:low molecular weight <1) when compared to FFP (ratio 2.5) with p value <0.05.

Conclusion

These results demonstrate that autologous shed hemothorax blood promotes a prothrombotic state due to increased tissue factor and microparticle content. Furthermore, clot localization may be impaired through inhibition of platelet aggregation by microparticles and degradation of vWF. These data suggest that randomized trials are indicated to establish the safety and efficacy of autologous shed traumatic hemothorax blood for trauma resuscitation.