(H-303) The Use of a Microfluidic Transection Model to Evaluate Coagulopathies and Hemostatic Resuscitation in Trauma

Introduction:: Bleeding secondary to trauma is one of the leading causes of death in the modern world. Hemostatic resuscitation with platelet products is an essential aspect of the therapeutic approach for traumatic bleeding. Effective in vitro methods to evaluate platelet product contribution to bleeding control would be a valuable tool for transfusion medicine. The objective of this study was to evaluate platelet product contribution to hemostasis ex vivo in a novel microfluidic model of vessel transection. We hypothesized that simulated resuscitation with platelet products used in clinical transfusion would rescue hemostatic function in induced coagulopathies. In addition, we also compared storage temperatures and duration, hypothesizing that function would diminish over the course of storage and that cold platelets would be more hemostatically active as is suggested by in vitro product assessment in the literature.

Materials and Methods:: We designed and fabricated a microfluidic mold using photolithography techniques to create a vascular channel and vessel transection leading to an extravascular space coated with collagen and tissue factor (Figure 1A). Heparinized whole blood was collected from healthy volunteers and used as recipient blood samples (n=17). Platelet products (n=6) stored in plasma were collected via apheresis from healthy donors per national blood collection standards. Platelets were labelled with fluorescent anti-CD41 using different fluorophores for donor and recipient. Whole blood was mixed with 60% normal saline to yield diluted blood (dWB), a common iatrogenic condition resulting from volume resuscitation (n=5), or thrombocytopenia (tWB) (n=12) was induced via centrifugation and resuspension to simulate platelet consumption. Samples were perfused through the microfluidic model by constant pressure via gravity (initial wall shear rate 10000/s, high shear appropriate to mimic injury). Outflow mass was measured to quantify blood loss and kinetic immunofluorescent images were captured. A plateau of mass outflow represents clot formation inducing occlusion of the injury site (i.e. cessation of bleeding). Outflow lines were perfused with EDTA to quench coagulation downstream of the injury site, ensuring a plateau in outflow mass was an indicator of true occlusion of the injury site. A volumetric equivalent of 4 platelet units was used to resuscitate recipient blood (1:5 ratio of platelet product:blood). Transfusion was simulated with platelet products at days 2, 5, and 7 of room temperature storage (current approved standard) and 2, 5, 7, 14, and 21 of cold storage (4C).

Results, Conclusions, and Discussions:: Dilutional coagulopathy samples resulted in an average of 2.55g of blood loss compared to 1.12g in matched whole blood (p=0.0289). Thrombocytopenic samples resulted in an average of 1.35g of blood loss compared to 1.07g in matched whole blood (p=0.0267). Clots that formed in the transected vessel space had high platelet content, and in simulated transfusion samples these clots predominantly incorporated platelets from the platelet product over the recipient blood sample (Figure 1B). Platelet products that were stored at room temperature for 7 days and used to resuscitate thrombocytopenic samples did not reduce blood loss compared to matched thrombocytopenic untreated controls (average of 1.48g vs. 1.51g, respectively; p=0.9941). Platelet products that were stored at cold temperatures for 21 days and used to resuscitate thrombocytopenic samples increased blood loss compared to matched thrombocytopenic untreated controls (average of 1.79g vs. 1.19g, respectively; p=0.0720) (Figure 1C-D). When platelet products were used to resuscitate tWB samples, day 2 units regardless of storage temperature had significantly lower blood loss compared to units at storage days 5-21 (p=0.0013) (Figure 1E).

In this study, platelet product contribution to hemostasis was evaluated ex vivo in conjunction with induced coagulopathies with blood loss as a primary microfluidic endpoint. While both dilution and induction of thrombocytopenia resulted in significant increases in blood loss, dilution resulted in more severe blood loss perhaps due to the altered viscosity of these samples. In these two induced coagulopathies studied, platelet product transfusion did not reduce blood loss. Surprisingly, cold-stored platelet products resulted in an increase in blood loss in our ex vivo transfusion simulation, perhaps indicating an inhibitory effect of the soluble mediators in the storage bag, or the critical role of resuscitating with a viscosity-restoring element in this model. As expected, later storage versus early storage also deteriorated the ability of platelet products to resuscitate coagulopathies, indicating the need to further evaluate the intrinsic function and biological mechanisms of platelet products over their storage duration.





Acknowledgements (Optional): :

References (Optional): :