Extracorporeal Membrane Oxygenation (ECMO) represents a promising technology for resuscitating trauma patients with severe lung and/or cardiac injury. Common ECMO-related complications of thrombosis and bleeding result in significant rates of mortality and morbidity. Multiple studies with ex-vivo ECMO devices using blood demonstrate an increased proinflammatory response. Mitochondrial DNA (mtDNA) is a known damage associated molecular pattern (DAMP) that has been shown to have proinflammatory effects. Recent studies have correlated poor clinical outcomes to increased levels of plasma mtDNA. The objective of this study was to determine the levels of mtDNA in an ECMO model as a function of shear stress and duration of pump flow and evaluate the possible impact of mtDNA on platelet function.


Blood was collected from healthy donors under a US Army Institute of Surgical Research approved standard operating procedure. We constructed an ex-vivo ECMO circuit that circulates heparinized human blood for six hours at differing flow rates (static control-0 L/min, 0.3, and 0.7 L/min). Quantification of mtDNA and assays of platelet function were performed on samples obtained from each group and were compared to static controls at 0, 60, 120, 240, and 360 min. mtDNA levels were assayed by measuring the copy number of the NADH dehydrogenase 1 gene using quantitative real-time PCR. Additional tests were performed on platelet count, aggregation, viability (as measured by calcein/CD41a), TEG, and Multiplate.


Increased flow rates and duration resulted in significantly higher levels of plasma mtDNA compared to static controls (static control, 0 L/min flow rate: 120 min = 3 x 105 ± 1.8 x 105, 240 min = 4 x 105 ± 2 x 105 copies/µL; 0.3 L/min flow rate: 0 min = 4 x 105 ± 0.8 x 105, 120 min = 8 x 105± 0.2 x 105, 240 min = 8 x 105 ± 1.6 x 105 copies/µL; 0.7 L/min flow rate: 0 min = 3 x 105 ± 0.9x 105, 120 min = 19 x 105± 4 x 105, 240 min = 35 x 105 ± 6 x 105 copies/µL; p < 0.001). Platelet count and viability were not significantly different. In contrast, both TEG platelet mapping and Multiplate demonstrated significant losses in ADP and collagen aggregation response (Multiplate at 120 min: static control 1564± 265, 0.3 L/min = 830 ± 474, 0.7 L/min = 1010 ± 261; 240 min: static control 874± 381, 0.3 L/min = 527 ± 384, 0.7 L/min = 405 ± 268; p < 0.01) when compared to static controls.


Increased mtDNA in surgical patients has been linked to poor outcomes. It has recently been shown that circulating mtDNA can trigger the innate immune system through toll-like receptor 9 (TLR9) and NF-κB activation, causing deleterious effects. In this study we found that mtDNA levels significantly increase as a function of shear stress and increased duration of pump flow. These increases were shown to be associated with loss in platelet function in both TEG and Multiplate measures of ADP and collagen agonist stimulated responses. Future ECMO studies will explore the correlation between circulating mtDNA levels and TLR9. mtDNA may be a clinically relevant biomarker and therapeutic target for efforts to mitigate life-threatening ECMO coagulation complications.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.