Effects of hyperoxia on platelet aggregation, inflammation and oxidative stress during ex-vivo extracorporeal membrane oxygenation

Abstract

Objective: Despite increasing use of extracorporeal membrane oxygenation (ECMO) for support of refractory cardiac and pulmonary failure, it remains a complex, high-risk intervention. Current practice often exposes patients to higher doses of oxygen (hyperoxia) during ECMO in order to avoid episodes of hypoxia. However, hyperoxia may play a role in dysregulating the coagulation pathways and immune responses which are seen in ECMO patients. Thus, we aimed to investigate the effects of two different oxygen exposures in altering the immune and haemostatic systems during ex-vivo ECMO. Methods: An ex-vivo model was used with fresh human whole blood circulating at 4LPM, 37°C for 6 hours with exposure to oxygen doses of either 21% (PO2 of 138±3.9 mmHg; n=5) (mild hyperoxia) or 100% (PO2 of 531±12.7 mmHg; n=4) (severe hyperoxia). Calcium, glucose, CO2 and pH levels were monitored and kept constant throughout the experiment. ACT was also maintained ≥200 seconds. Serial blood samples were taken at 0,1,2,4 and 6 hours for analysis of platelet aggregometry, immune, inflammatory and oxidative stress markers. Results: ECMO resulted in reduced platelet response towards ADP (p< 0.05) and TRAP (p< 0.05) to induce aggregation overtime. It also increased the release of, neutrophil elastase (activation marker) after 6 hours of ECMO (p=0.013). Similarly, we observed elevated levels of the inflammatory chemokine IL-8 (p< 0.05), as well as an increased SOD activity (p=0.002), a marker of oxidative stress. However, hyperoxia did not augment these responses, with no significant differences detected between mild and severe hyperoxia. Conclusions: Our ex-vivo model demonstrated that the ECMO circuit itself triggers a pro-inflammatory and oxidative stress response and activates neutrophil granule release. However, the data suggests that exposure to supranormal oxygen levels alone does not amplify that response. Extended-duration studies, and inclusion of normoxia control and an endothelial component into the model could be beneficial to characterise longerterm changes.L6337649662020-12-31 <br />