P2C140: Optimal Inspired Oxygen Concentration During Chest Compressions in an Ovine Model of Perinatal Cardiac Arrest

Background: The optimal inspired oxygen concentration during resuscitation and chest compressions (CC) of term infants with perinatal asphyxia remains controversial. Insufficient oxygen delivery can worsen cerebral hypoxia, while there is increased evidence that a relatively brief exposure to 100% oxygen increases systemic oxidative stress and lung oxidant injury. Solving this “oxygen dilemma” necessitates determining an optimal inspired oxygen concentration that minimizes cerebral hypoxia and oxidative stress during CC. We aim to determine the optimal balance between cerebral oxygen delivery, systemic oxidative stress and its effect on vascular reactivity in near-term asphyxiated lambs exposed to 21%, 50% and 100% inspired oxygen during CC.

Methods: This was a blinded randomized control trial. Near-term fetal lambs (N=21) in cardiac arrest were randomized into 3 groups: 21%, 50% and 100% inspired oxygen during CC. Those with return of spontaneous circulation (ROSC) were ventilated for 20 minutes after ROSC. The lambs were continuously monitored throughout the study. Blood samples were collected at various time points, and tissue samples were collected at the end of the study – lung, brain, and pulmonary vessels. These were used to measure the following oxidative stress markers: 1) superoxide dismutase (SOD), 2) total oxidant status (TOS), 3) 3-nitrotyrosine, and 4) isoprostane. The pulmonary arteries (PA) and veins (PV) were isolated to assess their vascular reactivity (a marker of lung oxidative stress) in vitro in tissue baths using relaxing agents, SNAP and A23187.

Results: The incidence and time to ROSC were similar among the 3 groups. After ROSC, cerebral oxygen delivery was significantly lower in the 21% group compared to the 100% group (p 0.0208 by ANOVA); it was similar between 50% group and the 100% group. Of note, the 50% group had significantly lower total oxygen exposure compared to the other 2 groups (50% vs. 100% p < 0.001, 50% vs 21% p 0.0115 by ANOVA). The SOD activity in plasma was higher in the 100% group compared to the 21% group 20 minutes post-ROSC (p 0.0340 by ANOVA). Total oxidant status, 3-nitrotyrosine and isoprostane levels were not significantly different among the 3 groups. The pulmonary arteries in the 100% group showed a lesser degree of relaxation to SNAP compared to the 21% group (p 0.02 by ANOVA repeated measures). Pulmonary vascular reactivity to A23187 was not statistically different between the 3 groups.

Conclusion: Cerebral oxygen delivery was low in near-term asphyxiated lambs in cardiac arrest exposed to 21% inspired oxygen during CC, while there was evidence of increased oxidative stress injury in those exposed to 100% inspired oxygen. Intermediate inspired oxygen at 50% during CC may provide the optimal balance between oxygen delivery to the brain while minimizing oxygen toxicity. Further studies are warranted to evaluate this result.