Quantitative assessment of brain microvascular and tissue oxygenation during cardiac arrest and resuscitation in pigs

Cardiac arrest is associated with a very high rate of mortality, in part due to inadequate tissue perfusion during attempts at resuscitation. Parameters such as mean arterial pressure and end-tidal carbon dioxide may not accurately reflect adequacy of tissue perfusion during cardiac resuscitation. We hypothesised that quantitative measurements of tissue oxygen tension would more accurately reflect adequacy of tissue perfusion during experimental cardiac arrest. Using oxygen-dependent quenching of phosphorescence, we made measurements of oxygen in the microcirculation and in the interstitial space of the brain and muscle in a porcine model of ventricular fibrillation and cardiopulmonary resuscitation. Measurements were performed at baseline, during untreated ventricular fibrillation, during resuscitation and after return of spontaneous circulation. After achieving stable baseline brain tissue oxygen tension, as measured using an Oxyphor G4-based phosphorescent microsensor, ventricular fibrillation resulted in an immediate reduction in all measured parameters. During cardiopulmonary resuscitation, brain oxygen tension remained unchanged. After the return of spontaneous circulation, all measured parameters including brain oxygen tension recovered to baseline levels. Muscle tissue oxygen tension followed a similar trend as the brain, but with slower response times. We conclude that measurements of brain tissue oxygen tension, which more accurately reflect adequacy of tissue perfusion during cardiac arrest and resuscitation, may contribute to the development of new strategies to optimise perfusion during cardiac resuscitation and improve patient outcomes after cardiac arrest.