Effects of systemic mild hypercapnia on brain function in a refractory cardiac arrest model treated with extracorporeal cardiopulmonary resuscitation

Abstract

Introduction Cardiac arrest (CA) is a major cause of mortality and neurologic disability. Extracorporeal cardiopulmonary resuscitation (ECPR) may be used in refractory CA to potentially reduce hypoxic-ischemic brain injuries. Currently, no recommendation exists regarding optimal carbon dioxide levels in ECPR patients. Methods Twelve pigs were submitted to 5 min of untreated ventricular fibrillation followed by 25 min of CPR; 30 min after CA induction, ECPR was initiated and defibrillations delivered until ROSC. Pigs were randomly assigned to normocapnia (NC; PaCO2 35–45 mmHg, n = 6) or hypercapnia (HC; PaCO2 50–55 mmHg, n = 6). Animals underwent continuous multimodal neurological and cardiovascular monitoring. The primary outcome of the study was the change of brain tissue oxygen tension (PbtO2) over time. Other parameters were secondary outcomes. At 12 h post-ROSC, animals were euthanized and brain tissues were harvested for molecular analyses. Results Mean PbtO2 tended to be higher in the HC group (49.9 vs 36.4 mmHg in the NC group; p = 0.13), while mean intracranial pressure was similar between the two groups (19.9 vs 22.6 mmHg, respectively – p = 0.42). Cerebral glucose concentrations were significantly higher in the HC group (p = 0.01), without differences in lactate/pyruvate ratio, glycerol or glutamate. Caspase-8 gene expression in the parietal lobe was significantly lower in the HC group, with no differences in EEG cortical activity or serum biomarkers of brain injury. Conclusions In this model of ECPR, HC was associated with increased cerebral glucose concentrations and reduced caspase-8 expression, without changes in PbtO2 levels, intracranial pressure, cortical activity, or neurobiomarkers.