Decrease in cerebral oxygenation at induction of therapeutic hypothermia seems related to decreased arterial CO2 tension

Abstract

Background: After cardiac arrest (CA), therapeutic hypothermia (TH) is the only therapy that improves neurological outcome and survival. However, hypothermia can alter CO2 production and solubility. This alteration can have an influence on cerebral perfusion and oxygenation in post‐CA patients. The FORE‐SIGHT absolute cerebral oximeter uses 4 wavelengths of laser light to determinethe levels of oxygenated and deoxygenated hemoglobin in the cerebral microvasculature. In this way, absolute cerebral oxygen saturation (SctO2) can be measured continuously and non‐invasively. In this study, cerebral oxygenation and pCO2 was measured during induction and maintenance of TH after CA.Methods: After IRB approval and with written informed consent, data were collected from 23 patients admitted to the hospital after CA. Therapeutic hypothermia (TH) (33°C) was induced by endovascular (Coolgard®, Alsius) or surface (Arctic Sun®, Medivance) cooling. All patients were sedated (propofol/remifentanil) for the duration of hypothermia. NIRS‐sensors were bilaterally applied to the frontotemporal area before the start of TH. Arterial blood gas values were collected every hour.Results: Twenty three patients were admitted to the intensive care unit after cardiac arrest. Cerebral tissue oxygenation values started at 68% (± 6). SctO2‐values decreased to 59% (± 3) within the first four hours after induction of TH. The decrease in cerebral oxygenation during induction of TH was could not be related to a change in hemodynamic parameters (MAP before induction of TH: 79 mmHg ± 19; at 33°C: 82 mmHg ± 9), nor to a change in SpO2 (start: 99% ± 1; 4h: 97% ± 3), or pO2 (start: 136,1 mmHg ± 73; 4h: 119 mmHg ± 30). However, SctO2 values showed a correlation with PaCO2. At the start of SctO2 monitoring, PaCO2 was 50,2 mmHg (± 9) but decreased to 36,1 mmHg (± 8) within four hours. Ventilatory management (tidal volume and ventilatory frequency) was not adjusted during monitoring.Conclusion: Non‐invasive monitoring of cerebral oxygenation showed a decrease in SctO2 during the first hours after inducing TH post ‐cardiac arrest. The observed decrease in PaCO2 with ensuing cerebral vasoconstriction could be responsible for the decrease in cerebral oxygenation.