Abstracts

Rationale: Cardiac arrest results in whole body hypoperfusion leading to subsequent temperature sensitive cell destruction, multiorgan failure, and neuronal apoptosis. Hypothermia has been shown to lessen these harmful effects, thus, providing a degree of protection to the brain and heart. However, hypothermia also has systemic risks of immunosuppression, electrolyte imbalance and increased bleeding.  It is standard of care for post cardiac arrest patients to receive actively induced cooling to a target temperature of 32-36ºC. Recent literature suggests cooling to 36ºC can limit the systemic side effects. We aimed to compare the EEG prognostic predictive value in post cardiac arrest patients with target temperatures of 33 and 36ºC.  Methods: We retrospectively reviewed the clinical data of all 33 adult post cardiac arrest patients at our institution who were on EEG during hypothermia protocol between February 2017 and April 2018 with a target temperature of either 33 or 36º C.  Patients remained at the target temperature for at least 24 hours and then were rewarmed while under EEG monitoring, with continued monitoring at normothermia for at least 5 hours. Of the 33 patients, 28 patients were on continuous EEG monitoring that was initiated either during the cooling or rewarming periods. Five of the 33 patients had only 30-minute routine EEGs performed during the rewarming period when patients are at greatest risk for seizures.  The EEGs were interpreted by epileptologists. Generalized periodic discharges (GPDs), clinical or electrographic seizures, myoclonus, or electrocerebral silence were considered poor prognostic indicators. Neurologic outcome was categorized by survival or death.  Results: Of the 33 patients reviewed, 16 patients were cooled to 36ºC (Group A) and 17 patients were cooled to 33 º C (Group B). In Group A, the EEG in 7 patients showed GPDs, clinical myoclonus, electrographic seizures or near electrocerebral silence (ominous EEGs). All of these patients died. The EEG in the remaining 9 patients showed diffuse generalized slowing (nonspecific EEGs). Of these 9 patients, 4 survived and 5 died. In Group B, the EEG in 2 patients had ominous EEGs and these patients died. Nonspecific EEGs were seen in the remaining 15 patients, 7 survived, and 8 died.  Thus, there was a trend towards better EEG prognostic predictive value in Group A (p value = 0.099) for death. There was no difference in prognostic predictive value for those with nonspecific EEGs (p =1.000) in either Group A or B.  There was no significant difference in the outcomes of Group A and Group B (p value = 0.464).  Conclusions: The EEG more accurately predicted death in those cooled to 36ºC.  Given these findings, we would suggest that EEG provides better prognostic predictive value for poor outcomes in patients cooled to 36ºC than in those cooled to 33 ºC.  This may help inform decision making for both medical teams and families.  In the future, we hope to have a larger patient pool and to review their comorbidities in conjunction with EEG findings for a more complete understanding of the EEG prognostic predictive value and overall patient outcomes in those cooled to varying target temperatures.  Funding: None