Maayke Hunfeld

75 Survey neuroprognostication Introduction Both in-hospital cardiac arrest (IHCA) and out-of-hospital cardiac arrest (OHCA) are uncommon in children. The incidence of paediatric OHCA ranges from 9.0 to 19.7 per 100,000 persons/year, including traumatic cardiac arrest (CA)(1-3). The incidence of paediatric IHCA widely ranges from 0.77 to 21 per 1000 hospital admissions (4-7). Whereas CA in adults is mostly of cardiac origin, in children it is commonly due to hypoxia (8). In our retrospective cohort study, 33% of 401 children with IHCA or OHCA had no return of spontaneous circulation (ROSC), whereas 34% died in the paediatric intensive care unit (PICU) due to severe neurological injury, brain death or respiratory/cardiovascular failure (9). Prognosis of CA in children is grim but over the past decades survival rates after CA have improved. This may be due to the combined effect of availability of early basic life support (BLS), the use of automated external defibrillators (AED) and improved post-ROSC care (10-13). However due to the improved survival rates numbers of survivors with neurological impairments due to hypoxic ischemic brain injury may increase (10, 14, 15). Predicting long-term outcome in children after ROSC remains challenging, especially for children who remain comatose for more than 24 hours (16). Different methods may be used in these comatose children in order to determine the neurological prognosis (neuro-prognostication): neurological examination (17, 18), routine and continuous electroencephalography (EEG) (16, 18-20), neuroimaging by computed tomography (CT) (21, 22) and magnetic resonance imaging (MRI) (23-26), serum biomarkers: Neuron-Specific Enolase (NSE) and S100 calcium-binding protein B (S100B) (27-29), and somatosensory evoked potential (SSEP) (17, 30). But the value for prognostication for these different methods is lacking. International guidelines for neuro-prognostication after CA are available for adults (31, 32), but not for children. Therefore, current practices may vary worldwide. It is crucial to predict neurological outcome as accurately as possible in these children in order to discuss further steps of treatment and to inform parents correctly. An inaccurate prediction of long-term outcome could lead to premature withdrawing of life-sustaining treatment (WLST) or at the other end of the spectrum severely disabled children with persistent vegetative state and high impact on resources and caregivers (33). The aim of our survey was to describe current practices in European PICUs regarding neuro-prognostication in comatose children after CA, in particular, the methods used, their timing, and end-of-life decision making. Due to the lack of international guidelines, we hypothesized that practices not only differ between various PICUs, but also may vary within PICUs. With this survey we aim to identify relevant research questions and priorities across Europe in order to optimize and standardise neuro-prognostication in European PICUs. 3