Maayke Hunfeld

85 Survey neuroprognostication European PICUs. Once decided that the prognosis is futile, consequences differed between respondents and countries. In order to determine neurological prognosis the majority of respondents (94%) thought that ancillary tests were needed besides neurological examination, being MRI and EEG as most useful. In children, there are few studies suggesting that absent brainstem reflexes and low GCS motor score within 12-24 hours after CA predicts a poor outcome (17, 18). In the past, there have been a number of studies regarding the prognostic value of other neuromonitoring modalities (EEG, MRI, SSEP, CT, biomarkers) in to predict outcome. An EEG with early (after 24 h) normal or continuous background pattern or with sleep spindles is associated with a good outcome, whereas burst suppression, flat, discontinuous background patterns and early epilepsy with poor outcome (16, 18-20). A recent study by Fung et al. concluded that EEG findings in children after CA must be used in overall clinical context to prognosticate early (35). A normal MRI after paediatric CA seems a predictor of a favourable outcome (between day 3-7 post-CA), whereas damage in multiple brain lobes and basal ganglia on T1/ T2 images, cytotoxic edema globally or in the basal ganglia or multiple cortical brain regions and low apparent diffusion coefficient (ADC) values on diffusion weighted imaging (DWI) MRI are associated with poor outcome (between day 3-7 post-CA) (23- 26, 36). Loss of gray-white matter differentiation (in particular in basal ganglia) and basilar cistern plus sulcal effacement on CT are associated with poor outcome (21, 22). Over the past 20 years, only two studies have been published regarding SSEP in children with hypoxic ischemic encephalopathy (17, 30) concluding that bilateral absence of the cortical N20 wave within 7 days after CA predicts poor outcome. However thresholds for SSEP are not defined yet. Increased levels of NSE and S-100B from respectively 24-48 h and 12-48 h post ROSC are associated with unfavourable outcome, however thresholds are lacking (27-29). All above mentioned studies were single center studies, mostly retrospective with small cohorts and clinicians that were not blinded for the results. Follow-up was mostly short-term (at hospital discharge) with gross outcome scales. Very recently (after the completion of this survey) a scientific statement has been published on paediatric post-cardiac arrest care by Topjian et al (37). They concluded that no single test (neurological examination, EEG, neuro-imaging, SSEP, biomarkers) was found to be sufficiently accurate and reliable for prognostication after paediatric CA. Multiple factors and ancillary tests should be considered when predicting outcome in children who achieve ROSC after CA. According to this statement we would like to recommend te next approach for clinical practice: When a child remains comatose after CA, clinicians should combine individual patient information (medical history, aetiology of CA, CPR variables like duration, witnessed arrest, bystander CPR etc) with the results of neurological examination (serial exams for at least 72 h) and ancillary 3