Maayke Hunfeld

112 Chapter 3 Can we improve prediction of neurological outcome in children after cardiac arrest? Cardiac arrest (CA) is rare in children however it occurs in out-of-hospital settings in up to 6000 children per year in USA; only 10-30% are successfully resuscitated and admitted to PICU [1]. Due to advances in intensive care medicine survival after paediatric CA has improved, but the reduction in mortality is reflected in the significant burden of living with lifelong neurological sequelae as a consequence of hypoxic-ischemic brain injury. It is therefore of utmost importance that accurate early prognostic guidelines after CA are made available for children, as they are for adults. In acute emergency situation all diagnostic and therapeutic procedures are focused maximally to save the child’s life. Along with neurologic examination, urgent CT head scan or MRI is done, not only to assess the brain damage but mainly to rule out the potential aetiology of cardiac arrest (trauma, intracranial haemorrhage etc.), if the cause of CA is not obvious from the history. Duration of initial resuscitation/time to return of spontaneous circulation (ROSC), biomarkers and early electroencephalography may provide additional tools to the paediatric intensivist when giving initial information to the family regarding the possible outcome [1,2]. Although combining CT imaging with clinical examination may improve the ability to prognosticate, in the era of advanced postresuscitation care neurological assessment should be undertaken with caution acknowledging the impact of hypothermia, sedation and ventilator support. However when a child remains comatose for more than 24 hours predicting long-term neurological outcome is challenging, especially when being aware that no diagnostic tests are 100% sensitive nor specific for neurologic outcome.To address these dilemmas in this issue, Hunfeld and colleagues, report the results of an anonymous online survey about current practices in European PICUs regarding neuro-prognostication in comatose children after CA [2]. They found that among paediatric intensivists and paediatric neurologists three methods were considered as most useful to predict neurological outcome: neurological examination (resp. 91% and 90%), brain MRI (resp. 78% and 87%) and cEEG (resp. 53% and 39%). In the literature there are a number of studies published about the prognostic value of these methods, but most of them are retrospective and single centre studies. Early head CT may to some extent help in prognosis of more unfavourable outcome: loss of gray-white matter (GWM) differentiation, in particular in basal ganglia or sulcal effacement had unfavourable neurologic outcomes as demonstrated by Starling et al., who analysed 78 patients with head CT performed within 24 hours of ROSC with a median time to CT of 3.3 hours [1]. Among children not treated with hypothermia after CA early electroencephalography (EEG) is routinely used in many PICUs as background activity reflects the degree of acute cerebral damage. In a