Predicting Outcomes Following Resuscitation

Abstract

Outcomes of pediatric resuscitation depend on the location, pre-arrest and arrest variables including quality of CPR. Out-of-hospital cardiac arrest has poorer survival and neurological outcomes due to the longer period of no-flow time, as well as the underlying etiologies of out-of-hospital cardiac arrest that are themselves associated with poor outcomes (e.g. Sudden Infant Death Syndrome or drowning). In contrast, more than 90 % of in-hospital pediatric cardiac arrests are witnessed or monitored, and CPR is provided. Half of in-hospital pediatric cardiac arrest victims are successfully resuscitated to return of spontaneous circulation and a quarter will survive to discharge. Sixty-five percent of survived children had favorable neurological outcomes. Pre-arrest and arrest variables are highly associated with survival and neurological outcomes. However, these pre-arrest and arrest variables tend to have a high false positive rate for predicting poor neurological outcomes. There are no reliable predictors of outcome in children. High quality of CPR is associated with short term survival outcomes. For post-arrest variables, absence of pupillary exams after 48 h is predictive for poor neurological outcome when therapeutic hypothermia is not induced. EEG finding with mild slowing and rapid improvement are associated with good outcomes, while burst suppression, electrocerebral silence, and lack of reactivity are associated with poor outcome. Somatosensory evoked potentials (SSEPs) are much less influenced by drugs, and resistant to environmental noise artifacts in contrast to bedside EEG. Bilateral absence of the N20 components in SSEPs is consistently associated with poor neurological outcomes. Serum neuron-specific enolase (NSE) and S-100B protein have been evaluated as prognostic indicators. NSE had higher discriminative ability for poor neurological outcomes compared to S100-B protein. For patients receiving therapeutic hypothermia, absence or extensor motor responses after achievement of normothermia is predictive for poor neurological outcomes. Neurological finding during therapeutic hypothermia is not reliable.