Search

Background and Purpose: Patients resuscitated from cardiac arrest are at risk of subsequent death or poor neurological outcome up to a persistent vegetative state. We investigated the prognostic value of several epidemiological and clinical markers and two neuroproteins, neuron-specific enolase (NSE) and S-100 protein (S-100), in 97 patients undergoing cardiopulmonary resuscitation (CPR) after non-traumatic cardiac arrest between 1998 and 2002., Results: 52.6% of the patients died, 28.8% survived with severe, moderate or without neurological disorders, and 18.6% remained in a persistent vegetative state. Unconsciousness>48 h after CPR predicted a 60.6-fold (95% CI 14.3287-257.205, p=0.001) and a Glasgow Coma Scale (GCS)<6 points after 72 h a 11.2-fold (CI 95%, 3.55-36.44, p<0.001) risk of poor neurological outcome. Serum levels>or=65 ng/ml for NSE and >or=1.5 microg/l for S-100 increased the risk of death and persistent vegetative state 16.8 (95% CI 2.146-131.520)- and 12.6 (95% CI 1.1093-99.210)-fold, respectively. By combination of the GCS with elevated serum concentrations of both neuroproteins above the cut off levels on third day after CPR a poor neurological outcome was predicted with a specificity of 100%., Conclusion: The combination of GCS with the serum levels of both neuroproteins at 72 h after CPR permit a more reliable prediction of outcome in post arrest coma than the single markers alone, independent of the application of anaesthetic agents.

Severe neurological deficits are common characteristics of patients surviving cardiopulmonary resuscitation (CPR) or isolated severe head trauma (SHT). For comparative evaluation of underlying pathomechanisms, 22 patients with out-of-hospital cardiac arrest and successful CPR as well as 10 patients with SHT were included in our prospective study. Circulating S-100B was determined as an indicator of cellular brain damage. Interleukin-8 (IL-8), soluble E-selectin (sE-selectin) and polymorphonuclear (PMN-) elastase were measured as markers of systemic inflammation following whole body ischaemia and reperfusion injury. Venous blood samples were drawn on scene (median time 11.0 min after starting basic life support) and in the intensive care unit (median time 12.5 h thereafter) in CPR patients and at admission to hospital (median time 43.8 min after trauma) and approx. 12 h later in SHT patients. Biochemical parameters in these samples were compared with specimens taken from 20 healthy volunteers. Initial median S-100B levels of the CPR and SHT patients were both significantly increased compared with the controls. Twelve hours later, significant falls in S-100B revealed no differences between the two patient groups, but did not reach control values. Median IL-8 and sE-selectin levels entry to the study were elevated in both patient groups compared with controls and showed further rises within the following 12 h. Finally, increased initial median levels of PMN-elastase revealed significant differences between the patient groups and between patients and controls. Twelve hours later, median PMN-elastase values were equally elevated in the CPR and SHT subjects. Our preliminary data suggest similar pathomechanisms occurring after both CPR and SHT. Both clinical entities seem to be associated with early transient cellular brain damage as shown by prolonged rapidly increasing and subsequent fall in S-100B serum levels. In contrast, the prolonged elevation of circulating IL-8, sE-selectin and PMN-elastase may indicate a very similar systemic inflammatory response by endothelial cells and neutrophils initiated by ischaemia and reperfusion injury in both conditions. Further studies should be carried out to determine the cause and the prognostic value of these biochemical parameters in relation to long-term neurological outcome.