Can early serum levels of S100B protein predict the prognosis of patients with out-of-hospital cardiac arrest?

Abstract

This study aims to know if the level of S100B protein at the initiation of cardiopulmonary resuscitation (CPR) and immediately after return of spontaneous circulation (ROSC) can predict clinical outcome. A prospective observational study from December 2004 to October 2006 was conducted in an urban tertiary hospital emergency department. Clinical demographics for out-of-hospital cardiac arrest patients were collected based on the Utstein style. Outcomes collected included ROSC for 20min, survival to admission, survival and Glasgow Outcome Scale (GOS) at 1 month. S100B protein was measured twice before starting CPR (first S100B) and immediately after ROSC (second S100B). We investigated the association between S100B protein levels and clinical outcomes using a multivariate logistic regression model. A total of 151 patients were included (age: 60.2+/-16.8 years, male: 64.2%). Of these, 60 (39.7%) had ROSC and 46 (30.5%) survived to admission. After 1 month, 12 (8.0%) survived and only three patients showed good GOS (>/=4 points). The S100B levels were not different for ROSC, survival to admission and 1-month survival between survivors and non-survivors (p>0.05, first and second S100 B level). For the witnessed out-of-hospital cardiac arrest (OHCA) group (N=87), only the first S100B (1.22+/-0.85mugl(-1) vs. 3.91+/-4.25mugl(-1), p<0.001) showed significant difference for 1-month survival between survivors and non-survivors. The first S100B showed significant association with survival to emergency department (ED) but not 1-month survival (adjusted odds ratio (OR)=0.905, 95% confidence interval=0.821-0.998). Higher levels of S100B at start of CPR were significantly associated with lower survival to admission, and not for 1-month survival.