S100B Serum Elevation Predicts In-Hospital Mortality After Brain Arteriovenous Malformation Rupture

Background and Purpose-S100B protein serum elevation has been associated with poor prognosis in neurologically ill patients. The purpose of this study is to determine whether elevation of S100B is associated with increased in-hospital mortality after brain arteriovenous malformation rupture. Methods-This is a retrospective study of patients admitted for brain arteriovenous malformation rupture. The study population was divided into derivation and validation cohorts. Univariate followed by multivariate logistic regression was used to determine whether elevation of S100B serum levels above 0.5 mu g/L during the first 48 hours after admission (S100Bmax48) was associated with in-hospital mortality. Results-Two hundred and three ruptures met inclusion criteria. Twenty-three led to in-hospital mortality (11%). Mean S100Bmax48 was 0.49 +/- 0.62 mu g/L. In the derivation cohort (n=101 ruptures), multivariate analysis found Glasgow coma scale score <= 8 (odds ratio, 21; 95% CI, 2-216; 0.001) and an S100Bmax48>0.5 mu g/L (odds ratio, 19; 95% CI, 2-188; P=0.001) to be associated with in-hospital mortality. When applied to the validation cohort (n= 102 ruptures), the same model found only S100Bmax48>0.5 mu g/L (odds ratio, 8; 95% CI, 1.5-44; P=0.01) to be associated with in-hospital mortality. Conclusions-Elevated S100B protein serum level is strongly associated with in-hospital mortality after brain arteriovenous malformation rupture.