Serum S100β as a predictor of severity and outcomes for mixed subtype acute ischaemic stroke

Abstract

Serum S100β levels are mostly used for predicting outcomes of large-vessel stroke. Its application to mixed subtypes of acute ischaemic stroke (AIS) has been limited. Patients with mixed subtypes of AIS who were aged over 18 years and presented within 24 hours of stroke onset were consecutively enrolled. Serum S100β levels at presentation (S100βb) and 72 hours (S100β72hrs), and corresponding National Institutes of Health Stroke Scale (NIHSSb and NIHSS72hrs, respectively) scores were assessed. Stroke outcomes were evaluated using the modified Rankin Scale (mRs) at 30 days (mRs30) and 90 days (mRs90). Correlations between S100βb and S100β72hrs, as well as differences between the two (∆S100β) and the corresponding NIHSS, mRs30 and mRs90 scores, were evaluated (p < 0.05). 35 patients were eligible for analysis. On univariate analysis, stroke outcomes had a significant association with S100βb, S100β72hrs, NIHSSb, NIHSS72hrs and ∆S100β. Both S100βb and S100β72hrs correlated with corresponding NIHSS values (ρb = 0.51, p < 0.001; ρ72hrs = 0.74, p < 0.001), mRs30 (ρb = 0.58, p < 0.001; ρ72hrs = 0.72, p < 0.001) and mRs90 (ρb = 0.51, p = 0.002; ρ72hrs = 0.68, p < 0.001). Correlations existed between ∆S100β and mRs30 (ρ = 0.74, p < 0.001) and mRs90 (ρ = 0.71, p < 0.001). Practical cut-off points for unfavourable outcomes (mRs 3-6) were S100β72hrs > 0.288 µg/L (sensitivity 92.3%, specificity 86.4%) and ∆S100β > 0.125 µg/L (sensitivity 100%, specificity 81.8%). High serum S100β is associated with unfavourable outcomes for mixed subtype AIS. Cut-off values of S100β72hrs and ∆S100β were optimal for predicting unfavourable stroke outcomes.