S100B protein as a screening tool for computed tomography findings after mild traumatic brain injury: Systematic review and meta-analysis

Primary objective: To determine whether S100B protein in serum can predict intracranial lesions on computed tomography (CT) scan after mild traumatic brain injury (MTBI). Research design: Systematic review and meta-analysis Methods and procedures: A literature search was conducted using Medline, Embase, Cochrane, Google Scholar, CINAHL, SUMSearch, Bandolier, Trip databases, bibliographies from identified articles and review article references. Eligible articles were defined as observational studies including patients with MTBI who underwent post-traumatic head CT scan and assessing the screening role of 51008 protein. Main outcomes and results: There was a significant positive association between S100B protein concentration and positive CT scan (22 studies, SMD=1.92, 95% Cl=1.29-2.45, I-2=100%; p<0.001). The pooled sensitivity and specificity values for a cut-point range=0.16-0.20 mu g L-1 were 98.65 (95% Cl=95.53-101.77; I-2=0.0%) and 50.69 (95% Cl=40.69-60.69; I-2=76.3%), respectively. The threshold for serum S100B protein with 99.63 (95% Cl=96.00-103.25; I-2=0.0%) sensitivity and 46.94 (95% Cl=39.01-54.87; I-2=95.5%) specificity was >0.20 mu g L-1. Conclusions: After MTBI, serum S100B protein levels are significantly associated with the presence of intracranial lesions on CT scan. Measuring the protein could be useful in screening high risk MTBI patients and decreasing unnecessary CT examinations.