Diagnostic Accuracy of S100B in Predicting Intracranial Abnormalities on CT Imaging Following Mild Traumatic Brain Injury: A Systematic Review and Meta-analysis.

Abstract

Traumatic brain injury (TBI) is a major cause of health loss and disabilities globally, burdening health care systems. Mild TBI is a common cause of emergency department visits. Computed tomography (CT) scans are the mainstay for acute TBI imaging. S100 calcium-binding protein B (S100B) biomarker is promising for predicting intracranial lesions on CTs in mild TBI. A comprehensive search of the literature was conducted on PubMed, Google Scholar, and Cochrane electronic databases to find eligible studies reporting the diagnostic performance of S100B. A meta-analysis was conducted to evaluate the predictive ability of S100B for CT imaging abnormalities. Of 1545 articles, 32 were included in our meta-analysis. At the threshold of 0.1 μg/L, a bivariate model showed a sensitivity of 89% (95% confidence interval [CI] 83-92) with a specificity of 32% (95% CI 26-39). The aggregate analysis containing all cutoffs showed the optimal cutoff of 0.751 μg/L with a sensitivity of 64% (95% CI 32-87) and a specificity of 85% (95% CI 76-92). The optimal diagnostic performance of S100B in patients with Glasgow Coma Scale 14-15 was estimated to be 0.05 μg/L, with a sensitivity of 98% (95% CI 92-99) and a negative predictive value of 99%. These findings indicate that S100B analysis could minimize the need for unnecessary CT scans in individuals with mild TBI. The test's diagnostic accuracy improves when the S100B analysis is done within 3 h of the injury. However, further research is warranted to validate its superiority to other biomarkers before considering it the standard routine for managing mild TBI.