The clinical value of serum S-100 protein measurements in minor head injury: a Scandinavian multicentre study

Purpose: This study of patients with minor head injury was designed to inve stigate the relation of S-100 protein measurements to computed tomograpy (C T) findings and patients outcomes. Increased serum levels of this protein w ere hypothetized to predict intracranial pathology and increased frequency of post-concussion symptoms. Methods: One hundred and eighty-two patients were studied with Glasgow Coma Scale scores of 13-15. The study recruited patients from three Scandinavia n neurotrauma centres. Serum levels of S-100 protein were measured at admit tance and CT scans of the brain were obtained within 24 hours postinjury in all patients. Outcome was evaluated with the Rivermead Postconcussion Symp toms Questionnaire (RPQ) 3 months after the injury. Results: Increased serum level of S-100 protein was detected in 69 (38%) pa tients, and CT scan demonstrated intracranial pathology in 10 (5%) (brain c ontusion in seven, epidural haematoma in two, traumatic subarachnoid haemor rhage in one). The proportion of patients with detectable serum level was s ignificantly ( p < 0.01) higher among those with intracranial pathology (90 %) compared to those without (35%). The negative predictive value of an und etectable S-100 level was 0.99. Sixty-two per cent reported one or more pos t-concussion symptoms at follow-up. A trend was observed towards an increas ed frequency of post-concussion symptoms among patients with detectable ser um levels. Conclusions: Undetectable serum level of S-100 protein predicts normal intr acranial findings on CT scan. Determination of S-100 protein in serum may b e used to select patients for CT scanning. Increased S-100 serum levels may be more related to post-concussion symptoms caused by mild traumatic brain injury than to symptoms of psychological origin.