Studies of patients with head trauma have demonstrated a correlation between a serum marker of brain tissue damage, namely S100B, and neuroradiological findings. It was recently demonstrated that the increases in serum S100B levels after heart surgery have extracerebral origins, probably surgically traumatized fat, muscle, and bone marrow. The current study examined multitrauma patients without head trauma, to determine whether soft-tissue and bone damage might confound the interpretation of elevated serum S100B concentrations for patients after head trauma. A commercial assay was used to determine serum S100B concentrations for a normal population (n = 459) and multitrauma patients without head injury (n = 17). Concentrations of the two subtypes of S100B (S100A1B and S100BB) were determined using separate noncommercial assays. The mean serum S100B concentration for a normal healthy population was 0.032 microg/L (median, 0.010 microg/L; standard deviation, 0.040 microg/L). The upper 97.5% and 95% reference limits were 0.13 and 0.10 microg/L, respectively. No major age or sex differences were observed. Among trauma patients, serum S100B levels were highest after bone fractures (range, 2-10 microg/L) and thoracic contusions without fractures (range, 0.5-4 microg/L). Burns (range, 0.8-5 microg/L) and minor bruises also produced increased S100B levels. S100A1B and S100BB were detected in all samples. Trauma, even in the absence of head trauma, results in high serum concentrations of S100B. Interpretation of elevated S100B concentrations immediately after multitrauma may be difficult because of extracerebral contributions. S100B may have a negative predictive value to exclude brain tissue damage after trauma. Similarly, nonacute S100B measurements may be of greater prognostic value than acute measurements.