The early fall in levels of S-100 beta in traumatic brain injury

Protein S-100 beta has been suggested as a prognostic marker in traumatic b rain injury, However, little is known of its behaviour in the immediate pos t-injury period. With Ethics Committee approval, we recruited 30 patients w ith a history of head injury presenting to our Accident and Emergency Depar tment. Blood was taken on arrival and at four hours post-injury. Serum S-10 0 beta was estimated using an immunoluminometric assay. Levels of S-100 bet a were seen to fall rapidly with time. Half-time was distributed non-parame trically with a median of 198 minutes. Using the Mann-Whitney U test we fou nd a statistically significant difference between non-desirable (Glasgow Ou tcome Score 1-3) and desirable (Glasgow Outcome Score 4-5) outcome on admis sion (p = 0.0155) but not at four hours (p = 0.1336). Levels of S-100 beta fell rapidly after its release following traumatic brain injury. Time after injury is therefore critical in assessing the significance of levels of S- 100 beta, and sampling should be as early as possible to gain maximum infor mation. If S-100 beta is to be assessed as a monitor of ongoing brain injur y in the intensive therapy unit sampling must be frequent (e.g. every 4 hou rs) to be able to detect rises in serum levels before they have decayed to baseline.