S100B and NSE analyses

As per clinical protocol since 2005, all TBI patients in our NICU have S100B and NSE sampled at admission and twice daily (06:00 and 18:00). From 2005 up until September 2008, the serum samples of S100B were analyzed at the Department of Clinical Chemistry, Karolinska University Hospital (fully blinded from any patient characteristic), using a quantitative automated immunoassay (LIAISON, DiaSorin, Saluggia, Italy). After that, the department changed method to an automated electrochemoluminescence assay (Modular E170, Elecsys, Roche Diagnostics, Basel, Switzerland) throughout the study period. S100B has been shown not to be significantly influenced by sample hemolysis [25].

NSE was analyzed throughout the whole period at the Department of Clinical Chemistry, Karolinska University Hospital, using a quantitative automated immunoassay (LIAISON, DiaSorin, Saluggia, Italy). Samples were not analyzed if the amount of hemoglobin exceeded 0.5 g/L which was assessed visually using a hemolysis scale. If the laboratory personnel determined that the amount of hemoglobin exceeded 0.5 g/L, the sample was discarded. The NSE samples were acquired simultaneously as S100B.

The samples were sent to the Department of Clinical Chemistry for immediate analysis. The admission sample was usually from a venous source while the subsequent samples were from arterial lines. The first three samples and the area under curve (AUC), calculated using these samples during the first 48 hours, were used. The lowest levels of detection (LLOD) are 0.02 μg/L for S100B and 0.04 μg/L for NSE on the LIAISON assay, and <0.005 μg/L for S100B on the Elecsys device. However, the lowest level of quantification (LLOQ, also known as functional sensitivity) has been shown to be 0.02 μg/L and 0.04 μg/L for S100B and NSE on the LIAISON, respectively, as well as <0.02 μg/L on the Elecsys assay, while lower concentrations sometimes also yield acceptable coefficients of variation [43–45]. Thus, all detection and quantification limits were lower than what were detected in our patient material, and we do not believe that the quantification range of the assays significantly altered our findings. The reference levels for healthy controls are generally considered to be <0.1 μg/L for S100B [11] and <13 μg/L for NSE [46], respectively.