BACKGROUND: CSF can be leaked from the nose or ear due to fractures, tumors or surgical procedures in the skull base region, and the threat of impending meningitis necessitates early identification of it. Since 2-transferrin occurs practically in cerebrospinal fluid (CSF) and not in other body fluid, its detection from the rhinorrhea or otorrhea can be used for the diagnosis of CSF leakage. We carried out immunofixation-silver stain (IF-SS) method for detection of 2-transferrin in the CSF in order to know optimal identification condition of specific cerebrogenic marker. METHODS: The fresh CSF sample was collected by spinal tapping. 2-Transferrin was estimated by quantifying the total transferrin by nephelomertry (Behring, Germany). 2-Transferrin of CSF was identified by electrophoresis using Titan gel high resolution protein system (Beckman, USA), immunofixation with anti-human transferrin antibody (Dako, Denmark) and then stained with silver nitrate. Serial dilutions of CSF were performed to know the detection limit of 2-transferrin. To know the influence of blood mixing, tests for mixed specimen of serum and hemolysate in CSF were performed. To evaluate the specimen storage condition, tests for different temperature and storage time were performed . RESULTS: By IF-SS method, identification limit of 2-transferrin was 0.5 mg/dL in 1:4 diluted CSF with distilled water. And 2-transferrin could be detected in condition of mixing serum protein (7.5 g/dL) or hemoglobin (13 g/dL) with CSF up to 6 : 4. At various sample storage condition, such as 37degrees C, room temperature, and 4degrees C, band intensity decreased abruptly after 1 day, and it was not detected 5 days later. Mean while, in -20degrees C and -70degrees C, 2-transferin band was detected after 10 days. CONCLUSIONS: IF-SS method was sufficiently sensitive and specific for invalidation by blood contamination, and seems to be used as effective identification of 2-transferrin in the CSF without sample concentration, less diagnostic test for CSF leakage.