Serum S100B in elderly patients with and without delirium.

OBJECTIVE Elevation of S100B has been shown after various neurologic diseases with cognitive dysfunction. The aim of this study was to compare the serum level of S100B of patients with and without delirium and investigate the possible associations with different subtypes of delirium. METHODS Acutely admitted medical patients aged 65 years or more were included from 2005 through 2008. Delirium was diagnosed by Confusion Assessment Method, delirium subtype by Delirium Symptom Interview and preexistent global cognitive function by the 'Informant Questionnaire on Cognitive Decline-short form'. S100B levels were determined in serum by electrochemiluminescence immunoassay. RESULTS Samples of 412 patients were included, 91 during delirium, 35 after delirium and 286 of patients without delirium. Patients with delirium (31%) were significantly older, 81.5 versus 76.6 years (p < 0.001) and experienced significantly more often preexistent cognitive and functional impairment (p < 0.001). S100B level differed significantly (p = 0.004) between the three groups: median 0.07 microg/L (inter-quartile ranges: 0.05-0.14 microg/L) during delirium, 0.12 microg/L (0.05-0.29 microg/L) after delirium and 0.06 microg/L (0.03-0.10 microg/L) in patients without delirium. Combining the impact of cognitive impairment, infection and age on S100B, highest S100B was observed in the oldest patients after delirium with preexistent cognitive impaired and infection. Delirium subtype and S100B level were not significantly correlated. CONCLUSION Higher S100B levels were found in patients with delirium than in patients without delirium, with highest levels of S100B in samples taken after delirium. Future studies are needed to elucidate the mechanism responsible for the increase of S100B and the possible association with long term cognitive impairment.