The aim of the study was to assess adjunctive intravenous dexamethasone in adult community-acquired bacterial meningitis (BM) in daily practice. Analysis of consecutive patients (1990-2009) with acute community-acquired bacterial meningitis in a single centre in Zagreb, Croatia, N = 304. Adjusted relative risks [RR, dexamethasone vs. no dexamethasone (control)] of Glasgow Outcome Scale (GOS) = 1 (death) and GOS = 5 (full recovery) at discharge/end of specific treatment were estimated considering demographics; co-morbidity; BM pathogenesis and on-admission characteristics, and cerebrospinal fluid (CSF) inflammation markers; causative agent and antibiotic use. Two hundred forty (79%) patients had proven BM (43.1% Streptococcus pneumoniae, any other agent ≤ 8.2%). No independent effects of dexamethasone on GOS = 1 or GOS = 5 were observed in the entire cohort (dexamethasone n = 119, control n = 185; RR = 1.06, 95% CI 0.77-1.45 and RR = 0.99, CI 0.83-1.20, respectively), microbiologically proven disease (dexamethasone n = 104, control n = 136; RR = 0.97, CI 0.69-1.38 and RR = 1.03, CI 0.82-1.28), pneumococcal disease (dexamethasone n = 71, control n = 60; RR = 0.95, CI 0.53-1.70 and RR = 0.82, CI 0.57-1.18), and also in other BM, subgroups based on consciousness disturbance, CSF markers, prior use of antibiotics and timing of appropriate antibiotic treatment. CSF markers did not predict the outcomes. Our experience does not substantiate the reported benefits of adjunctive dexamethasone in adult BM. Socio-economic and methodological factors do not seem to explain this discrepancy. Empirical use of dexamethasone in this setting appears controversial.