PURPOSE:With the changing epidemiology of the etiological agents causing bacterial meningitis in Canada, the purposes of this study were to determine what empirical antibiotic(s) is/are advocated, and whether dexamethasone is recommended as adjunctive therapy by paediatric infectious diseases specialists for suspected meningitis caused by Streptococcus pneumoniae, Neisseria meningitides, Haemophilus influenzae type b, and partially treated (PT) bacterial meningitis. METHODS: A questionnaire invoking the specialists' preferred choices in treating bacterial meningitis was circulated by e-mail in March 1999 to 41 paediatric infectious diseases specialists or microbiologists (Paediatric Investigators Collaborative Network on Infections in Canada [PICNIC] members) who practised in 13 paediatric centres located in eight provinces. RESULTS: Thirty-two (78% response rate) replies were received by July 1999 from 11 paediatric centres in seven provinces. Dexamethasone was recommended by 11 of 32 (34%) respondents for suspected S pneumoniae, four of 32 (12%) for suspected N meningitides, 18 of 32 (56%) for suspected H influenzae type b and five of 32 (16%) for suspected PT meningitis. A trend for more frequent recommended use of dexamethasone for S pneumoniae, H influenzae type b and N meningitides meningitis was found in eastern provinces (Ontario, Quebec and Nova Scotia), as opposed to the western provinces (Manitoba, Saskatchewan, Alberta and British Columbia). The most commonly recommended first line empirical antibiotic therapy was vancomycin plus 3rd-generation cephalosporin (V+3C). This was recommended by 27 of 32 (84%) respondents for suspected S pneumoniae, seven of 32 (22%) for suspected N meningitides, six of 32 (19%) for suspected H influenzae type b and 24 of 32 (75%) for PT meningitis; the remainder recommended using a 3C empirically for each type of meningitis. The major reason cited by respondents for using V+3C for suspected S pneumoniae was the presence of high-and intermediate-level penicillin-resistant S pneumoniae in their centres. No differences were found in the choice of antibiotic(s) between practitioners in the eastern and western provinces. CONCLUSIONS: Although regional differences exist, the majority of paediatric infectious diseases experts no longer favour the use of dexamethasone for empirical therapy in most cases of bacterial meningitis. V+3C has become the regimen of choice in Canada for most cases of suspected bacterial meningitis.