Clinical variation in the early assessment and management of suspected community-acquired meningitis: a multicentre retrospective study.

Background: Bacterial meningitis is a medical emergency and timely management has been shown to improve outcomes. The aim of this study was to compare the early assessment and management of adults with suspected community-onset meningitis between hospitals and identify opportunities for clinical practice improvement.

Methods: This retrospective cohort study was conducted at three principal referral hospitals in Sydney, Australia. Adult patients with suspected meningitis undergoing cerebrospinal fluid sampling between 1 July 2018 and 31 June 2019 were included. Relevant clinical and laboratory data were extracted from the medical record. Differences between sites were analysed and factors associated with time to antimicrobial therapy were assessed by Cox regression.

Results: In 260 patients, the median time from triage to antibiotic administration was 332 min with a difference of up to 147 min between hospitals. Median time from triage to lumbar puncture (LP) was 366 min with an inter-hospital difference of up to 198 min. Seventy per cent of patients had neuroimaging prior to LP, and this group had a significantly longer median time to antibiotic administration (367 vs 231 min; P = 0.001). Guideline concordant antibiotics were administered in 84% of patients, with only 39% of those administered adjunctive corticosteroids. Seven (3%) patients had confirmed bacterial meningitis. Modifiable factors associated with earlier antimicrobial administration included infectious diseases involvement (adjusted hazard ratio [aHR], 1.50 [95% confidence interval (CI), 1.01-2.24]) and computed tomography (CT) scanning (aHR, 0.67 [95% CI, 0.46-0.98]).

Conclusion: Opportunities for improvement include reducing the time to LP and antibiotic administration, improving coadministration of corticosteroids and avoiding potentially unnecessary CT scanning.