Background Since the release of linezolid vancomycin has been downgraded as an alternative in the treatment of central nervous system (CNS) infections caused by Gram-positive bacteria on account of its bad penetration and high incidence of nephrotoxicity. There are no studies with enough patients to support this trend.

Purpose To evaluate the efficacy and safety of vancomycin in CNS infections, and the impact of monitoring its pharmacokinetics.

Material and methods Descriptive retrospective study which included all patients with CNS infections treated with vancomycin and monitored. Patients aged under 18 and those who received less than 5 days’ treatment with vancomycin were excluded.

Results A total of 62 patients were included, 39 with previous surgical intervention (SI) in the CNS.

The most common diagnoses in the group with prior SI were bacterial meningitis (51%), fistula of cerebrospinal fluid (CSF) (21%) and shunt infection (21%). All had baseline neurological disease – neoplasms (46.2%) and subarachnoid haemorrhage (25.6%).

Most patients without prior SI (n = 23) were diagnosed with bacterial meningitis (n = 21) and just 2 with a brain abscess. The infective pathogen was isolated in 39 samples of CSF. All isolated microorganisms were sensitive to vancomycin. 63.7% of the isolated microorganisms were coagulase-negative Staphylococcus, with a MIC = 2 in 23.7%.

The initial and adjusted mean doses of vancomycin were 35.6 ± 9.3 mg/kg/day and 39.9 ± 15.2 mg/kg/day respectively. The median initial and adjusted Cmin were 10.04 (6.16) mcg/ml and 14.67 (3.66) mcg/ml respectively.

Laboratory-confirmed CSF clearance was obtained in 26 of the 39 isolates, 73.1% during the first 10 days of treatment.

The overall mortality was 5.8%, but only one death was related to the CNS infection.

Although Cmin above 20 mcg/ml was recorded in 15 patients, none developed nephrotoxicity.

Conclusion Vancomycin is still an agent of choice for CNS infections. Vancomycin trough concentrations of 15–20 mcg/ml are recommended to achieve clinical effectiveness for CNS infections without causing nephrotoxicity.

References and/or acknowledgements No conflict of interest.