Empiric treatment of healthcare-associated central nervous system infections in Denmark: do we need carbapenems?

Abstract

Background

Carbapenems are widely used for empiric treatment of healthcare-associated central nervous system (CNS) infections. We investigated the feasibility of a carbapenem-sparing strategy, utilising a third-generation cephalosporin (ceftriaxone or cefotaxime) (combined with vancomycin) for the empirical treatment of healthcare-associated CNS infections in Eastern Denmark.

Methods

The departments of neurosurgery and neuro-intensive care at Copenhagen University Hospital Rigshospitalet. First, we analysed local microbiological data (1st January 2020–31st August 2022) to identify microorganisms non-susceptible to third-generation cephalosporin. Subsequently, we assessed all carbapenem prescriptions over a three-month period for their indication and justification.

Results

In total, 25,247 bacterial cultures were identified, of which 2,563 CNS-related, were included in the analysis. The positivity rate was 10.5% (n = 257/2439) for cerebrospinal-fluid samples and 75.8% (n = 95/124) for brain parenchyma. CNS samples from five individual patients revealed bacteria non-susceptible to third generation cephalosporins (Enterobacter spp. (n = 3), Pseudomonas spp. (n = 2), Klebsiella spp. (n = 2), Citrobacter freundii (n = 1)). All five patients had been hospitalised for ≥10days at the time-point of antibiotic therapy. Out of 11,626 sets of blood cultures, a total of 10 individual patients had Gram-negative blood-stream infections with resistance to ceftriaxone and piperacillin/tazobactam. 140 days-of-therapy (32%) with carbapenem in 18 patients (36%) were definitively or possibly indicated according to guidelines, none were indicated for healthcare-associated CNS-infections.

Conclusion

An empiric treatment strategy relying on a third-generation cephalosporin appears suitable for healthcare-associated CNS infections at our tertiary hospital, serving a population of 2.6 million. However, in patients with prolonged hospitalization (≥10 days), immunosuppression, prior broad-spectrum antibiotic use, or history of resistant Gram-negative bacteria, empirical prescription of carbapenem may be needed.

Keywords:

• Antimicrobial stewardship
• CNS-infections
• Denmark
• Carbapenem-sparing
• Cephalosporin

Author contributions

Conceptualisation: CK, FBH, BRO, KM, ÅBA, JDK; Data analysis: CK, FH, JK; writing/draft: CK; data interpretation: all authors; writing editing/review: all authors.

Disclosure statement

No potential conflict of interest was reported by the author(s).