Systematic review and meta-analysis of intraventricular antibiotics for neonatal meningitis and ventriculitis

Abstract

Purpose: We aimed to determine the safety and effectiveness of intraventricular antibiotics in neonates with meningitis and/or ventriculitis and analyze the quality of available evidence.

Methods: DESIGN: Systematic review and meta-analysis.

Data sources: PubMed, EMBASE, LILACS, and SCOPUS up to 17 February 2023.

Eligibility criteria for selecting studies: Randomized experimental and observational studies were included. The Cochrane methodology was used for systematic reviews.

Results: Twenty-six observational studies and one randomized clinical trial involving 272 patients were included. The risk of bias in both pediatric and neurosurgical studies was high, and the quality of evidence was low (evidence level C). In the pediatric studies, no significant differences in mortality were found between intraventricular antibiotics and only systemic antibiotic [25.4% vs 16.1%, OR = 0.96 (0.42-2.24), P = 0.93]. However, when analyzing the minimum administered doses, we found a lower mortality when a minimum duration of 3 days for intraventricular antibiotics was used compared to only systemic antibiotic [4.3% vs 17%, OR = 0.22 (0.07-0.72), P = 0.01]. In the neurosurgical studies, the use of intraventricular antibiotics in ventriculitis generally results in a mortality of 5% and a morbidity of 25%, which is lower than that in cases where intraventricular antibiotics were not used, with an average mortality of 37.3% and a morbidity of 50%.

Conclusion: Considering the low quality of evidence in pediatric and neurosurgical studies, we can conclude with a low level of certainty that intraventricular antibiotics may not significantly impact mortality in neonatal meningitis and ventriculitis. However, reduced mortality was observed in cases treated with a minimum duration of 3 days of intraventricular antibiotic, particularly the multidrug-resistant or treatment-refractory infections. Higher-quality studies are needed to improve the quality of evidence and certainty regarding the use of intraventricular antibiotics for treating neonatal meningitis and ventriculitis.

Keywords: Intraventricular antibiotic; Meningitis; Neonates; Ventriculitis.