RESUMEN
OBJECTIVES: The objective of this review was to methodologically identify, appraise, and synthesise the primary research reporting the effectiveness of interventions to reduce ventriculostomy-associated infections in adult and paediatric neurosurgical patients with an external ventricular drain (EVD). REVIEW METHOD USED: Systematic review DATA SOURCES: A systematic search of five databases was conducted: MEDLINE, CINAHL Plus, Scopus, PubMed, and Cochrane Central. REVIEW METHODS: Key search terms and their variations included external ventricular drain and ventriculostomy-associated infection. The search was limited to studies published in English from 1980 to 2021. Screening, quality appraisal, and data extraction occurred in duplicate by the reviewers. The final search was conducted in June, 2021. RESULTS: A total of 11 699 records were identified from database searches. Fifty-three articles met inclusion criteria. Thirty-eight studies investigated individual interventions, and 15 investigated multi component interventions. Nineteen studies reported interventions effective in reducing the incidence of ventriculostomy-associated infections. For individual interventions, examples included the frequency of sampling of cerebrospinal fluid for biochemical markers, the use of silver-impregnated and antibiotic-impregnated EVDs, different insertion techniques, the application of 2-octyl cyanoacrylate dressings, and the administration of prophylactic intrathecal vancomycin antibiotics. For multi-component interventions (n = 15), examples included barrier precautions, EVD routine exchanges, frequency of cerebrospinal fluid sampling, and impregnated EVDs. CONCLUSIONS: Fifty-three studies were included in this review, and 19 reported single-component or multi component interventions effective in reducing ventriculostomy-associated infection in patients with an EVD. The use of antibiotic- impregnated and silver-impregnated EVDs was reported to be most effective and the majority of these studies were assessed as having the lowest risk of bias across the individual interventions.