Lumbar drain infection rates: a comprehensive risk factor analysis from a multicenter retrospective study of 1000+ cases
No Thumbnail Available
Authors
Baig Mirza, Asfand
Velicu, Maria Alexandra
Vastani, Amisha
Fayez, Feras
Georgiannakis, Ariadni
Rashed, Sami
Sharma, Chaitanya
El Sheikh, Mustafa
MacCormac, Oscar
Lam, Pak Yin
Check for full-text access
Issue Date
2026
Type
Article
Language
Keywords
Alternative Title
Abstract
BACKGROUND AND OBJECTIVES: Lumbar drain (LD) insertion is a common cerebrospinal fluid (CSF) diversion method in neurosurgery; however, infection remains a major complication with significant morbidity. We evaluated the incidence, etiology, and associated risk factors of LD infection across 4 neurosurgical units over 15 years. METHODS: This retrospective multicenter cohort study included all adults requiring a LD between January 2009 and February 2024. Demographic, clinical, and microbiological characteristics were analyzed. LD infections were defined by positive CSF cultures and clinical symptoms. Risk factors were assessed by multivariate logistic regression analysis using IBM SPSS®. RESULTS: A total of 1017 patients required a LD, and the overall infection rate was 11.4% (116 infections). Significant risk factors for LD infection identified by univariate analysis were preoperative use of oral steroids ( P < .001), previous CSF drainage ( P = .019), LD insertion for 2 or more days ( P = .001), out-of-hours surgery ( P = .008), and CSF leak at the operation site ( P = .007). Conversely, factors reducing the risk of infection were LD insertion during the primary surgery ( P = .015) and the reason for insertion ( P = .029). Multivariate analysis confirmed increased incidence of LD infection with oral steroid use ( P = .01), LD insertion >7 days after the primary surgery ( P = .019), no previous CSF drainage ( P = .029), LD removal ≥2 days ( P = .002), out-of-hours primary surgery ( P = .024), CSF leak from the LD puncture site >2 days after LD insertion ( P = .074), LD disconnection >3 days postinsertion ( P = .028), and bleeding from the LD puncture site >2 days after drain insertion ( P = .026). CONCLUSION: We report a large patient series evaluating the factors associated with LD infections across multiple neurosurgical subspecialties. To reduce infection risk, LDs should be inserted during primary surgery, kept for the shortest duration, and promptly removed if disconnected, avoiding unnecessary sampling.
Description
Citation
Publisher
License
Journal
Neurosurgery
Volume
98
Issue
3