Abstract
BACKGROUND AND OBJECTIVES:
External ventricular drainage (EVD) care bundle approaches are associated with reduced infection rates after EVD insertion. However, awareness and standardization of such protocols remain limited, and the current literature is sparse and has a high variability in analysis. We conducted a systematic review and meta-analysis of appropriate studies to identify the key components of an effective EVD care bundle for reducing EVD-related infections (ERIs).
METHODS:
A systematic review of the literature was conducted to identify any studies presenting patient cohorts undergoing EVD placement and reporting the introduction of bundle care protocols including postoperative infection rates. Major scientific databases (PubMed/MEDLINE, EMBASE, and Scopus) were systematically searched for studies published up to December 31, 2024. Studies were screened, and data were extracted independently by 2 authors.
RESULTS:
Twenty-eight studies were included for analysis. ERI rates before and after the implementation of care bundle protocols were observed. Different care bundles were identified and categorized as follows: insertion bundles, manipulation bundles, dressing bundles, and others. Although the literature concerning the implementation of care bundles for ERI is deemed poor, our results show a significant reduction in the postoperative ERI rates; a meta-analysis of the studies comparing cohorts before and after the implementation of care bundle protocols reveals significant differences in the postoperative infection rate after EVD placement (odds ratio 3.70, 95% CI 2.89-4.74, P < .0001 for the overall effect, heterogeneity I2 = 46%).
CONCLUSION:
Implementation of care bundle approaches for EVD insertion is associated with a significant reduction in ERIs. The implementation of care bundle approaches and the strict knowledge, training, and adherence to these protocols permit a standardization of the procedure and a better postoperative management diminishing complication rates. The available evidence supports the broader adoption of care bundle protocols as a new standard of care in clinical practice.
KEY WORDS: External ventricular drainage, Ventriculostomy, Care bundle, Infection, External ventricular drainage related infection, Meningitis
ABBREVIATIONS:
- ERI
EVD-related infection
- EVD
external ventricular drainage
- RCT
randomized controlled trial.
The operative insertion of an external ventricular drainage (EVD) is a common neurosurgical procedure used for intracranial pressure monitoring and urgent cerebrospinal fluid (CSF) diversion. It is routinely applied in conditions such as intraventricular or subarachnoid hemorrhage and traumatic brain injury1,2 or treatment of central nervous system infections.2-5
EVD-related infections (ERIs) are a relatively frequent complication after EVD placement, with incidence rates ranging between 0% and 27%,6 which can be as high as 32% in low- and middle-income countries7 and are associated with severe patient morbidity, mortality, and prolonged hospital stays.8-10 Several risk factors associated with ERIs have been identified including duration of EVD, CSF leak, and excessive sampling, as well as EVD manipulation and periodical reinsertion.6,10,11 Attempts to reduce the incidence of ERIs have included the use of antibiotic-coated catheters,12,13 diminishing the duration of drainage or changing the location of the EVD periodically,14,15 or increasing the length of the subcutaneous tunnel from the site of placement to the site of externalization.16,17 Some of these interventions have been widely adopted by the neurosurgical community, whereas others require further study.
As it remains poorly understood which factors significantly contribute to the appearance of ERIs, standardized ERI prevention protocols and strategies are lacking.1 One solution is the introduction of a care bundle. A care bundle in a healthcare setting is defined as a small, structured set of 3 to 5 evidence-based interventions or practices that are grouped together and applied collectively to manage a specific condition or patient population. Their key features include evidence-based individual elements, consistent collective application, measurability, targeted use, and improved outcomes when applied reliably.
The published literature regarding EVD care bundle approaches strongly suggests that the implementation of the standardized protocol greatly reduces the infection rates after EVD placement.18,19 However, care bundle approaches remain relatively unknown and underutilized,8 and published protocols exhibit a high variability in the measures taken; there is ultimately no clear consensus regarding which elements are essential for an adequate bundle approach.20 Hence, establishing a standardized, broadly adopted protocol would help ensure consistent and safe EVD placement and improve overall infection prevention and management.21 Our aim is to systematically review the literature on the role and the advantages of EVD care bundle approaches and to identify and summarize the main factors in successful ERI reduction.
METHODS
Overview
A systematic review of the literature was conducted to identify controlled studies assessing the effectiveness and/or safety of bundle approaches to reduce infections in adult patients with EVD. Commonly reported aspects included perioperative preparation, insertion technique, manipulation, antibiotic use, and CSF sampling. The clinical outcome investigated was infection number (8) and rate (9). Screening of titles and abstracts, full-text assessment, and data extraction were performed independently by 2 reviewers (M.T.F.N. and V.P.) using Covidence systematic review software (Veritas Health Innovation). Disagreements at any stage were resolved by discussion and consensus. Persistent inconsistencies were clarified with a third reviewer (V.E.S.). This systematic review followed the methodological framework described by Arksey and O′ Malley for systematic reviews and the Preferred Reporting Items for Systematic reviews and Meta-Analyses statement. This systematic review was not registered in any systematic review database.
Search Strategy
The PubMed/MEDLINE, EMBASE, and Scopus databases were searched to identify eligible articles. The search strategy was as follows: (“external ventricular drain” OR “external ventricular drainage” OR “ventriculostomy” OR “EVD”) AND (“bundle” OR “package” OR “best practice” OR “quality improvement” OR “strategy” OR “evidence-based” OR “protocol”).
Where appropriate, searches were expanded to include relevant word variants and exploded medical subject headings. Reference lists of relevant articles were also manually reviewed to identify additional eligible studies. The search included articles published between 2000 and 2024 in order for included studies to represent the contemporary healthcare environment.
Study Selection
Only in vivo studies in English were considered, involving adult patients. Only controlled studies comparing a bundle/care package/best practice guideline/infection reduction protocol with standard care were considered for inclusion, including retrospective and prospective cohort studies, quasi-experimental designs, and randomized controlled trials (RCTs). Systematic reviews, case reports, and single-group studies reporting data exclusively on bundle of care or standard care were excluded. To be eligible for inclusion, studies had to report the rate of ERI along with a clear definition of what was considered as ERI. This was the primary end point.
Data Extraction and Quality Assessment
The following data were extracted from all included publications: Study design and year of publication, number of patients, mean patient age and sex distribution, indication for EVD insertion, details of the bundle of care for infection reduction, use of antibiotic-coated catheters or not, and infection rate along with its definition. Secondary outcomes such as length of stay, costs, mortality, and need for permanent CSF shunting were recorded when available. The methodological quality of the included studies was graded using the Newcastle-Ottawa Quality Assessment Scale for Cohort Studies and the Cochrane Risk-of-Bias tool for RCTs.
Statistical Meta-Analysis
Given the expected clinical and methodological heterogeneity and the limited availability of large randomized trials, random-effects models were used for all meta-analyses. Odds ratios for dichotomous outcomes were calculated using Mantel-Haenszel tests, whereas continuous outcomes were analyzed using inverse variance models. Statistical heterogeneity was assessed using Cochran's Q test and quantified with the I2 statistic, with a P-value of ≤0.1 used as a cutoff for relevant heterogeneity. Reporting bias was assessed by visual inspection of funnel plots. The meta-analyses were performed using R statistical software.22 P values ≤ .05 in 2-tailed tests were considered statistically significant for the assessment of the overall effect.
RESULTS
Literature Search
The results of the literature search and study selection are illustrated in the Preferred Reporting Items for Systematic reviews and Meta-Analyses flowchart (Figure 1). Studies included for quantitative meta-analysis reported ERI rates before and after introducing EVD care bundles. For studies reporting multiple postimplementation ERI rates (eg, different time points or after adjustments to the care bundle), we combined these data into a single postimplementation ERI rate per study to facilitate meta-analysis.23-27
FIGURE 1.
Preferred Reporting Items for Systematic reviews and Meta-Analyses flowchart.
Included Study Characteristics and Quality
Table 1 summarizes the basic characteristics of the included studies and their quality assessment, whereas the description and categorization of the different bundles are detailed in Table 2. The outcomes regarding ERIs are shown in Table 3. Quality assessment of the abovementioned studies was organized into basic study characteristics, introduced care bundles, and definitive outcomes regarding ERIs.
TABLE 1.
Basic Study Characteristics
| Studies | Study design | Journal | No. of patients | Male number and % | Age, years, mean (range or SD) | Indication for EVD | Pathology | EVD-duration days | Quality assessment a |
|---|---|---|---|---|---|---|---|---|---|
| Hoefnagel et al28 | Quasi-experimental; prospective cohort for bundle and retrospective for control | BioMed Central Neurology | 409 patients. Preprotocol: 228 Postprotocol: 181 |
Preprotocol: 50%; postprotocol: 44% | Preprotocol: 60 (49-70); postprotocol: 56 (47-66) | Obstructive hydrocephalus | SAH/IVH (pre: 48% (109), post: 60% (109)). Tumor (pre: 10% (23), post: 12% (21)). Trauma (pre: 3% (7), post: 9% (16)). Other (pre: 39% (89), post: 19% (35)) | Preprotocol: median 5.5 (2-12), postprotocol: 7 (3-12) | 4/2/3 |
| Katzir et al29 | Retrospective, cohort | World Neurosurgery | 142 patients; preprotocol group (Group A: Elective EVD replacement): 43 patients Postprotocol group (Group B: Clinically indicated EVD replacement): 99 patients aPatients presenting with preinfected CSF confirmed by CSF analysis were excluded as well as those with a positive CSF culture from the first EVD, and patients with any CSF leak |
Group A: 25 (58%); Group B: 56 (57%) | Group A: 48.8 ± 17.7 Group B: 51.9 ± 17.4 |
Obstructive/malresorptive hydrocephalus | Group A: Head injury: 8 (19%), shunt malfunction: 5 (12%), aneurysmal SAH: 11 (26%), IVH: 5 (12%), ICH: 7 (16%), tumor with HCP: 3 (7%), stroke with HCP: 3 (7%), obstructive HCP: 1 (2%) Group B: Head injury: 20 (20%), shunt malfunction: 6 (6%), aneurysmal SAH: 25 (25%), IVH: 10 (10%), ICH: 15 (15%), tumor with HCP: 16 (16%), stroke with HCP: 2 (2%), obstructive HCP: 5 (5%) |
Median Group A: 14 (12-24); Group B: 8 (5-12) | 4/2/3 |
| Piccirilli et al30 | Retrospective, cohort | World Neurosurgery | 196 EVD placed in 161 patients. Preprotocol group A: 86; postprotocol group B: 75; aexcluded patients with ETV and postinfectious hydrocephalus | N/A | Subgroup A: 66.8 ± 7.8 y Subgroup B: 65.3 ± 7.7 y |
Obstructive hydrocephalus | Group A: SAH: 36 (41.9%), ICH: 17 (19.8%), TBI: 13 (15.1%), tumor: 10 (11.6%), others: 10 (11.6%); Group B: SAH: 29 (38.7%), ICH: 15 (20.0%), TBI: 11 (14.7%), tumor: 9 (12.0%), others: 11 (14.7%) | Group A: 12.7 (±3.8, 7-24) Group B: 12.5 (±3.6, 9-21) |
4/2/3 |
| Chatzi et al19 | Prospective observational | Critical Care Medicine | 139 preprotocol group: 82 patients Postprotocol group: 57 patients; aexclusion of patients with a previous EVD-associated ventriculitis diagnosis and previous EVD placement |
N/A | Preprotocol group: 53.5 (50.1-56.9) Postprotocol group: 50.3 (47.8-56.4) |
Obstructive/malresorptive hydrocephalus | Preprotocol group: SAH: 26 (31.7%), brain trauma: 11 (13.4%), ICH: 42 (51.2%), tumor surgery: 3 (3.7%) Postprotocol group: SAH: 20 (35.1%), brain trauma: 11 (19.3%), ICH: 23 (40.3%), tumor surgery: 3 (5.3%) |
Preprotocol group: 15.6 Postprotocol group: 15.2 |
4/2/3 |
| Omar et al7 | Prospective observational | Malaysian Journal of Medical Sciences | 87 (Group A: Tunneling <5 cm: 35; Group B: Tunneling >5 cm: 52) aexclusion of CSF leakage and CNS infection before the procedure | Total: 40 males (46.0%) | 50 (14-73) | Obstructive hydrocephalus | Hypertensive bleeding with intraventricular extension associated with hydrocephalus: 28 patients (32.2%) SAH: 25 patients (28.7%) ICP monitoring in trauma cases: 20 patients (23.0%) Brain tumor and other factors (including posterior fossa and supratentorial ischemia or infarct): least common indications |
<10 d: 51 >10 d: 36 |
4/1/3 |
| Williams et al31 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Journal of Neurosurgery | 382 preprotocol group: 206. Postprotocol group: 176 aPatients with pre-existing CNS infections were excluded |
Preprotocol group: 61% Postprotocol group: 59% |
Preprotocol group: 47 ± 18.3 Postprotocol group: 45 ± 18.7 |
Obstructive hydrocephalus | Preprotocol group: SAH: 87 (42%), traumatic head injury: 60 (29%), ICH: 29 (14%), other: 30 (15%) Postprotocol group: SAH: 58 (33%), traumatic head injury: 63 (36%), ICH: 25 (14%), other: 30 (17%) |
Median preprotocol group: 4.9 (2.6-7.9) Postprotocol group: 5.8 (2.7-8.1) |
4/2/3 |
| Choo et al32 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Journal of Korean Medical Science | 183 (preprotocol group: 84, postprotocol group: 99) aPrevious CNS infections among those with EVD removal <3 d were excluded |
Preprotocol group: 42 (50%) Postprotocol group: 41 (41.4%) |
Mean preprotocol group: 57 (17-88) Postprotocol group: 59 (16-88) |
Obstructive/malresorptive hydrocephalus | Preprotocol group: ICH/IVH: 37 (44.0%), SAH: 22 (26.2%), tumor/mass: 16 (19.0%), TBI: 3 (3.6%), other: 6 (7.1%) Postprotocol group: ICH/IVH: 46 (46.5%), SAH: 29 (29.3%), tumor/mass: 22 (22.2%), TBI: 0 (0%), other: 2 (2%) |
Median preprotocol: 11.61 (3-47) Postprotocol group: 12.59 (3-64) |
4/2/3 |
| Talibi et al33 | Quasi-experimental; prospective cohort for bundle and retrospective for control | British Journal of Neurosurgery | 275 preprotocol group: 120 Postprotocol group: 155 aexcluded EVD preinfections and reinsertions because of malpositioning |
Preprotocol group: 69 (58%) Postprotocol group: 86 (55%) |
Median preprotocol group: 53 (43-64) Postprotocol group: 56 (44-67) |
Obstructive/malresorptive hydrocephalus | Preprotocol group: SAH 50 (42%), TBI: 35 (29%), ICH: 20 (17%), other: 15 (12%) Postprotocol group: SAH 65 (42%), TBI: 45 (29%), ICH: 25 (16%), other: 20 (13%) |
Median preprotocol group: 9 (5-15) Postprotocol group: 11 (6-16) |
4/1/3 |
| Thien et al23 | Quasi-experimental; prospective cohort for bundle and retrospective for control | World Neurosurgery | 551 preprotocol group (2014): 165 patients Postprotocol group (2015): 189 patients Postprotocol group (2016): 197 patients aBy providing separate data for each year postprotocol implementation, the study aims to show whether the new practices and procedures continued to be effective and possibly improved further as they became more established and refined over time |
Preprotocol: 2014: 96 (58.2%) Postprotocol: 2015: 109 (57.7%) 2016: 111 (56.3%) |
Preprotocol: 2014: 57 ± 14 2015:56 ± 16 2016: 57 ± 16 |
Obstructive/malresorptive hydrocephalus | Preprotocol 2014: SAH: 51 (30.9%), IVH: 18 (10.9%), ICH: 55 (33.3%), tumor: 13 (7.9%), TBI: 4 (2.4%), other: 24 (14.5%) Postprotocol 2015: SAH: 58 (30.7%), IVH: 27 patients (14.3%), ICH: 56 (29.6%), tumor: 21 (11.1%), TBI: 5 (2.6%), other: 22 (11.6%) Postprotocol 2016: SAH: 51 (25.9%), IVH: 44 (22.3%), ICH: 41 (20.8%), tumor: 9 (4.6%), TBI: 5 (2.5%), other: 47 (23.9%) |
Preprotocol 2014: 11 ± 8 Postprotocol 2015: 10 ± 6 Postprotocol 2016: 10 ± 5 |
4/2/3 |
| Levy et al34 | Quasi-experimental; prospective cohort for bundle and retrospective for control | American Journal of Infection Control | 310 preprotocol group: 129 postprotocol group: 181; aexcluded previous CNS infections, outside hospital EVD placement Pregnancy, ICU length of stay <48 h |
Preprotocol group: 55 (43%) Postprotocol group: 63 (35%) |
Preprotocol group: 55.99 Postprotocol group: 58.85 |
Obstructive/malresorptive hydrocephalus | Preprotocol group: SAH: 73 (57%), ICH: 27 (21%), tumor: 8 (6%), IVH: 5 (4%), other: 16 (12%) Postprotocol group: SAH: 103 (57%), ICH: 38 (21%), tumor: 11 (6%), IVH: 7 (4%), other: 22 (12%) |
Median preprotocol group: 12.2 Postprotocol group: 16.2 |
4/1/3 |
| Dasgupta et al24 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Journal of Neurosurgery | 291 preprotocol group: 200 postprotocol group 2016: 43 (67 EVDs) Postprotocol group (2017): 48 (59 EVDs) |
Preprotocol group: N/A Postprotocol group (2016): 21 (49%) Postprotocol group (2017): 28 (58%) |
Postprotocol 2016: 54 ± 15 Postprotocol group 2017: 54 y ± 18 |
Obstructive/malresorptive hydrocephalus | Preprotocol group: N/A Postprotocol group (2016): ICH: 37 (55%), tumor: 6 (9%), other: 10 (15%), infection: 14 (21%) Postprotocol group (2017): ICH: 32 (54%), tumor: 9 (15%), other: 9 (15%), infection: 9 (15%) |
Preprotocol group: N/A Postprotocol group (2016): 9.1 ± 5.3 Postprotocol group (2017): 8.5 ± 4.9 |
4/1/3 |
| Whyte et al35 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Clinical Neurology and Neurosurgery | 145 preprotocol: 91; postprotocol: 54; apre-existing infections were excluded | Preprotocol: 54 (59.3%) Postprotocol: 30 (55.6%) |
Preprotocol: 55.6 ± 16.1 Postprotocol: 60.5 ± 17.3 |
Obstructive/malresorptive hydrocephalus | Preprotocol: ICH 23 (25.3%), SAH: 17 (18.7%), SDH: 16 (17.6%), TBI: 12 (13.2%), tumor: 12 (13.2%), stroke: 6 (6.6%), other: 5 (5.5%) Postprotocol: ICH: 13 (24.1%), SAH: 12 (22.2%), SDH: 16 (29.6%), TBI: 3 (5.6%), tumor: 6 (11.1%), stroke: 1 (1.9%), other: 3 (5.6%) |
Median preprotocol: 7 (IQR 7) Postprotocol:9 (IQR 7) |
4/2/3 |
| Flint et al36 | Retrospective | Neurosurgery | 262 preprotocol: 143; postprotocol: 119 | N/A | Median preprotocol: 61 (50-73) Postprotocol group: 61 (48-72) |
Obstructive/malresorptive hydrocephalus | Preprotocol: ICH/IVH: 52 (36.4%), SAH: 63 (44.1%), tumor: 23, (16.1%), TBI: 1 (0.7%), other: 4 (2.8%) Postprotocol: ICH/IVH: 48 (40.3%), SAH: 45 (37.8%), tumor: 10 (8.4%), TBI: 5 (4.2%), other: 11 (9.2%) |
Median preprotocol: 8 (4-12) Postprotocol: 11 (4-15) |
4/1/3 |
| Champey et al37 | Retrospective, multicenter | Journal of Neurosurgery | 462 patients Preprotocol: 241 (Saint-Etienne and Marseille combined); Postprotocol: 221 (Grenoble); aexcluded patients with pre-existing infections | Preprotocol: 136 (56%) Postprotocol: 108 (49%) |
Median preprotocol: 58 (47-66) Postprotocol: 58 (46-67) |
Obstructive/malresorptive hydrocephalus | Preprotocol: SAH: 120 (50%), ICH: 58 (24%), TBI: 18 (7%), other: 45 (19%) Postprotocol: SAH: 122 (55%), ICH: 50 (23%), TBI: 27 (12%), other: 22 (10%) |
Median preprotocol: 12 (6-18) Postprotocol: 14 (9-23) |
4/2/3 |
| Kubilay et al18 | Prospective, observational | Journal of Neurosurgery | 2928 preprotocol: 217 (estimated from the infection rate); postprotocol: 2711; aprevious infections were excluded | N/A | 56 | Obstructive/malresorptive hydrocephalus | Hydrocephalus, elevated intracranial pressure, ICH, and intrathecal administration of medications | 8-14 | 4/1/3 |
| Huang et al38 | Retrospective, multicentre | Neurocritical Care | 742 preprotocol: 446 Postprotocol: 296; aprevious infections were excluded |
362 (48.8%) | 61.4 ± 15.4 | Obstructive/malresorptive hydrocephalus | ICH: 347 (46.9%), aneurysmal SAH: 229 (30.9%), tumor: 105 (14.2%), AVM: 34 (4.6%), trauma: 57 (7.7%) | 12.7 ± 1.2 | 4/2/3 |
| Lwin et al25 | Prospective observational | Singapore Medical Journal | 234 preprotocol (Phase I): 82. Postprotocol (Phase II): 79. Postprotocol (Phase III): 73 | 152 (65%) | 31-69 | N/A | N/A | Preprotocol 8.6. Postprotocol Phase II 7.4. Postprotocol Phase III 7.3 | 4/1/3 |
| Lackner et al39 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Neurocritical Care | 39 preprotocol: 20; postprotocol: 19a 1 patient was excluded because of EVD replacement | N/A | Preprotocol: 62 y ± 11 Postprotocol: 59 y ± 14 |
Obstructive/malresorptive hydrocephalus | Preprotocol: SAH: 12 (60%), cerebellar infarction: 4 (20%), ICH: 4 (20%) Postprotocol: SAH: 10 (52.6%), cerebellar infarction: 4 (21.1%), ICH: 3 (15.8%), tumor: 1 (5.3%), SDH: 1 (5.3%) |
Preprotocol: median 15.5 (5.5) Postprotocol group: median 14 (5) |
4/1/3 |
| Flint et al26 | Retrospective | World Neurosurgery | 451 preprotocol (baseline, 2005-2007): 143 Postprotocol (initial 3-y period, 2009-2011): 119 Postprotocol (4-y follow-up, 2012-2015): 189 EVD in 173 patients |
N/A | Preprotocol: Median 61 (50-73) Postprotocol 3-y period: 61 (48-72) Postprotocol 4-y period: 59 (49-69) |
Obstructive/malresorptive hydrocephalus | Preprotocol: ICH/IVH: 52 (36.4%), SAH: 63 (44.1%), tumor: 23 (16.1%), TBI: 1 (0.7%), other: 4 (2.8%) Postprotocol 3-y period: ICH/IVH: 48 (40.3%), SAH: 45 (37.8%), tumor: 10 (8.4%), TBI: 5 (4.2%), other: 11 (9.2%) Postprotocol 4-y period: ICH/IVH: 74 (42.8%), SAH: 77 (44.5%), tumor: 11 (6.4%), TBI: 3 (1.7%), other: 8 (4.6%) |
Preprotocol median 8 (4-12) Postprotocol 3-y period: 11 (4-15) 4-y period: 12 d (8-15) |
4/1/3 |
| Phan et al40 | Retrospective | Journal of Clinical Neuroscience | 153 preprotocol (2006-2007): 43 Postprotocol (2007-2010): 110 |
Preprotocol: 14 (32.6%) Postprotocol: 74 (67.2%) |
Preprotocol: 56 (19-87) Postprotocol: 53 (12-81) |
N/A | N/A | Preprotocol: infection: 14.9 ± 7.5, noninfection: 7 ± 6.6 Postprotocol: infection: 7.8 ± 3.8, noninfection: 7.7 ± 3.8 |
4/2/3 |
| Hong et al8 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Acta Neurochirurgica | 349 preprotocol: 141 Postprotocol: 208 |
Preprotocol: 77 (54.6%) Postprotocol: 96 (46.2%) |
Preprotocol: 48.6 ± 21.6 Postprotocol: 50.9 ± 21.2 |
Obstructive/malresorptive hydrocephalus | Preprotocol: SAH: 65 (46.1%), tumor: 39 (27.7%), TBI: 14 (9.9%), other: 22 (14.9%) Postprotocol: SAH: 142 (68.3%), tumor: 44 (21.2%), TBI: 7 (3.4%), other: 14 (6.8%) |
Preprotocol: 18.8 ± 13.0 Postprotocol: 14.3 ± 12.1 |
4/1/3 |
| Rojas-Lora et al41 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Annals of Clinical Microbiology and Antimicrobials | 198 preprotocol: 87 Postprotocol: 111 |
Preprotocol: 40 (46%) Postprotocol: 52 (47%) |
Preprotocol: 53 ± 14 Postprotocol: 57 ± 15 |
Obstructive/malresorptive hydrocephalus | Preprotocol: SAH: 35 (40%), ICH: 13 (15%), tumor: 20 (23%), TBI: 4 (5%), other: 15 (17%) Postprotocol: SAH: 50 (45%), ICH: 34 (31%), tumor: 8 (7%), TBI: 4 (4%), other: 15 (13%) |
Median preprotocol: 9 (4-15) Postprotocol: 12 (7-15) |
4/2/3 |
| Harrop et al27 | Prospective observational | Neurosurgery | 1961 Preprotocol (Period 1): 327 Postprotocol (Period 2): 281 Postprotocol catheter A (Period 3): 195 Postprotocol without antibiotic catheters (Period 4): 157 Postprotocol with catheter B (Period 5): 1001 |
N/A | N/A | N/A | N/A | N/A | 4/1/3 |
| Dasic et al42 | Quasi-experimental; prospective cohort for bundle and retrospective for control | British Journal of Neurosurgery | 113 Preprotocol: 41 patients, 51 EVD; postprotocol: 54 patients, 62 EVD | Preprotocol: 27 Postprotocol: 32 |
Preprotocol: 53 Postprotocol: 56 |
N/A | N/A | Preprotocol: 11.4 Postprotocol: 10.2 |
4/1/3 |
| Lozano et al43 | Retrospective, cohort | Cureus | 58 preprotocol: 33 Postprotocol: 25 |
Preprotocol: 24 (72.73%) Postprotocol: 14 (56.00%) |
Preprotocol: 48.45 ± 20.11 Postprotocol: 48.24 ± 17.59 |
N/A | N/A | Preprotocol: 17.61 ± 16.64 Postprotocol: 10.68 ± 8.75 |
4/0/3 |
| Korinek et al44 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Acta Neurochirurgica | 306 preprotocol: 131 patients (161 EVDs) Postprotocol: 175 patients (216 EVDs) |
Preprotocol: 79 (60%) Postprotocol: 114 (65%) |
Preprotocol: 49 ± 16 Postprotocol: 44 ± 17 |
Obstructive/malresorptive hydrocephalus | Severe head trauma: 43.9% SAH/IVH: 37.6% Other: not specified |
Preprotocol: 10.2 ± 6.5 Postprotocol: 17.4 ± 8.6 |
4/2/3 |
| Rahman et al45 | Retrospective, cohort | Joint Commission Journal on Quality and Patient Safety | Preprotocol: 217; postprotocol: 2911 | N/A | N/A | N/A | N/A | N/A | 4/1/3 |
| Zakaria et al21 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Journal of Hospital Infection | 239 patients Preprotocol: 159 patients, 234 EVD Postprotocol: 107 patients, 132 EVD |
Preprotocol: 82 (51.6%) Postprotocol: 52 (48.6%) |
Preprotocol: 52.1 Postprotocol: 50.3 |
N/A | N/A | Preprotocol: 9.7 for noninfected, 22.1 for infected Postprotocol: 10.2 for noninfected, 22.6 for infected |
4/0/3 |
CSF, cerebrospinal fluid; EVD, external ventricular drainage; ETV, endoscopic third ventriculostomy; HCP, hydrocephalus; ICH, intracerebral hematoma; ICP, intracranial pressure; IVH, intraventricular hemorrhage; N/A, not available; SAH, subarachnoid hemorrhage; TBI, traumatic brain injury.
Quality Assessment according to the Newcastle-Ottawa Assessment Scale.
TABLE 2.
Bundles of Care
| Authors | Study design | Journal | No. of patients | Protocol innovations and key points of the study | Insertion protocol | Dressing protocol | Manipulation bundle | Other | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Standard | Bundle | Standard | Bundle | Standard | Bundle | Standard | Bundle | |||||
| Hoefnagel et al28 | Quasi-experimental; prospective cohort for bundle and retrospective for control | BioMed Central Neurology | 409 preprotocol: 228 Postprotocol: 181 |
EVD handling protocol, propensity score matching cohort | Preprotocol sterile techniquesa, antibiotic prophylaxis (flucloxacillin or amoxicillin/clavulanic acid), Rickham reservoir | Postprotocol sterile techniquesa, antibiotic prophylaxis (Cefazolin), tunneled EVD >5 cm, rigorous disinfectiona, and standardized CSF sampling protocola | No formal recommendation | Sterile dressing as long as EVD remains in situ, changed every 72 h | No formal recommendation | Sterile gloves, multiple disinfection moments before CSF tapping | No formal recommendation | Multidisciplinary meetings with infectiologists for the meningitis treatment strategy, use of sterile gloves and multiple disinfection moments for EVD flushing |
| Katzir et al29 | Retrospective, cohort | World Neurosurgery | 142; preprotocol group (Group A: Elective EVD replacement): 43 patients Postprotocol group (Group B: Clinically indicated EVD replacement): 99 patients |
Switch from routine EVD replacement in preprotocol every 5 d to EVD replacement only when clinically indicated. EVD catheters were replaced only in the case of device malfunction or migration, or the presence of laboratory values or clinical signs of meningitis. Emphasis on reducing the number of EVD replacements to decrease the risk of infection | Routine EVD replacement every 5 d Antibiotic prophylaxis with 1 g of cefazolin intravenously before insertion |
EVD replacement only when clinically indicated (in the case of device malfunction or migration or the presence of laboratory values or clinical signs of meningitis) | Semiocclusive adhesive dressing replaced every 48 h | Same as preprotocol | N/A | N/A | Attention to drain duration and reduction of unnecessary drain openings and replacements | |
| Piccirilli et al30 | Retrospective, cohort | World Neurosurgery | 196 EVD placed in 161 patients. Preprotocol group A: 86; postprotocol group B: 75 | Introduction of the “CLD-assisted” technique to prevent ventricular catheter displacement and infection; removal of EVD as soon as possible | Linear incision based on preoperative CT scan. Catheter inserted to a depth of approximately 5 cm until CSF was readily collected Catheter passed under the skin about 5 cm backward from the insertion site and secured to the skin using a nonabsorbable suture |
Curvilinear incision to create a skin flap for effortless catheter removal. Catheter slid through a biocompatible CLD using a disposable applier. Proper positioning of the CLD on the ventricular catheter based on preoperative CT scan. Final steps similar to the standard technique for securing the catheter | N/A | Site care with alcoholic chlorhexidine every 8 h, EVD insertion site left open without dressing unless oozing, daily site cleaning, and no routine use of occlusive dressings | N/A | Continuous monitoring and assessment by the infection control committee Use of a standardized EVD kit containing all necessary components for sterile placement |
N/A | N/A |
| Chatzi et al19 | Prospective observational | Critical Care Medicine | 139 preprotocol group: 82 patients Postprotocol group: 57 patients |
Re-education of ICU personnel on infection control related to EVD Meticulous EVD catheter handling, including hand hygiene and barrier precautions CSF sampling only when clinically necessary Routine replacement of the drainage catheter on the seventh drainage day if still necessary |
Standard sterile technique with perioperative antibiotic prophylaxis (ceftriaxone and teicoplanin given 30 min before and 6-8 h after insertion) | Similar to preprotocol with additional focus on meticulous catheter handling and routine replacement on the seventh day | N/A | N/A | N/A | Hand hygiene: Strict hand hygiene practices were enforced for all healthcare personnel handling the EVD. This included washing hands with soap and water or using an alcohol-based hand sanitizer before and after any contact with the patient or the EVD system Barrier precautions: When accessing the EVD system (eg, for CSF sampling or adjusting the drainage system), healthcare personnel were required to use full barrier precautions. This included wearing sterile gloves, a mask, a gown, and sterile drapes to create a sterile field around the EVD insertion site EVD handling: Covering: The EVD insertion site was covered with a sterile dressing. The dressing was checked regularly and changed if it became soiled or loose Dressing for the wound exit site: A sterile dressing was applied to the EVD exit site, and the dressing was replaced at regular intervals or when it became compromised Flushing and cleaning standards: The peripheral part of the EVD was flushed to prevent blockages. Flushing was performed using a sterile technique to avoid introducing infections. Cleaning standards were strictly followed to maintain sterility. This included using disinfectants to clean the area around the EVD and the equipment used for CSF sampling or EVD adjustments Connecting three-way valve: When opening the connecting three-way valve of the EVD system, healthcare personnel used sterile techniques. This involved disinfecting the valve with an appropriate antiseptic solution before accessing it CSF sampling: CSF samples were taken only when clinically necessary, aiming to minimize the frequency of sampling to reduce the risk of infection. When sampling was required, it was done using an aseptic technique to ensure sterility |
N/A | Isolation techniques for patients with multidrug-resistant bacteria, written antibiotic treatment protocol, continuous surveillance of nosocomial infections, daily oral decontamination with hexetidine, and no routine changes of respiratory circuits |
| Omar et al7 | Prospective observational | Malaysian Journal of Medical Sciences | 87 (Group A: Tunneling <5 cm: 35; Group B: Tunneling >5 cm: 52) | Tunneling the catheter for more than 5 cm under the scalp, from the burr hole to the exit site of the skin Duration of the ventricular catheterization limited to 10 d or less to reduce infection risk |
Group A: Tunneling <5 cm | Tunneling >5 cm | N/A | N/A | N/A | N/A | N/A | N/A |
| Williams et al31 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Journal of Neurosurgery | 382 preprotocol group: 206; postprotocol group: 176 | Change from daily CSF sampling to once every 3 d | EVD inserted in the operating room or ICU by a neurosurgeon using standard practice. Chlorhexidine 2% in alcohol 70% used for skin preparation Hair clipped or shaved around the burr hole. EVD catheter not tunneled |
Similar to preprotocol | Transparent, nonadherent, occlusive dressing covering the insertion site, changed every 3rd day or if dislodged | Same as preprotocol | CSF samples taken daily from the tubing aspiration bung | CSF drainage taken from the drainage bag once every 3 d | N/A | N/A |
| Choo et al32 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Journal of Korean Medical Science | 183 (preprotocol group: 84; postprotocol group: 99) | Skin preparation using 2% chlorhexidine instead of povidone–iodine Avoidance of routine prophylactic antibiotics for the entire period of catheterization; single dose of first-generation cephalosporin given before skin incision Application of transparent adhesive CHG gel pad instead of daily dressing, which was checked daily without breaking the seal and changed weekly if no abnormalities were observed |
Type and maintenance period of prophylactic antibiotics differed according to the attending physician | Skin preparation using 2% chlorhexidine instead of povidone–iodine Avoidance of routine prophylactic antibiotics; single dose of first-generation cephalosporin given before skin incision |
Daily dressing using povidone–iodine or chlorhexidine cotton balls and aseptic adhesive dressing pads | Application of transparent adhesive CHG gel pad checked daily and changed weekly if no abnormalities were observed | CSF sampling daily through a 3-way stopcock with sterile gloves without a full drape | EVD bag kept separate from the floor and maintained in a vertical position during transport Staff wore sterile gloves and a mask when emptying the chamber CSF profile test conducted daily on CSF in the chamber from the third postoperative day When CSF culture was required, the sample was collected from a 3-way stopcock at the access port, maintained aseptically under a closed system Once the port was removed, the cap was discarded and replaced with a new, sterile one Luer fitting inside the port cleaned multiple times with a disposable chlorhexidine swab CSF draw or flush performed using a strict sterile technique |
N/A | N/A |
| Talibi et al33 | Quasi-experimental; prospective cohort for bundle and retrospective for control | British Journal of Neurosurgery | 275 preprotocol group: 120 Postprotocol group: 155 |
Establishment of a protocol for aseptic CSF sampling including hand-washing, surgical scrub and preparation, and cleaning of EVD access ports. Implementation of an electronic clinical noting system to document all CSF sampling episodes | Techniques for EVD insertion and maintenance were at the discretion of the operating neurosurgeon with no standardized protocol apart from generic aseptic techniques | Standardized technique for aseptic CSF sampling including the hand-washing technique, use of surgical scrub, preparation, and cleaning of EVD access ports, along with gown, gloves, and mask | N/A | N/A | No standardized protocol | Standardized aseptic CSF sampling protocol including hand-washing, surgical scrub, preparation, and cleaning of EVD access ports, and documentation of all sampling episodes in the electronic clinical noting system | N/A | Implementation of an electronic clinical noting system to document all CSF sampling episodes |
| Thien et al23 | Quasi-experimental; prospective cohort for bundle and retrospective for control | World Neurosurgery | 551 preprotocol group (2014): 165 patients Postprotocol group (2015): 189 patients Postprotocol group (2016): 197 patients |
Introduction of a pragmatic risk stratification pathway Use of plain, silver-impregnated, and antibiotic-impregnated catheters based on risk stratification. Implementation of structured training and monitoring compliance through audits and meetings |
Disparate EVD handling protocols across different sites | EVD tunneling ≥5 cm | No standardized protocol across sites | Use of double gloves and changing outer gloves before handling EVD Application of Tegaderm for window dressing, changed if soiled, wet, or integrity breached |
No standardized protocol across sites | Aseptic technique for CSF sampling Routine sampling on Mondays and Thursdays (Lozier “A”) Minimal sampling unless clinically indicated (Lozier “B”) CSF sampling from the 3-way tap using the strict aseptic technique |
N/A | Implementation of an electronic clinical noting system to document all CSF sampling episodes |
| Levy et al34 | Quasi-experimental; prospective cohort for bundle and retrospective for control | American Journal of Infection Control | 310 preprotocol group: 129; postprotocol group: 181 | Addition of a 70% alcohol-impregnated Curos cap to the proximal EVD retrieval port | Standardized EVD placement using a sterile technique, 1 dose of cephalosporin or glycopeptide before skin incision, tunneling of all catheters after burr hole placement | Same as the preprotocol with the addition of a 70% alcohol-impregnated Curos cap to the proximal EVD retrieval port | Sterile adhesive dressing placed over the catheter insertion site | Same as the preprotocol | Default surveillance sampling every 3 d, unless specifically requested at more frequent intervals by a neurosurgery attending physician or neurointensivist EVD maintenance sheet filled out daily |
Same as the preprotocol | N/A | N/A |
| Dasgupta et al24 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Journal of Neurosurgery | 291 preprotocol group: 200; postprotocol group 2016: 43 (67 EVDs) Postprotocol group (2017): 48 (59 EVDs) |
Simulation workshops to train surgeons on the safe sampling technique, administration of intrathecal drugs, and a standardized operative technique | No specific care bundle | Silver-impregnated catheter used Use of prophylactic antibiotics as per hospital policy Minimal shaving with clippers only Preparation with alcoholic chlorhexidine left to air dry before draping Frontal burr hole at Kocher's point, perpendicular approach to the ventricle, catheter placement at a depth of 5.5 cm from the cortical surface. The personnel is trained using the simulation “Rowena head surgical model” Catheter tunneled for 10 cm and secured with an approved fixation device or modified Roman sandal knot |
N/A | Three layers of permeable, transparent, water-resistant, spray-on dressing (Opsite spray) | N/A | CSF sampling protocol: Hand hygiene before the procedure; use of sterile gloves; use of sterile equipment; preparation of the access site by careful disinfection (eg, alcoholic chlorhexidine); use of sterile drapes; minimization of manipulation; use of closed systems; ensuring that the procedure is performed by trained personnel through simulation workshops. Regularly monitoring the patient for any signs of infection or complications. Documentation: proper documentation of the procedure, including any deviations from the standard protocol and the reasons for them | N/A | N/A |
| Whyte et al35 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Clinical Neurology and Neurosurgery | 145 preprotocol: 91; postprotocol: 54 | Implementation of an EVD bundle to standardize nursing care Antibiotic duration changed to 1 preoperative dose for all neurosurgical drains |
N/A | N/A | N/A | Performed when dressing is loose or soiled. Sterile barrier used. Sterile hat and mask worn throughout the procedure. Clean gloves worn for removal of old dressing and sterile gloves worn for dressing change. Surrounding area cleaned with an antiseptic solution of chlorhexidine gluconate and isopropyl alcohol for 30 s and allowed to dry completely. Chlorhexidine-eluting patch placed over the catheter exit site. Benzoin tincture applied to surrounding skin and allowed to dry completely Transparent dressing placed over the catheter site and secured with sterile adhesive strip |
Antibiotics continued for the duration of drain placement | Single preoperative antibiotic dose. Sterile CSF sampling only by properly trained staff. Nursing critical skills assessed regularly by simulation workshops, competency checklists, supervised practice, peer review, regular audits to identify any deviation from the protocol, generation of key performance indicators, and continuous feedback via educational sessions | N/A | N/A |
| Flint et al36 | Retrospective | Neurosurgery | 262 preprotocol: 143; postprotocol: 119 | Hand washing, draping; wearing sterile gown and gloves, cap, and mask; chlorhexidine skin preparation; application of a sterile dressing; use of antibiotic-impregnated catheters | Techniques for EVD insertion and maintenance were up to the operating physician Nonantibiotic-impregnated catheters used |
Formal EVD infection control policies and procedures were adopted Use of antibiotic-impregnated catheters Comprehensive sterile technique including hand washing, chlorhexidine preparation, and sterile draping |
N/A | Catheter secured using surgical staples Chlorhexidine-eluting patch applied over the catheter exit site Medium-sized transparent dressing film applied, secured with sterile adhesive strips |
N/A | Strict sterile technique for CSF draws or flushing | N/A | Education through face-to-face meetings and web-based multimedia. Summary posters placed in the neurosurgery clinic and neuro-ICU |
| Champey et al37 | Retrospective, multicenter | Journal of Neurosurgery | 462 patients, preprotocol: 241 (Saint-Etienne and Marseille combined); postprotocol: 221 (Grenoble) | Implementation of daily CSF sampling and intraventricular administration of antibiotics before EVD removal at the Grenoble site (postprotocol) | Standardized EVD insertion under sterile conditions | Same as the preprotocol | N/A | N/A | CSF sampling based on clinical suspicion of infection | Daily CSF sampling and intraventricular antibiotics before EVD removal | N/A | N/A |
| Kubilay et al18 | Prospective, observational | Journal of Neurosurgery | 2928 preprotocol: 217 (estimated from the infection rate); postprotocol: 2711; aprevious infections were excluded | Hand hygiene before the procedure + prophylactic antibiotics (oxacillin, cefazolin, or vancomycin) + sterile gloves changed between preparation, draping, and procedure + hair removal by clipping + skin preparation using iodine povacrylex (0.7% available iodine) and isopropyl alcohol (74%) + full body and head drape Full surgical attire for the surgeon and other bedside providers + Antimicrobial-impregnated catheter (Codman Bactiseal antimicrobial catheter system) |
Techniques and procedures were variable and not standardized | Use of antimicrobial-impregnated catheters and adherence to strict sterile techniques | N/A | Use of chlorhexidine-eluting patch over the catheter exit site and medium-sized transparent dressing film secured with sterile adhesive strips | N/A | Strict sterile technique for CSF draws or flushing, including cap and mask, sterile gown, and sterile gloves, and cleaning the access port and surrounding tubing with chlorhexidine | N/A | Use of a procedural checklist to confirm proper insertion and monitor practice Continuous feedback and training of staff |
| Huang et al38 | Retrospective, multicenter | Neurocritical Care | 742 preprotocol: 446 Postprotocol: 296 |
Implementation of chlorhexidine gluconate bathing and needle-free Y-connectors Standard EVD insertion protocols and care bundles |
Complete hair shaving, sterile draping of the head and full-body draping, prophylactic antibiotics, nonimpregnated catheters | Same as the preprotocol | Dressing changes every 3 d, maintaining a closed system | Same as the preprotocol with chlorhexidine gluconate bathing | CSF sampling based on clinical suspicion using the standard aseptic technique | Modified to use needle-free Y-connectors and chlorhexidine gluconate bathing | N/A | N/A |
| Lwin et al25 | Prospective observational | Singapore Medical Journal | 234 preprotocol (Phase I): 82; Postprotocol (Phase II): 79; Postprotocol (Phase III): 73 | Phase II: Meticulous surgical techniques; enforced hand washing for at least 2 min; limiting the number of personnel in the operating theater Minimizing operation duration; training junior doctors on aseptic CSF sampling; SOP for EVD nursing management EVD care workshops for nurses; competency skill checks for neurosurgical ward nurses Phase III: Introduction of silver nanoparticle-impregnated EVD catheters |
Standard surgical techniques without specific emphasis on asepsis or duration of procedures | Phase II: Meticulous surgical techniques, enforced hand washing, limiting personnel, minimizing operation duration Phase III: Introduction of silver nanoparticle-impregnated EVD catheters |
N/A | Phase II: SOP for nursing management, EVD care workshops, competency skill checks Phase III: Same as Phase II |
N/A | Phase II: Training on aseptic CSF sampling, sterile prepacked disposable culture sets, thorough cleaning of sampling ports with antiseptic solution Phase III: same as Phase II |
N/A | N/A |
| Lackner et al39 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Neurocritical Care | 39 preprotocol: 20; Postprotocol: 19; a1 patient was excluded because of EVD replacement | Use of silver nanoparticle-impregnated EVD catheters in the treatment group Standardized catheter insertion and maintenance protocols Regular CSF sampling |
Standard catheter insertion without silver nanoparticles | Use of silver nanoparticle-impregnated EVD catheters | Standard routine dressing changes with topical antiseptics (Octenisept) | Same as the preprotocol | CSF samples drawn 3 times a week | CSF samples drawn daily under sterile conditions | N/A | N/A |
| Flint et al26 | Retrospective | World Neurosurgery | 451 preprotocol (baseline, 2005-2007): 143 Postprotocol (initial 3-y period, 2009-2011): 119 Postprotocol (4-y follow-up, 2012-2015): 189 EVD in 173 patients |
Broad hair clipping Full draping and barrier precautions Strict sterile technique Multiple chlorhexidine skin preps Use of rifampin-minocycline-impregnated EVD catheter EVD tunneling by 3-5 cm Stabilization of the catheter with a question-mark array of surgical staples Use of a chlorhexidine-eluting patch Large transparent dressing Adhesive strips Strict sterile technique for any manipulation (CSF sampling, flushing, transduction line attachment) |
Standard practices without the infection control protocol | Broad hair clipping, full draping, barrier precautions, strict sterile technique, multiple chlorhexidine skin preps, use of rifampin-minocycline-impregnated EVD catheter, tunneling by 3-5 cm, stabilization with surgical staples, use of chlorhexidine-eluting patch | Standard dressing practices without the infection control protocol | Use of a chlorhexidine-eluting patch, large transparent dressing, and adhesive strips | Standard manipulation practices without the infection control protocol | Strict sterile technique for CSF sampling, flushing, and transduction line attachment | N/A | N/A |
| Phan et al40 | Retrospective | Journal of Clinical Neuroscience | 153 preprotocol (2006-2007): 43; postprotocol (2007-2010): 110 | EVD transducer systems set up in the theater Cessation of prophylactic antibiotics after 24 h CSF samples taken second or third daily Discontinuation of elective EVD changes |
Standard practices without specific infection control measures | EVD transducer systems set up in the theater Broad hair clipping, full draping, barrier precautions, strict sterile technique, multiple chlorhexidine skin preps |
Standard practices without specific infection control measures | Use of a chlorhexidine-eluting patch, large transparent dressing, and adhesive strips | Standard practices without specific infection control measures | CSF samples taken on the second or third day using the strict sterile technique | N/A | N/A |
| Hong et al8 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Acta Neurochirurgica | 349 preprotocol: 141 Postprotocol: 208 |
Intravenous analgosedation; Wide shaving of the insertion site; strict hand washing; preprocedural prophylactic intravenous antibiotics; multistage preparation with disinfection; full draping and barrier precautions; surgical disinfection for 5 min; full body drape; meticulous sterile technique with nursing assistance EVD tunneling; EVD attached directly to a closed drainage system; Occlusive wound dressing; Indications and Timing of EVD Change: Routine EVD replacements are not performed; EVD is replaced only because of dislocation or occlusion; Other Relevant Innovations or Study Points: Continued antibiotic prophylaxis over the entire EVD period; CSF samples obtained from a 3-way tap rather than directly from the EVD |
No specific protocol | Wide shaving, strict hand washing, preprocedural prophylactic antibiotics, multistage disinfection, full draping and barrier precautions, surgical disinfection, full body drape, sterile technique, EVD tunneling, EVD attached to a closed drainage system | N/A | Occlusive wound dressing | N/A | Hand disinfection and sterile gloving before any EVD manipulation, daily drainage changes by a physician, scalp disinfection at dressing changes, no flushing of occluded EVDs, CSF samples obtained from a 3-way tap | N/A | N/A |
| Rojas-Lora et al41 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Annals of Clinical Microbiology and Antimicrobials | 198 preprotocol: 87 Postprotocol: 111 |
Implementation of a checklist for protocol systematization; staff training and re-education; reduction in unnecessary catheter sampling; stricter antiseptic measures during catheter insertion and maintenance; use of chlorhexidine dressing; updated protocol for sampling CSF | N/A | Checklist completed by a nursing assistant; informed consent signature; date of insertion; location of insertion; operator's record; number of people in the room; surgeon's hand washing and use of cap, mask, gloves, and sterile gown; patient preparation (wide shaving; washing with soap and water; betadine painting; sterile field; and antibiotic prophylaxis administration); drainage tunneling 3-5 cm from the insertion point; fixation of the catheter; chlorhexidine spray; connection to the collector system; sterile protection of the first key of the collecting system | N/A | Head washing every 4 d with chlorhexidine soap, insertion point care every 4 d or when dirty, wet, or unhooked (including patient mask placement, sterile drape and gauze, physiological saline, chlorhexidine antiseptic solution, and chlorhexidine dressing) | N/A | Hand disinfection and sterile gloving before any EVD manipulation, daily drainage changes by a physician, scalp disinfection at dressing changes, no flushing of occluded EVDs, CSF samples obtained from a 3-way tap, replacement of drainage bag when 3/4 full, sampling from the 7th day and every 4 d thereafter | N/A | N/A |
| Harrop et al27 | Prospective observational | Neurosurgery | 1961 Preprotocol (Period 1): 327 Postprotocol (Period 2): 281 Postprotocol catheter A (Period 3): 195 Postprotocol without antibiotic catheters (Period 4): 157 Postprotocol with catheter B (Period 5): 1001 |
Implementation of an evidence-based standardized catheter insertion protocol Use of antibiotic-impregnated catheters |
N/A | Standardized protocol including elimination of room traffic, use of electric clippers, surgical scrub, full barrier precautions, disinfection with povidone–iodine, use of monofilament suture | N/A | Use of bio-occlusive dressing | N/A | Strict sterile technique for CSF sampling, flushing, and transduction line attachment | N/A | N/A |
| Dasic et al42 | Quasi-experimental; prospective cohort for bundle and retrospective for control | British Journal of Neurosurgery | 113 preprotocol: 41 patients, 51 EVD; postprotocol: 54 patients, 62 EVD | Routine EVD replacements are not performed; EVD is replaced only because of dislocation or occlusion | N/A | Theater doors remain closed during the procedure; all staff wear masks; minimal staff in the theater; EVDs inserted in the operating theater unless in an emergency Administration of cefuroxime and gentamicin STAT on induction; shaving and prepping with alcoholic betadine and chlorhexidine; double gloving by the surgeon Use of drapes and Opsite; burr hole or twist drill formation; gloves changed/outer gloves removed before handling EVD; EVD insertion and CSF sample collection Administration of intrathecal vancomycin if possible; EVD tunneling for a minimum of 10 cm; EVD secured and attached to a closed drainage system |
N/A | Use of bio-occlusive dressing, sterile techniques maintained | N/A | Strict sterile technique for CSF sampling, flushing, and transduction line attachment | N/A | N/A |
| Lozano et al43 | Retrospective, cohort | Cureus | 58 preprotocol: 33 Postprotocol: 25 |
Introduction of an EVD care bundle including standardized procedures, associated checklists, monitoring requirements | N/A | The procedure is performed under full sterile conditions All team members involved in the procedure wear sterile gowns, gloves, masks, and caps EVD catheter is tunneled subcutaneously for a distance of >5 cm from the entry point |
N/A | The insertion site is covered with a transparent, semipermeable dressing to allow for inspection without removing the dressing | N/A | The catheter is connected to a closed drainage system CSF samples are taken using the sterile technique as required |
N/A | N/A |
| Korinek et al44 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Acta Neurochirurgica | 306 preprotocol: 131 patients (161 EVDs) Postprotocol: 175 patients (216 EVDs) |
Complete hair clipping Betadine scrub shampoo Scalp disinfection with iodine alcohol twice Lateral tunneling of the catheter Prophylactic antibiotics only for emergency EVDs |
N/A | Hair entirely clipped and shampooed with betadine scrub Scalp disinfected with iodine alcohol twice Catheter inserted into the lateral ventricle through a 4-mm burr hole EVD catheter tunneled laterally under the scalp Connected to a drainage system with sterile protections |
Head dressing changed every 2 d Hair not shampooed during EVD duration Daily emptying of the drainage bag System unblocked by syringe aspiration or saline injection if needed |
Head dressing changed every 3 d Hair shampooed every 6 d Closed drainage system strictly respected Clamping of drainage tubing for switching between pressure monitoring and CSF drainage Avoidance of system manipulations |
N/A | N/A | N/A | N/A |
| Rahman et al45 | Retrospective, cohort | Joint Commission Journal on Quality and Patient Safety | Preprotocol: 217; postprotocol: 2911 | Turn the patient bed 45° to improve access and remove patient monitor cables and ventilator tubing from the sterile field Use a large procedure table for establishing a sterile field Ensure that mobile procedure lights are available Limit EVD placement to the operating room, certain ICUs, and the emergency department Equip these units with stocked “ventriculostomy carts” Use antibiotic-coated ventricular catheters (0.15% clindamycin and 0.054% rifampicin) Administer a dose of cefazolin before the procedure (clindamycin for patients with severe penicillin or cephalosporin allergies) No routine postprocedure antibiotics Full cranial drape used to create a large sterile field Junior residents allowed to insert an EVD independently only after 5 supervised insertions |
N/A | Turn bed 45° Confirm procedure light in place Wide hair clipping with electric clippers Alcohol skin prep followed by an antiseptic solution Full cranial drape to create a sterile field Administer prophylactic antibiotics Insert an antibiotic-coated catheter Use of a checklist to ensure adherence to the protocol |
N/A | EVD dressings maintained by the neurosurgical team and changed every 72 h | N/A | Sterile gloves and antiseptic solution used whenever accessing the EVD CSF sampled only if patients developed fever or other signs/symptoms of infection |
N/A | N/A |
| Zakaria et al21 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Journal of Hospital Infection | 239 patients Preprotocol: 159 patients, 234 EVD Postprotocol: 107 patients, 132 EVD |
Insertion of drains under strict aseptic conditions in the operating room Use of prophylactic antibiotics at induction; use of silver ion-coated catheters as standard Limiting sampling to when clinically indicated; A weekly microbiology round to review all patients with drains |
Drains inserted under standard aseptic conditions with aqueous betadine skin preparation Prophylactic antibiotics given at induction |
Insertion of drains under strict aseptic conditions in the operating room Prophylactic antibiotics given at induction (cefuroxime, teicoplanin, or vancomycin) Drains inserted through a burr hole at Kocher's point, tunneled under the skin, and connected to an external collecting system Use of silver ion-coated catheters as standard |
N/A | Dressings changed every 72 h by the neurosurgical team | N/A | Sterile gloves and antiseptic solution used whenever accessing the EVD CSF sampled only if patients developed fever or other signs/symptoms of infection |
N/A | N/A |
CHG, chlorhexidine gluconate; CLD, catheter-locking device; CSF, cerebrospinal fluid; CT, computed tomography; EVD, external ventricular drainage; ICU, intensive care unit; N/A, not available; SOP, standard operating procedures.
TABLE 3.
EVD-Related Infections Before and After the Implementation of Bundles of Care
| Authors | Study design | Journal | No. of patients | Infection definition | Infection time of onset days | Infection number before bundles | Infection rate before bundles (%) | Infection number after bundles | Infection rate after bundles |
|---|---|---|---|---|---|---|---|---|---|
| Hoefnagel et al28 | Quasi-experimental; prospective cohort for bundle and retrospective for control | BioMed Central Neurology | 409 patients, preprotocol: 228 Postprotocol: 181 |
Positive CSF culture and at least 2 symptoms of meningitis, based on CDC criteria | Preprotocol: 8.0 ± 9.0; postprotocol: 9.5 ± 10.0 | 53 | 23 | 16 | 9% |
| Katzir et al29 | Retrospective, cohort | World Neurosurgery | 142 patients, preprotocol group (Group A: Elective EVD replacement): 43 patients Postprotocol group (Group B: Clinically indicated EVD replacement): 99 patients |
Positive CSF culture and at least one of the following: fever >38°C with no other recognized cause, elevated CSF protein level (>40 g/dL), hypoglycorrhachia (<45 mg/dL), or organisms found on CSF gram staining | N/A | 14 | 32 | 8 | 8% |
| Piccirilli et al30 | Retrospective, cohort | World Neurosurgery | 196 EVDs placed in 161 patients. Preprotocol group A: 86; postprotocol group B: 75 | A catheter-related CSF infection was defined by the presence of any of the following criteria: Clinical symptoms: Fever (>38°C) with no other recognized cause Meningismus (symptoms of meningitis such as headache, neck stiffness, photophobia) CSF analysis: Positive CSF culture (growth of microorganisms from the CSF sample) Elevated CSF white blood cell count (>5 cells/µL) Elevated CSF protein level (>50 mg/dL) Low CSF glucose concentration (<2/3 of blood glucose level) Other laboratory and imaging findings: CSF Gram stain showing microorganisms Evidence of infection around the catheter insertion site or along the catheter tract, possibly confirmed by imaging studies (eg, MRI, CT scan) |
N/A | 7 | 8.1 | 0 | 0% |
| Chatzi et al19 | Prospective observational | Critical Care Medicine | 139 patients, preprotocol group: 82 patients Postprotocol group: 57 patients |
Detection of a pathogen in the CSF or the combination of at least one typical clinical sign indicating ventriculitis plus pathological CSF findings (elevated cell count and/or decreased glucose level) | 12.8 (9.9-15.8) | 23 | 28 | 6 | 10.5% |
| Omar et al7 | Prospective observational | Malaysian Journal of Medical Sciences | 87 patients, Group A: Tunneling <5 cm: 35; Group B: Tunneling >5 cm: 52 | Positive CSF culture and Gram stain. Presence of supportive CSF laboratory findings: Pleocytosis with WBC count >11/mm3. Decreased CSF glucose level (normal level = 2.5 mmol/L). Increased CSF protein level (normal level = 0.4 g/L) | Group A: drainage >10 d: 19 of 36 (52.8%) developed EVD-related infection. Group B: drainage <10 d 9/51 (17.6%) infections | 22 | 62.9 | 6 | 11.5% |
| Williams et al31 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Journal of Neurosurgery | 382 patients, preprotocol group: 206; postprotocol group: 176 | Positive EVD CSF culture showing growth of microorganisms. Laboratory findings support the clinical diagnosis of ventriculitis. Infection considered only as “proven ventriculitis” with confirmatory laboratory findings | Median 6 (5-9) | 20 | 9.7 | 6 | 3.3% |
| Choo et al32 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Journal of Korean Medical Science | 183 patients, (preprotocol group: 84; postprotocol group: 99) | Positive CSF culture or catheter tip culture, or at least one of the following CSF abnormalities: Low glucose level <40 mg/dL or <50% serum glucose High protein level >50 mg/dL High cell count (white blood cell) >1000/mm3 with 50% or more polymorphonuclear leukocytes Two of the following symptoms without another clear source of infection: fever >38°C, 1 meningeal sign (eg, headache, neck stiffness, new onset seizure), new or worsening deterioration of consciousness |
Median preprotocol group: 11.61, postprotocol group: 12.59 d | 14 | 16.7 | 4 | 4% |
| Talibi et al33 | Quasi-experimental; prospective cohort for bundle and retrospective for control | British Journal of Neurosurgery | 275 patients, preprotocol group: 120 Postprotocol group: 155 |
A single positive CSF culture obtained from a ventricular catheter or CSF from lumbar puncture | N/A | 32 | 27 | 16 | 10% |
| Thien et al23 | Quasi-experimental; prospective cohort for bundle and retrospective for control | World Neurosurgery | 551 patients, preprotocol group (2014): 165 patients Postprotocol group (Groups of 2015 and 2016): 386 patients |
CSF infection defined as organisms seen on Gram stain or culture | N/A | 8 | 4.8 | 11 (groups of 2015 and 2016) | 2.8% (groups of 2015 and 2016) |
| Levy et al34 | Quasi-experimental; prospective cohort for bundle and retrospective for control | American Journal of Infection Control | 310 patients, preprotocol group: 129; postprotocol group: 181 | Ventriculitis defined as clinical suspicion with positive CSF cultures, excluding probable contaminants | 10 (4-17) | 9 | 7 | 8 | 4.4% |
| Dasgupta et al24 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Journal of Neurosurgery | 291 patients, preprotocol group: 200; postprotocol group 2016 and 2017: 91 (126 EVDs) | Presence of positive CSF cultures and clinical signs of infection | N/A | 17 | 8.5 | Postprotocol (groups 2016 and 2017): 6 | Postprotocol (groups 2016 and 2017): 6.5% |
| Whyte et al35 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Clinical Neurology and Neurosurgery | 145 patients, Preprotocol: 91; postprotocol: 54 | Positive CSF culture or escalation of antibiotics for suspected ventriculitis for >48 h. The presence of subgaleal/subdural drains was supported by signs of cellulitis or purulent discharge and escalation of antibiotics for greater than 48 h or evidence of possible infection on postoperative head CT or MRI | N/A | 5 | 6 | 3 | 5.50% |
| Flint et al36 | Retrospective | Neurosurgery | 262 patients, preprotocol: 143; postprotocol: 119 | Growth of 1 or more organisms from a CSF sample obtained from a ventriculostomy catheter placed in the ICU Ventriculitis: Clinical documentation of suspected ventriculitis and/or treatment with antibiotics for this purpose and the presence of a positive CSF culture or a CSF profile consistent with infection |
Preprotocol: Median 8 (4-12 d) Postprotocol: 11 |
14 | 9.8 | 1 | 0.8% |
| Champey et al37 | Retrospective, multicenter | Journal of Neurosurgery | 462 patients, Preprotocol: 241 (Saint-Etienne and Marseille combined); postprotocol: 221 (Grenoble) | One or more positive CSF cultures or Gram stain with CSF pleocytosis and biochemical abnormalities Ventriculitis: CSF pleocytosis and biochemical abnormalities with degradation of neurological status and fever |
N/A | 20 | 8.3 | 3 | 1.4% |
| Kubilay et al18 | Prospective, observational | Journal of Neurosurgery | 2928 patients, preprotocol: 217 (estimated from the infection rate); postprotocol: 2711; aprevious infections were excluded | Presence of positive CSF culture with an increase in CSF WBC or an increase in CSF WBC with clinical signs of infection in the setting of negative CSF culture | 8-14 d | 20 | 9.2 | 43 | 1.5% |
| Huang et al38 | Retrospective, multicenter | Neurocritical care | 742 patients, preprotocol: 446 postprotocol: 296 |
Fever (>37.5°C) after EVD placement, CSF pleocytosis (CSF WBC count >100/mm3), ratio of CSF to serum glucose level <0.5, and positive CSF Gram stain and culture results | 24.1 ± 7.2 | 16 | 3.6 | 3 | 1% |
| Lwin et al25 | Prospective observational | Singapore Medical Journal | 234 patients, preprotocol (Phase I): 82; postprotocol (Phase II and III): 152 | Catheter-associated meningoventriculitis confirmed by positive CSF culture, CSF pleocytosis, high protein level, or low glucose level in addition to positive culture | N/A | 5 | 6.1 | Phase II/III: 3 | Phase II/III: 1.9% |
| Lackner et al39 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Neurocritical Care | 39 patients, preprotocol: 20; postprotocol: 19 a 1 patient was excluded because of EVD replacement | Proven by positive CSF culture | >10 | 5 | 25 | 0 | 0% |
| Flint et al26 | Retrospective | World Neurosurgery | 435 patients, preprotocol (baseline, 2005-2007): 143 Postprotocol (initial 3-y period, 2009-2011, and 4-y follow-up, 2012-2015): 292 patients |
Growth of 1 or more organisms from a CSF sample obtained from an EVD catheter placed in the ICU, or a positive Gram stain from such a sample Ventriculitis: Clinical documentation of suspected ventriculitis and/or treatment with antibiotics for this purpose, along with either a positive CSF culture or a CSF profile consistent with infection |
N/A | 14 | 9.8 | Postprotocol: 1 | Postprotocol: 0.3% |
| Phan et al40 | Retrospective | Journal of Clinical Neuroscience | 153 preprotocol (2006-2007): 43; postprotocol (2007-2010): 110 | Positive microbial growth from a CSF sample | N/A | 9 | 20.9 | 13 | 11.5% |
| Hong et al8 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Acta Neurochirurgica | 349 patients, preprotocol: 141 Postprotocol: 208 |
Positive if at least one of the criteria from (a) to (d) and at least one criteria from (e) to (g): (a) Decreased CSF glucose level; (b) increased CSF protein level; (c) increased CSF cell count; (d) increased inflammatory factors in blood count; (e) clinical presentation of meningitis; (f) positive bacterial culture or Gram staining in CSF sampling; (g) radiological signs of ventriculitis | N/A | 41 | 29.1 | 10 | 4.8% |
| Rojas-Lora et al41 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Annals of Clinical Microbiology and Antimicrobials | 198 patients, preprotocol: 87 Postprotocol: 111 |
Positive CSF cultures, microorganisms on Gram stain, and suggestive symptoms, with cultures considered for EVDRI if positivity detected from 24 h after implantation to 5 d after removal | N/A | 18 | 21 | 5 | 4% |
| Harrop et al27 | Prospective observational | Neurosurgery | 1961 patients, preprotocol: 327 Postprotocol: 1634 |
Two positive CSF cultures from ventriculostomy catheters with a concurrent increase in CSF white blood cell count | N/A | 22 | 6.7 | Postprotocol: 46 | Postprotocol: 2.8% |
| Dasic et al42 | Quasi-experimental; bundle prospective cohort for and retrospective for control | British Journal of Neurosurgery | 113 patients, preprotocol: 41 patients, 51 EVD; postprotocol: 54 patients, 62 EVD | Positive microbial growth from a CSF sample | N/A | 14 | 27 | 7 | 12% |
| Lozano et al43 | Retrospective, cohort | Cureus | 58 patients, preprotocol: 33 Postprotocol: 25 |
Positive CSF cultures requiring treatment | N/A | 7 | 21.21 | 0 | 0 |
| Korinek et al44 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Acta Neurochirurgica | 306 patients, preprotocol: 131 patients (161 EVDs) Postprotocol: 175 patients (216 EVDs) |
Defined as EVD-related ventriculitis if a febrile patient (temperature >38.5°C) had a positive CSF culture, pleocytosis (>15 cells/mm3), and a CSF/plasma glucose ratio of <0.5, with no other detectable source of CNS infection | Control period: 8.7 (2-24) Study period: Mean 17.7 (5-31) |
16 | 12.2 | 10 | 5.7% |
| Rahman et al45 | Retrospective, cohort | Joint Commission Journal on Quality and Patient Safety | 3128 patients, preprotocol: 217; postprotocol: 2911 | Defined as either: a positive CSF culture with an increase in CSF WBC Significantly increased CSF WBC with clinical signs of infection in the setting of a negative CSF culture |
N/A | 20 | 9.2 | 35 | 1.2% |
| Zakaria et al21 | Quasi-experimental; prospective cohort for bundle and retrospective for control | Journal of Hospital Infection | 239 patients Preprotocol: 159 patients, 234 EVD Postprotocol: 107 patients, 132 EVD |
Defined as per CDC criteria: organisms cultured from CSF or clinical signs/symptoms with elevated CSF WBC, protein, and/or decreased glucose | N/A | 54 | 33.96 | 18 | 16.82% |
CDC, Centers for Disease Control and Prevention; CNS, central nervous system; CSF, cerebrospinal fluid; CT, computed tomography; EVD, external ventricular drainage; EVDRI, external ventricular drainage-related infection; ICU, intensive care unit; N/A, not available; WBC, white blood cells.
Basic Study Features
The review comprises data from 28 studies published between 2005 and 2024, examining the effectiveness of infection control protocols for EVDs in neurosurgical patients. The studies were primarily based on quasi-experimental designs, comparing pre- and postprotocol infection rates. Sample sizes in the different analyzed studies varied widely, ranging from 39 to 2928 patients. They were conducted in various countries and published in neurosurgical and critical care journals. Patient demographics across the publications showed mean ages ranging from 31 to 67 years, with male patients comprising 40% to 65% of study populations. The duration of EVD placement varied among the selected studies, with the median duration typically ranging from 5 to 15 days, although some studies reported longer durations up to 24 days. The pathologies treated with EVDs were diverse, with many studies reporting a mix of conditions. Subarachnoid hemorrhage was often the most common indication, followed by intracerebral hematoma, traumatic brain injury, and tumors. Some studies also included patients with intraventricular hemorrhage, stroke, and other less common conditions requiring CSF diversion (Table 1). The data from included studies are heterogeneous.
Care Bundle Approaches for ERIs
Innovations and New Control Protocols
Key components commonly reported are related to standardized insertion techniques and perioperative antibiotic prophylaxis (78.6%). Chlorhexidine skin preparation (50.0%) and catheter tunneling longer than 5 cm (35.7%) were moderately common, among studies reporting tunneling as protocol innovation. Antibiotic-impregnated or silver-coated catheters were reported in 25.0% of studies. Enhanced simulation-based education was less common (17.9%). Some studies included catheter-locking devices and electronic documentation systems. Strategies like risk stratification pathways, multidisciplinary collaboration, and documentation of all CSF sampling events were variably implemented. Checklists, audits, and feedback mechanisms were common to enhance protocol adherence. An overview of major care bundle components is displayed in Figure 2.34,40
FIGURE 2.
Comprehensive overview of evidence-based components included in effective EVD care bundles, categorized by insertion techniques, system manipulation, dressing protocols, and organizational strategies. Created using BioRender. Vasella, F. (2025) https://BioRender.com/e8ocbr7. CSF, cerebrospinal fluid; EVD, external ventricular drain.
EVD Insertion Protocols
Insertion protocols in the included studies consistently emphasized strict sterile technique and barrier precautions. Catheter tunneling of 5 to 10 cm was reported in 46.4% of studies, whereas perioperative antibiotic prophylaxis was used in 64.3%. Specific catheter types were reported in 67.9%, and detailed sterile insertion procedures were described in 53.6%. Chlorhexidine skin preparation and hair clipping/shaving were each reported in 32.1% of studies. Less frequent measures included full-body draping or closed drainage at insertion (10.7% each). A few studies described innovative approaches, such as biocompatible locking devices or curvilinear incisions, in narrative form.
EVD Dressing Protocols
Dressing protocols commonly incorporated chlorhexidine-eluting patches at the catheter exit site and transparent semipermeable dressings to facilitate inspection. Regular dressing changes every 3 to 7 days, or sooner if soiled, were widely practiced. Some bundles specified meticulous scalp disinfection, use of sterile barriers (eg, mask, gloves), and application of antiseptic solutions for defined durations. Additional innovations, including spray-on dressings or benzoin tincture to enhance adherence, were less frequently reported.
EVD Manipulation Protocols
The strict aseptic technique during EVD handling was emphasized universally across studies, including hand hygiene, sterile gloves, and antiseptic cleaning of access ports before use. Many protocols reduced CSF sampling frequency, often shifting from daily sampling to scheduled intervals such as every 3 days or only when clinically indicated. Closed drainage systems were commonly used (>75%) to minimize contamination risk. Less frequent measures included the use of needle-free connectors and standardized CSF collection and handling protocols (each <25%). Several studies also stressed minimizing overall manipulation of the EVD system and maintaining a closed circuit for all procedures, including CSF sampling, flushing, and transduction line attachment.
Organizational Strategies and Other Care Bundle Components
Organizational measures frequently addressed education, training, and adherence monitoring. Multidisciplinary staff training or re-education and the use of checklists for protocol adherence were each reported in 21.4% of studies. Continuous monitoring, audits, or surveillance were implemented in 17.9%. Standardized EVD kits, infection control committee involvement, written infection management protocols, and simulation-based education were each described in 10.7%. Electronic documentation systems to track EVD care and CSF sampling were used in 7.1% of studies.
Several studies additionally described simulation workshops for ongoing training and procedural checklists to ensure correct insertion practices and consistent monitoring. Practical strategies reported included patient positioning during insertion, use of mobile procedural lighting, and equipping clinical units with dedicated ventriculostomy carts. Comprehensive education programs were also detailed, combining face-to-face sessions, web-based multimedia resources, and strategically placed summary posters in clinical areas.
Reduction of ERIs After Introducing Effective Care Bundles
Implementation of EVD care bundles consistently resulted in reduced ERI rates across the included studies, as summarized in Table 3. Before bundle implementation, ERI rates varied substantially, ranging from 3.6% to 62.9%, with most studies reporting rates between 8% and 30%. After the introduction of care bundles, infection rates notably decreased, typically falling between 0% and 10%. Definitions of infection differed slightly among studies but generally included positive CSF cultures combined with clinical symptoms or additional laboratory findings. The reported onset of infection ranged from 8 to 17 days after EVD insertion.
Several studies demonstrated marked reductions in ERI rates. For example, Piccirilli et al30 reported a decrease from 8.1% to 0%, whereas Flint et al36 described a reduction from 9.8% to 0.8%. By contrast, some studies observed minimal changes, such as the study by Whyte et al,35 who reported a reduction from 5.6% to 5.5%. Long-term follow-up data provided further evidence of sustained improvements: Flint et al26 observed a progressive reduction from an initial rate of 9.8% down to 0% over a four-year period, and Thien et al23 reported improvements from 4.8% down to 2.0% over 2 years. Larger-scale studies conducted by Kubilay et al18 and Rahman et al45 similarly demonstrated significant reductions in ERI rates, dropping from 9.2% to 1.5% and 9.2% to 1.2%, respectively.
Overall, these findings support the effectiveness of care bundles across varied healthcare environments. The meta-analysis of all 28 included studies indicated statistically significant reductions in ERI rates after bundle implementation, with an overall effect P-value of <.0001. However, heterogeneity among studies was moderate to high, as indicated by I2 (Figure 3). Publication bias was assessed by visual inspection of a funnel plot. There was no apparent skew or asymmetry suggestive of publication bias (Figure 4).
FIGURE 3.
Forest plots detailing the meta-analysis of external ventricular drainage–related infections in the study cohorts before the introduction of care bundles and after their implementation.
FIGURE 4.
Funnel plot of all studies included in the meta-analysis showing no publication bias.
DISCUSSION
ERIs remain a significant concern in neurosurgical practice, with infection rates varying widely across different clinical settings. This systematic review and meta-analysis aimed to demonstrate the effectiveness of care bundle approaches in reducing ERI rates and to identify common, effective components that could inform standardized practices. Our analysis of 28 studies provides robust evidence supporting the implementation of care bundles as a successful strategy to reduce ERIs, across diverse healthcare environments.
These findings align with the previous literature, including a meta-analysis by Sieg et al,46 which reported significant reductions in infection rates after standardized EVD placement and management protocols. Notably, Hoefnagel et al28 reported the complete elimination of infections after protocol implementation for patients with secondary meningitis post-EVD placement. Multicenter studies conducted by Champey et al37 and Huang et al38 also demonstrated significant reductions in infection rates, reinforcing the effectiveness and generalizability of standardized protocols.
Key Components of Effective Care Bundles
Insertion Protocols
Our analysis indicates that standardized insertion techniques are essential for minimizing ERIs. Critical components consistently identified included a full sterile technique with barrier precautions, chlorhexidine or iodine-based skin preparation, catheter tunneling, and the use of antibiotic- or silver-coated catheters. Korinek et al44 demonstrated that adherence to a detailed insertion protocol significantly predicts lower infection rates. Further supporting these findings, Omar et al7 emphasized the benefits of catheter tunneling, whereas Katzir et al29 recommended clinically indicated rather than routine EVD replacements. Randomized trials and subsequent meta-analyses have validated the effectiveness of coated catheters, particularly silver-coated variants, as reported by Lackner et al39 and confirmed by a recent meta-analysis.47
Manipulation Protocols
Strict aseptic technique during EVD manipulation is universally emphasized. Several studies highlighted the strict aseptic technique, reduction in the frequency of CSF sampling, and the use of closed drainage systems as critical manipulation components. Thien et al23 and Lozier et al6 emphasized the role of reduced CSF sampling frequency in significantly lowering ERI rates. Maintenance of closed drainage systems was further validated by Lackner et al,39 reinforcing these practices as essential components of effective EVD care bundles.
Dressing Protocols
Effective dressing protocols were characterized by the use of chlorhexidine-eluting patches and transparent semipermeable dressings with scheduled dressing changes. Flint et al26 and Hong et al8 demonstrated significant infection reductions associated with meticulous scalp disinfection and the use of antiseptic dressings such as chlorhexidine-eluting patches. These findings are supported by Rojas-Lora et al,41 who also emphasized regular dressing changes combined with reduced catheter sampling frequency and stringent antiseptic measures.
Organizational Strategies
Multidisciplinary training, staff education, and adherence strategies emerged as critical organizational components. Studies consistently reported sustained ERI rate reductions associated with educational interventions and structured feedback mechanisms.48 Electronic documentation systems were also highlighted as effective tools for protocol adherence and error reduction.23 Despite these clear benefits, challenges remain, particularly in resource-limited settings, underscoring the importance of accessible and adaptable educational approaches. Choo et al32 and Talibi et al33 demonstrated the effectiveness of comprehensive staff training combined with systematic documentation in achieving sustainable reductions in ERIs. Others described individual training through web seminars and in-person meetings and through continuous feedback from the staff in charge of implementing the protocols.18,36
Implications for Clinical Practice and Future Directions
The consistent reduction in ERI rates observed across diverse healthcare settings highlights the universal applicability of care bundles in EVD management. These protocols not only standardize practices but also foster a culture of accountability among healthcare providers, ensuring adherence to evidence-based guidelines. However, variability in bundle components across institutions underscores the need for further research to identify the most critical elements for infection prevention. The Neurocritical Care Society has established guidelines for EVD insertion which explicitly recommend bundle approaches.2 In a hospital setting, bidirectional communication between neurosurgeons, infectious disease specialists, and intensive care staff is paramount. EVD insertion is often performed at the patient's bedside in emergency settings, where strict adherence to care bundle protocols is essential to prevent ERIs.31 Even minor protocol changes can significantly reduce common complications associated with EVD placement, as demonstrated in retrospective analyses.42,43
Recommendations for a Standardized Care Bundle
Based on the frequency, effectiveness, and ease of implementation of components identified in this review, we recommend that an EVD care bundle ideally includes the following: (1) standardized sterile insertion techniques with barrier precautions, (2) perioperative antibiotic prophylaxis, (3) catheter tunneling >5 cm, (4) antiseptic catheter coatings such as antibiotic or silver-based coatings, (5) reduced frequency of CSF sampling, (6) use of closed drainage systems, (7) regular dressing changes with antiseptic solutions or chlorhexidine-eluting patches, (8) structured multidisciplinary training and education programs, and (9) implementation of structured protocol adherence tools.
Limitations
Limitations of our study include significant heterogeneity in study designs, definitions of ERI, and variations in bundle components, complicating direct comparisons. While RCTs remain the gold standard, their practical limitations in evaluating complex multicomponent interventions suggest that large-scale observational studies and quality improvement initiatives may be more feasible and informative. Future research should aim to refine bundle components further through multicenter collaborations and cost-effectiveness analyses, as well as the integration of new technologies such as advanced antimicrobial-coated catheters or innovative dressing materials.
CONCLUSION
ERIs appear to be significantly reduced by the adequate implementation of care bundle approaches for EVD insertion. By integrating standardized insertion techniques, strict aseptic manipulation protocols, optimized dressing practices, and comprehensive staff education programs, these bundles address multiple risk factors for infection simultaneously. As supported by our meta-analysis and current literature, widespread adoption of care bundles has the potential to significantly improve patient outcomes while reducing healthcare costs associated with ERIs. In agreement with other publications, the implementation of care bundle approaches and the strict knowledge, training, and adherence to these protocols permit a standardization of the procedure and consequently a better postoperative management diminishing complication rates.
Acknowledgments
Author contributions: Conception and design: MTFN, VES, FV; literature search: MTFN, VP; statistical analysis: MB; manuscript drafting: MTFN, VES, FV; critical review of the manuscript: all authors; reviewed submitted version of manuscript: all authors; study supervision: VES, FV.
Contributor Information
Mateo Tomas Fariña Nuñez, Email: mateo-91@hotmail.com.ar.
Veronica Percuoco, Email: percuocoveronica@gmail.com.
Richard Parvin, Email: richard.parvin@gmx.de.
Massimo Barbagallo, Email: Massimo.Barbagallo@usz.ch.
Stefanos Voglis, Email: stefanos.voglis@usz.ch.
Victor Gabriel El-Hajj, Email: victor.gabriel.elhajj@stud.ki.se.
Adrian Elmi-Terander, Email: adrian.elmi.terander@ki.se.
Erik Edström, Email: erik.edstrom.1@ki.se.
Lewis Thorne, Email: lewisthorne@nhs.net.
Carlo Serra, Email: Carlo.Serra@usz.ch.
Luca Regli, Email: luca.regli@usz.ch.
Victor E. Staartjes, Email: VictorEgon.Staartjes@usz.ch.
Funding
This study did not receive any funding or financial support.
Disclosures
Victor E. Staartjes is supported by the Prof. Dr Max Cloetta Foundation. The other authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.
COMMENTS
The authors conducted a systematic review and meta-analysis of studies looking at the effects of care bundle implementation to reduce infections related to external ventricular drain (EVD) placement, which was as high as 32% prior to care bundle implementation in some studies. Overall, the authors found significantly reduced infection rates across 28 studies incorporating data on over 14 000 patients without evidence of publication bias. Care bundles are structured sets of 3 to 5 evidence-based practices applied collectively. The authors rightly identify that randomized controlled trials are practically limited and that cohort and pre-post comparison studies are likely more suited for evaluating complex, multi-component interventions such as care bundle implementation.
The authors identified perioperative antibiotic prophylaxis, standard sterile technique (ie, prepping, barrier protection), using pathogen-resistant catheters, tunneling greater than 5 cm, antiseptic dressing changes, taking steps to minimize cerebrospinal fluid leak (ie, with closure, drain stitch etc.), minimizing manipulation/accessing (sampling) the closed drainage system, taking steps to reduce duration of CSF drainage (weaning), (provider/nurse) education, and structured protocol adherence tools as essential components of care bundles to reduce ERIs.
EVD placement is special among neurosurgical procedures in that it is one of if not the most commonly performed; it is often emergent and life-saving, and it is usually the first procedure specific to the practice of neurosurgery in which a neurosurgical trainee becomes proficient. For these reasons especially, good technique can make a big difference at reducing potentially devastating complications such as ventriculitis. At the same time, the components of safe and effective EVD placement are directly applicable to the safe and effective practice of neurosurgery in general, including nuances of positioning, fundamentally good surgical hygiene practices, and use of standardized instrumentation (ie, standardized EVD equipment kits).
The authors concluded that the consistent reduction in ERIs reported across diverse healthcare settings supports universal applicability of a care bundle approach. It is possible that opportunities for standardized education of neurosurgery trainees (ie, resident boot camps, the ABNS board certification process) can be an opportunity to actively implement and reinforce best practice care bundles such as those the authors identify.
Christopher Carr, William Ledford, and Fernando L. Vale
Augusta, Georgia, USA
REFERENCES
- 1.Walek KW, Leary OP, Sastry R, Asaad WF, Walsh JM, Mermel L. Decreasing external ventricular drain infection rates in the neurocritical care unit: 12-year longitudinal experience at a single institution. World Neurosurg. 2021;150:e89-e101. [DOI] [PubMed] [Google Scholar]
- 2.Fried HI, Nathan BR, Rowe AS, et al. The insertion and management of external ventricular drains: an evidence-based consensus statement: a statement for healthcare professionals from the Neurocritical Care Society. Neurocrit Care. 2016;24(1):61-81. [DOI] [PubMed] [Google Scholar]
- 3.Hughes JD, Puffer R, Rabinstein AA. Risk factors for hydrocephalus requiring external ventricular drainage in patients with intraventricular hemorrhage. J Neurosurg. 2015;123(6):1439-1446. [DOI] [PubMed] [Google Scholar]
- 4.Dey M, Jaffe J, Stadnik A, Awad IA. External ventricular drainage for intraventricular hemorrhage. Curr Neurol Neurosci Rep. 2012;12(1):24-33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.El-Hajj VG, Pettersson I, Gharios M, et al. Detection and management of elevated intracranial pressure in the treatment of acute community-acquired bacterial meningitis: a systematic review. Neurocrit Care. 2024;41(1):228-243. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lozier AP, Sciacca RR, Romagnoli MF, Connolly ES. Ventriculostomy-related infections: a critical review of the literature. Neurosurgery. 2002;51(1):170-182. [DOI] [PubMed] [Google Scholar]
- 7.Omar MA, Mohd Haspani MS. The risk factors of external ventricular drainage-related infection at hospital Kuala Lumpur: an observational study. Malays J Med Sci. 2010;17(3):48-54. [PMC free article] [PubMed] [Google Scholar]
- 8.Hong B, Apedjinou A, Heissler HE, et al. Effect of a bundle approach on external ventricular drain-related infection. Acta Neurochir (Wien). 2021;163(4):1135-1142. [DOI] [PubMed] [Google Scholar]
- 9.Hersh EH, Yaeger KA, Neifert SN, Kim J, Dangayach NS, Weiss N. Patterns of health care costs due to external ventricular drain infections. World Neurosurg. 2019;128:e31-e37. [DOI] [PubMed] [Google Scholar]
- 10.Hoefnagel D, Dammers R, Ter Laak-Poort MP, Avezaat CJJ. Risk factors for infections related to external ventricular drainage. Acta Neurochir (Wien). 2008;150(3):209-214. [DOI] [PubMed] [Google Scholar]
- 11.Hader WJ, Steinbok P. The value of routine cultures of the cerebrospinal fluid in patients with external ventricular drains. Neurosurgery. 2000;46(5):1149-1155. [DOI] [PubMed] [Google Scholar]
- 12.Winkler KML, Woernle CM, Seule M, Held U, Bernays RL, Keller E. Antibiotic-impregnated versus silver-bearing external ventricular drainage catheters: preliminary results in a randomized controlled trial. Neurocrit Care. 2013;18(2):161-165. [DOI] [PubMed] [Google Scholar]
- 13.Rivero-Garvía M, Márquez-Rivas J, Jiménez-Mejías ME, Neth O, Rueda-Torres AB. Reduction in external ventricular drain infection rate. Impact of a minimal handling protocol and antibiotic-impregnated catheters. Acta Neurochir (Wien). 2011;153(3):647-651. [DOI] [PubMed] [Google Scholar]
- 14.Mayhall CG, Archer NH, Lamb VA, et al. Ventriculostomy-related infections. A prospective epidemiologic study. N Engl J Med. 1984;310(9):553-559. [DOI] [PubMed] [Google Scholar]
- 15.Holloway KL, Barnes T, Choi S, et al. Ventriculostomy infections: the effect of monitoring duration and catheter exchange in 584 patients. J Neurosurg. 1996;85(3):419-424. [DOI] [PubMed] [Google Scholar]
- 16.Khanna RK, Rosenblum ML, Rock JP, Malik GM. Prolonged external ventricular drainage with percutaneous long-tunnel ventriculostomies. J Neurosurg. 1995;83(5):791-794. [DOI] [PubMed] [Google Scholar]
- 17.Collins CDE, Hartley JC, Chakraborty A, Thompson DNP. Long subcutaneous tunnelling reduces infection rates in paediatric external ventricular drains. Childs Nerv Syst. 2014;30(10):1671-1678. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Kubilay Z, Amini S, Fauerbach LL, Archibald L, Friedman WA, Layon AJ. Decreasing ventricular infections through the use of a ventriculostomy placement bundle: experience at a single institution: clinical article. J Neurosurg. 2013;118(3):514-520. [DOI] [PubMed] [Google Scholar]
- 19.Chatzi M, Karvouniaris M, Makris D, et al. Bundle of measures for external cerebral ventricular drainage-associated ventriculitis. Crit Care Med. 2014;42(1):66-73. [DOI] [PubMed] [Google Scholar]
- 20.Lewis A, Czeisler BM, Lord AS. Variations in strategies to prevent ventriculostomy-related infections: a practice survey. Neurohospitalist. 2017;7(1):15-23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Zakaria J, Jusue-Torres I, Frazzetta J, et al. Effectiveness of a standardized external ventricular drain placement protocol for infection control. World Neurosurg. 2021;151:e771-e777. [DOI] [PubMed] [Google Scholar]
- 22.R Foundation for Statistical Computing. R: A Language and Environment for Statistical Computing; 2021. Accessed April 30, 2025. https://www.R-project.org/ [Google Scholar]
- 23.Thien A, Soh S, Lock C, et al. The national neuroscience institute external ventricular drain study: a pragmatic multisite risk-stratification pathway to reduce ventriculostomy-related infection. World Neurosurg. 2020;135:e126-e136. [DOI] [PubMed] [Google Scholar]
- 24.Dasgupta D, D’Antona L, Aimone CatD, et al. Simulation workshops as an adjunct to perioperative care bundles in the management of external ventricular drains: improving surgical technique and reducing infection. J Neurosurg. 2019;131(5):1620-1624. [DOI] [PubMed] [Google Scholar]
- 25.Lwin S, Low SW, Choy DKS, Yeo TT, Chou N. External ventricular drain infections: successful implementation of strategies to reduce infection rate. Singapore Med J. 2012;53(4):255-259. [PubMed] [Google Scholar]
- 26.Flint AC, Toossi S, Chan SL, Rao VA, Sheridan W. A simple infection control protocol durably reduces external ventricular drain infections to near-zero levels. World Neurosurg. 2017;99:518-523. [DOI] [PubMed] [Google Scholar]
- 27.Harrop JS, Sharan AD, Ratliff J, et al. Impact of a standardized protocol and antibiotic-impregnated catheters on ventriculostomy infection rates in cerebrovascular patients. Neurosurgery. 2010;67(1):187-191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Hoefnagel D, Volovici V, Dos Santos Rubio EJ, et al. Impact of an external ventricular shunt (EVD) handling protocol on secondary meningitis rates: a historical cohort study with propensity score matching. BMC Neurol. 2023;23(1):36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Katzir M, Lefkowitz JJ, Ben-Reuven D, Fuchs SJ, Hussein K, Sviri GE. Decreasing external ventricular drain-related infection rates with duration-independent, clinically indicated criteria for drain revision: a retrospective study. World Neurosurg. 2019;131:e474-e481. [DOI] [PubMed] [Google Scholar]
- 30.Piccirilli M, Scafa AK, Marchese E, Gallo M, Santoro A. Preventing ventricular catheter displacement and infection with the “catheter-locking device-assisted” technique: a retrospective study of 231 patients. World Neurosurg. 2024;182:e236-e244. [DOI] [PubMed] [Google Scholar]
- 31.Williams TA, Leslie GD, Dobb GJ, Roberts B, Van Heerden PV. Decrease in proven ventriculitis by reducing the frequency of cerebrospinal fluid sampling from extraventricular drains: clinical article. J Neurosurg. 2011;115(5):1040-1046. [DOI] [PubMed] [Google Scholar]
- 32.Choo YH, Shim Y, Kim H, et al. Significant reduction in external ventricular drain-related infections after introducing a novel bundle protocol: a before and after trial. J Korean Med Sci. 2023;38(50):e386. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Talibi SS, Silva AH, Afshari FT, et al. The implementation of an external ventricular drain care bundle to reduce infection rates. Br J Neurosurg. 2020;34(2):181-186. [DOI] [PubMed] [Google Scholar]
- 34.Levy ZD, Ledoux DE, Lesser ML, et al. Rates of iatrogenic ventriculitis before and after the use of an alcohol-impregnated external ventricular drain port cap. Am J Infect Control. 2017;45(1):92-93. [DOI] [PubMed] [Google Scholar]
- 35.Whyte C, Alhasani H, Caplan R, Tully AP. Impact of an external ventricular drain bundle and limited duration antibiotic prophylaxis on drain-related infections and antibiotic resistance. Clin Neurol Neurosurg. 2020;190:105641. [DOI] [PubMed] [Google Scholar]
- 36.Flint AC, Rao VA, Renda NC, Faigeles BS, Lasman TE, Sheridan W. A simple protocol to prevent external ventricular drain infections. Neurosurgery. 2013;72(6):993-999. [DOI] [PubMed] [Google Scholar]
- 37.Champey J, Mourey C, Francony G, et al. Strategies to reduce external ventricular drain–related infections: a multicenter retrospective study. J Neurosurg. 2019;130(6):2034-2039. [DOI] [PubMed] [Google Scholar]
- 38.Huang TF, Su YK, Su IC, et al. Risk, predictive, and preventive factors for noninfectious ventriculitis and external ventricular drain infection. Neurocrit Care. 2024;41(1):109-118. [DOI] [PubMed] [Google Scholar]
- 39.Lackner P, Beer R, Broessner G, et al. Efficacy of silver nanoparticles-impregnated external ventricular drain catheters in patients with acute occlusive hydrocephalus. Neurocrit Care. 2008;8(3):360-365. [DOI] [PubMed] [Google Scholar]
- 40.Phan K, Schultz K, Huang C, et al. External ventricular drain infections at the Canberra Hospital: a retrospective study. J Clin Neurosci. 2016;32:95-98. [DOI] [PubMed] [Google Scholar]
- 41.Rojas-Lora M, Corral L, Zabaleta-Carvajal I, et al. External ventriculostomy-associated infection reduction after updating a care bundle. Ann Clin Microbiol Antimicrob. 2023;22(1):59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Dasic D, Hanna SJ, Bojanic S, Kerr RSC. External ventricular drain infection: the effect of a strict protocol on infection rates and a review of the literature. Br J Neurosurg. 2006;20(5):296-300. [DOI] [PubMed] [Google Scholar]
- 43.Lozano M, Wang AS, Siddiqi I, et al. Extraventricular drain care bundle decreases cerebrospinal fluid infection rates associated with extraventricular drain-related procedures and systemic infection. Cureus. 2024;16(1):e52440. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Korinek AM, Reina M, Boch AL, Rivera AO, De Bels D, Puybasset L. Prevention of external ventricular drain ? Related ventriculitis. Acta Neurochir (Wien). 2005;147(1):39-46. [DOI] [PubMed] [Google Scholar]
- 45.Rahman M, Whiting JH, Fauerbach LL, Archibald L, Friedman WA. Reducing ventriculostomy-related infections to near zero: the eliminating ventriculostomy infection study. Jt Comm J Qual Patient Saf. 2012;38(10):459-464. [DOI] [PubMed] [Google Scholar]
- 46.Sieg EP, Schlauderaff AC, Payne RA, Glantz MJ, Simon SD. Impact of an external ventricular drain placement and handling protocol on infection rates: a meta-analysis and single institution experience. World Neurosurg. 2018;115:e53-e58. [DOI] [PubMed] [Google Scholar]
- 47.Atkinson RA, Fikrey L, Vail A, Patel HC. Silver-impregnated external-ventricular-drain-related cerebrospinal fluid infections: a meta-analysis. J Hosp Infect. 2016;92(3):263-272. [DOI] [PubMed] [Google Scholar]
- 48.Camacho EF, Boszczowski Í, Freire MP, et al. Impact of an educational intervention implanted in a neurological intensive care unit on rates of infection related to external ventricular drains. PLoS One. 2013;8(2):e50708. [DOI] [PMC free article] [PubMed] [Google Scholar]