Hi, my name is Eric. I'm one of the neurosurgeons at Temple University Hospital in Philadelphia, presenting data regarding the effect of ventricular catheter type and drainage system gentamicin flushes on ventriculoscopy associated infections. We have no disclosures. EVD placement is a common neurosurgical procedure utilized for ICP monitoring, as well as therapeutic CSF diversion. There is institutional variation and protocols to optimize EVD placement and management. One of the unified aims is to minimize infection rates with various strategies, including the use of antibiotic impregnated ventricular catheters, as well as dressing methods. At Temple, we have utilized both antibiotic and non-antibiotic impregnated ventricular catheters, depending largely on the presence of intraventricular hemorrhage. In these situations, a non-antibiotic impregnated catheter with a larger inner diameter would be placed with the thought that this would help maintain a patent system. At Temple, we have also used surgical glue to dress the EVD incision and exit site, based on a prior study performed at our institution in 2012, which demonstrated a decrease in infection rate. We have also utilized gentamicin flush solution only within the distal drainage system. In theory, none of the flush should be reaching the ventricular catheter or the patient. In order to promote antibiotic stewardship, our ICU ceased utilization of gentamicin flushes, given the lack of supporting evidence. During this period, a new FDA-approved non-antibiotic impregnated floral oligomeric polymer-coated ventricular catheter, designed to be antithrombogenic, was introduced into our practice as well. We then observed a subsequent rise in EVD infections, prompting a review of our EVD practices. We then performed a retrospective review of all EVDs maintained in the ICU from June 2017 through August 2019. Dates for implementation of the flush and FOP catheters are as listed. We excluded patients with infection as indication for EVD placement. CSF sampling was only performed if patient developed fevers or otherwise demonstrated clinical concern for ventriculitis. Infections were defined as positive CSF cultures, and then recorded other various data variables. Eight infections were identified of the 123 ventricular drains reviewed. One of the infections was with an antibiotic impregnated catheter. Two patients had gunshot wounds to the head. Almost all had intraventricular hemorrhage. And the most common organism isolated was staph epidermidis. We then looked at the distribution of catheter types, infection percentage within that particular catheter type, and percentage of catheters within that group that received gentamicin flushes. We saw that the FOP catheters infection rate of 29.41% was significant. We found that non-antibiotic impregnated catheters as a whole had significantly increased infection frequency. But this was not significant if we removed the FOP catheter from analysis. We found that percentage of gentamicin flush was not significantly different between the groups. When looking at the gentamicin flushes and infection rates, we did find that there was a trend towards decreased infections with the gentamicin flushes. We then broke down the different combinations between catheter types and presence of gentamicin flushes. We see on the far right that the FOP catheters without gentamicin flush comprise the smallest percentage of overall catheters, but accounted for 37.5% of all infections. We then performed logistic regression analysis. Controlling for gentamicin flushes, the FOP catheter had a significant 23.32% increased odds of infection compared to the antibiotic impregnated catheters. Lack of gentamicin flush demonstrated increased odds ratio of infection, but this was not statistically significant. Interestingly, the gentamicin flushes becoming a significant factor in non-antibiotic impregnated catheter situation requires further evaluation, particularly if other catheter qualities may be beneficial and subsequently reduce the risk and complications associated with manipulating the system or replacing a catheter manipulating the system or replacing occluded ventricular drains. There's limited power to the study with eight infections and six different combinations of catheter types and presence of gentamicin flushes. We could have a deeper look at the role pathology may play as well as developing a more reliable method to track access of the ventricular system. Our data is consistent with the growing body literature that an antibiotic impregnated catheter is critical to an infection reducing EVD protocol. We'd like to thank doctors Fisher and Ramsey for their aid in statistical analysis. Thank you for your time.

neurosurgeon

ventricular catheter

gentamicin flushes

ventriculoscopy-associated infections

antibiotic-impregnated catheters