Object: Treatment of ventriculoperitoneal shunt infections frequently requires placement of an external ventricular drain (EVD). Surveillance specimens obtained from antibiotic-impregnated (AI) EVDs may be less likely to demonstrate bacterial growth, potentially resulting in undertreatment of an infection. The purpose of this study was to assess whether AI EVDs had any significant effect on bacterial culture results compared with nonantibiotic-impregnated (NAI) EVDs.
Methods: In vitro assays were performed using AI EVDs containing minocycline and rifampin (VentriClear II, Medtronic) and NAI EVD controls (Bioglide, Medtronic). The presence of antibiotics was evaluated via capillary electrophoresis of sterile saline drawn from AI and NAI EVDs after predefined incubation intervals. Antimicrobial activity was assessed by evaluating zones of inhibition created by the catheter aspirates on plates inoculated with a quality control strain of Staphylococcus epidermidis (American Type Culture Collection strain 12228). To determine the effects of cultures drawn through AI compared with NAI EVDs, the quality control strain was then incubated within 4 new AI and 4 new NAI EVDs for predefined intervals before being plated on culture media. Spread and streak plate culture results from each type of catheter were compared at each time interval.
Results: Capillary electrophoresis showed that more minocycline than rifampin was eluted from the AI EVDs. Sterile saline samples incubated within the AI EVDs demonstrated zones of growth inhibition when placed on plates of S. epidermidis at all time intervals tested. No zones of inhibition were noted on NAI EVD control plates. When a standardized inoculum of S. epidermidis was drawn through AI and NAI EVDs, antimicrobial effects were observed after incubation in the AI EVD group only. Colony counting demonstrated that significantly fewer colonies resulted from samples drawn through AI compared with NAI EVDs at the multiple time intervals. Similarly, streak plating yielded a statistically significant number of false-negative results from AI compared with NAI EVDs at 2 time intervals.
Conclusions: The findings in the current study indicate that the risk of a false-negative culture result may be increased when a CSF sample is drawn through an AI catheter. In the management of a known shunt infection, a false-negative result from an EVD culture specimen may lead to an inappropriately short duration of antibiotic therapy. These data have significant clinical implications, particularly given the widespread use of AI drains and the current high rates of shunt reinfection after EVD use worldwide.