Enhancing Surveillance for Infections Associated with External Ventricular Devices in a University Hospital

ISSUE: External ventricular devices (EVD) are commonly used in acute neurosciences patients. They can be associated with infections due to prolonged insertion times, or deficient insertion and manipulation techniques. Historically, our cases of nosocomial ventriculomeningitis were determined based on positive cerebrospinal fluid (CSF) cultures in patients who met the National Nosocomial Infection Surveillance System (NNIS) definition for meningitis; however, EVD-associated infections have not been specifically defined. Due to cliniciansʹ perceived high rates of EVD-associated CSF infections, we established a prospective surveillance program to assess the burden of EVD-associated infections. PROJECT: In order to develop a standardized surveillance plan for EVD-associated CSF infections, a multidisciplinary team involving Infection Control and Epidemiology, Neurology, Anesthesiology, and Neurosurgery developed a novel surveillance definition based upon literature review and current NNIS definitions. This new definition included the presence of an EVD for at least 24 hours before symptom onset. We also began counting the number of EVD days so a rate could be calculated. Using the definition, infection control practitioners prospectively monitored all potential CSF infections in patients with an EVD to ascertain the number of EVD-associated CSF infections, report our findings to physician leaders, and ultimately decrease these infections through targeted prevention initiatives. RESULTS: From August 2004 to November 2005, fourteen EVD-associated infections were identified, compared to 2 infections noted in the 3 years prior to establishment of formal surveillance. The infection rate for the surveillance period was 12.7 infections per 1000 device days. There were no comparison rates since device days were not counted prior to project implementation. Feedback of infection data to frontline clinicians led to increased CSF cultures in patients with an EVD, which led to dedicating a nurse practitioner to collect CSF samples. LESSONS LEARNED: Through development of a standardized definition for EVD-associated infections and a targeted surveillance program, additional device-associated infections were detected. Continued monitoring will enable us to calculate rates for comparison. The project also provided a foundation for a new infection prevention initiative.