Observational Retrospective Cohort Two Centred Study on External Ventricular Drain-Related Infections in US and Turkey.

Aim: To compare external ventricular drain-related infection (EVD-RI) rates of two Academic Medical Centers in Turkey and the US in order to determine the key factors.

Material And Methods: We performed an observational retrospective cohort study to compare the EVD-RI rates between two hospitals (Hospital T in Turkey; Hospital A in US). We analyzed data gathered from 736 patients (Hospital T, n=237; Hospital A, n=499), in a total of 736 EVD cases which occurred between January 1, 2013 and December 31, 2018. Electronic records of hospitals were searched according to the procedure code ?external ventricular drain?. The study protocol was approved by the audit and clinical governance committees of both participating hospitals.

Results: EVD-RI rates were 18.3 per 1000 days for a total of 3227 catheter days in Hospital T, whereas 4.0 per 1000 cases for a total of 7010 catheter days in Hospital A. The prolonged use of EVD catheter, length of stay, and frequency of number of cerebrospinal fluid sampling were associated with EVD-RI in both hospitals. Cerebrospinal fluid culture of Hospital A was dominant for gram positive 32.6% and gram negative 46.1% rods, whereas for Hospital T, gram positive 39.0% and gram negative 33.9% rods were the main microorganisms for the EVD-RI. There was a correlation, between the duration of antibiotic use and EVD-RI in Hospital T. The nursing care facilities of the hospitals were significantly different. All-cause mortality found to be 12.0% versus 3.7% in Hospital A and Hospital T respectively (p < 0.05).

Conclusion: In both institutions, the EVD-RI was associated with EVD characteristics rather than patients? personal characteristics. Early drain removal and patient discharge should be goals whenever medically appropriate. The wide range infection and mortality rates between Hospital A and T is most likely attributable to many factors, including the variety of populations evaluated and the use of dissimilar methodologies in the management of EDVs.