ABSTRACT
BACKGROUND: External ventricular drain (EVD)-associated cerebrospinal fluid infection (EACI) remains a major complication associated with EVD. Length of EVD tunnel, an overlooked but modifiable factor, can be associated with increased risk of EACI. The aim of this study is to find the tunnel length associated with least chances of EACI by performing a network meta-analysis. METHODS: A comprehensive search of different databases was performed to retrieve studies that studied the rates of EACI with different EVD tunnel lengths and a Bayesian network meta-analysis was performed. RESULTS: Six studies met the inclusion criteria and were included in the network meta-analysis. With 0 cm tunnel length as reference, the odds ratio (OR) for developing EACI was minimum for tunnel length 5-10 cm (OR, 0.027). It was followed by tunnel length of 5 cm (OR, 0.060) and 10 cm (OR, 0.075). The surface under the cumulative ranking curve plot showed that the probability of the tunnel length 5-10 cm (ranked first), 5 cm (ranked second), and 10 cm (ranked third) for being the best EVD tunnel length was found to be 86%, 64%, and 61%, respectively. CONCLUSIONS: The length for which an EVD is tunneled may have an impact on the rate of EACI. Our network meta-analyses showed that the tunnel length of 5-10 cm was associated with the lowest rates of EACI, with 86% probability of being the best EVD tunnel length. The probability of a patient with 5-10 cm EVD length developing EACI was 2.7% compared with zero tunnel length.