Analysis of the potential risk of central intravenous lines and/or total parenteral nutrition with ventriculoatrial shunts.

objectThe distal catheter of a ventriculoatrial (VA) cerebrospinal fluid shunt is potentially exposed to bacterial seeding from a subclavian central line. The risk of blood stream infections (BSIs) from central lines increases with administration of total parenteral nutrition (TPN). The potential risks of shunt malfunction or infection in patients with a VA shunt and a concurrent subclavian central line and/or TPN administration have not been studied.

methodsA retrospective review of 49 pediatric patients with placement of a VA shunt was performed. Three outcome measures were studied: shunt malfunction, shunt infection, and bacteremia/fungemia requiring shunt removal. All outcomes were measured by 1 year after shunt insertion. We analyzed the following potential risk factors: age at shunt insertion, prior ventriculoperitoneal (VP) shunt, prior shunt infection, abdominal infection/necrotizing enterocolitis (NEC), concurrent subclavian central line, and administration of TPN. The association between each risk factor and outcome was evaluated using Fisher's exact test to generate the relative risk. Additionally, a logistic regression analysis was performed to evaluate the odds ratio of the outcomes to risk factors considering age as a covariate.

resultsThe average age at shunt insertion was 6.3 ± 7.6 years. The most common diagnosis was posthemorrhagic hydrocephalus of prematurity (53.1 %). Fifteen patients (30.1 %) had a shunt malfunction within 1 year, 6 (12.2 %) had a shunt infection, and 3 (6.1 %) required removal of the shunt due to bacteremia/fungemia. The age at shunt insertion was not a statistically significant independent risk factor for any of the three outcomes. Prior shunt infection predicted an increased risk for both future shunt malfunction and infection in both the associative relative risk analysis and the age-dependent logistic regression analysis, although the correlation did not reach statistical significance. The presence of a subclavian central line or TPN administration did not statistically increase the risk over baseline for any of the outcomes in either analysis.

conclusionsThe relatively small number of patients limits the power of the study. Considering this limitation, the data suggests that the presence of a concurrent subclavian central line or administration of TPN does not increase the risk of shunt malfunction or infection over the baseline of this high-risk cohort.