Cerebrospinal fluid shunt infection: risk factors and long-term follow-up

Shunt infection (SI) is an enduring problem in pediatric neurosurgery. Its occurrence is variable in the different series that were published, according to the definition retained. In addition, long-term data, which could help to evaluate the incidence of delayed SI, as well as the developmental outcome after SI, are scarce in the literature. We reviewed retrospectively children shunted for hydrocephalus during the last 20 years to evaluate the incidence of SI, including late SI, the risk factors and sources of contamination, and the late outcome after SI. We treated 1,173 patients who were followed-up for a mean duration of 7.0 years. During that period, 158 patients presented with a total number of 190 episodes of infection, 19 of which occurred more than 1 year after surgery. The infection rates per patient and per procedure were 13.6 and 5.9%, respectively. Age below 4 months at shunt insertion [odds ratio (OR)=1.81], antenatal diagnosis (OR=2.23), myelomeningocele (OR=2.14), and post-hemorrhagic hydrocephalus (OR=1.98) were significantly correlated with SI. SI was mostly due to intraoperative contamination; however, delayed SI was mostly caused by blood-borne contamination and abdominal sepsis. The mortality related to SI was 10.1%; the Glasgow Outcome Score, as well as schooling, was significantly and independently affected by SI. Long-term follow-up of shunted children is necessary to evaluate the real incidence of SI and the functional outcome after SI.