Central Nervous System Infections in Patients With Ventriculoperitoneal Shunts Admitted for Primary Abdominal Infections: A US Nationwide Cohort Analysis

Abstract

BACKGROUND AND OBJECTIVES: Although ventriculoperitoneal (VP) shunts are a common treatment for hydrocephalus, there are complication risks including infections. Late complications such as ventriculitis from ascending abdominal infections can have severe consequences. However, the incidence of central nervous system (CNS) infections in VP shunt patients with abdominal infections is not well understood. We aimed to determine the incidence of CNS infections in VP shunt patients admitted with abdominal infections. METHODS: Using the National Inpatient Sample, we studied patients from 2016 to 2019 to determine the incidence of CNS infections in VP shunt patients admitted with abdominal infections. Results were compared with VP shunt patients admitted for primary pneumonia. RESULTS: Among 725 VP shunt patients presenting with abdominal infections, 20 (2.8%) had CNS infections. Chronic obstructive pulmonary disease, hypertension, older age, and a smoking history were more common in patients with CNS infections and primary abdominal infection (P < .05). Patients who developed CNS infection had a significantly higher likelihood of both blood transfusion and coma but a lower likelihood of seizures. VP shunt patients with CNS infections were more likely to undergo shunt removal (odds ratio [OR] = 23.167, P < .001). 4.1% of VP shunt patients with primary abdominal infections died during admission. In a population of primary abdominal infection and pneumonia patients with VP shunts, a multivariate logistic regression analysis controlling for age, sex, and comorbidities identified abdominal infection as an independent risk factor for both CNS infection (OR = 51.208, P < .001) and inpatient death (OR = 3.417, P < .001). Among 6620 VP shunt patients admitted with primary pneumonia, only 5 (0.1%) had CNS infection compared with 20 (2.8%) in those with a primary abdominal infection (OR = 37.532, P < .001), and mortality was 1.6% vs 4.1% for those with a primary abdominal infection (P < .05). CONCLUSION: CNS infections in VP-shunted patients with abdominal infections are relatively rare but may lead to increased risks of death and other serious adverse outcomes.