Objective: To evaluate risk factors and predictors of cerebrospinal ventricular shunt revisions in children.
Methods: A retrospective, longitudinal cohort of 1307 children ages 0 to 18 years undergoing initial ventricular shunt placement in the year 2000, with follow-up through 2005, from 32 freestanding children's hospitals within the Pediatric Health Information Systems database was studied. Rates of ventricular shunt revision were compared with patient demographic, clinical, and hospital characteristics with use of bivariate and multivariate regression accounting for hospital clustering.
Results: Thirty-seven percent of children required at least one shunt revision within 5 years of initial shunt placement; 20% of children required two or more revisions. Institutional rates of first shunt revision ranged from 20 to 70% of initial shunts placed among the 32 hospitals in the cohort. Hospitals where one to 20 initial shunt placements per year experienced the highest initial shunt revision rate (42%). Hospitals performing over 83 initial shunt placements per year experienced the lowest revision rate (22%). We found that children undergoing shunt placement in the Midwest were more likely to experience multiple shunt revisions (odds ratio, 1.25; 95% confidence interval, 1.06-1.47) after controlling for hospital volume, shunt type, age, and diagnosis associated with initial shunt placement.
Conclusion: Higher hospital volume of initial shunt placement was associated with lower revision rates. Substantial hospital variation in the rates of ventricular shunt revision exists among children's hospitals. Future prospective studies are needed to examine the reasons for the variability in shunt revision rates among hospitals, including differences in specific processes of care.