Rate and Risk Factors for Shunt Revision in Pediatric Patients with Hydrocephalus—A Population-Based Study

• Published in: World neurosurgery Vol. 101; pp. 615 - 622
• Main Authors: Tervonen, Joona, Leinonen, Ville, Jääskeläinen, Juha E., Koponen, Susanna, Huttunen, Terhi J.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.05.2017
• Subjects: Child; Child, Preschool; Female; Finland - epidemiology; Follow-Up Studies; Humans; Hydrocephalus; Hydrocephalus - diagnosis; Hydrocephalus - epidemiology; Hydrocephalus - surgery; Male; Pediatric; Population Surveillance - methods; Population-based; Registries; Reoperation - trends; Retrospective Studies; Revisions; Risk Factors; Shunting; Ventriculoperitoneal Shunt - trends; CD; Hydrocephalus; MRI; Shunting; Revisions; Population-based; EVD; CSF; VPS; IVH; HC; Pediatric; KUH
• ISSN: 1878-8750; 1878-8769
• DOI: 10.1016/j.wneu.2017.02.030

Ventriculoperitoneal shunt (VPS) is a common treatment for patients with hydrocephalus (HC). VPS is associated with complications that may lead to shunt revisions. We studied the surgical outcome of pediatric patients with HC in a population-based setting. The medical charts and imaging findings of 80 patients ≤16 years old who required VPS secondary to HC were studied. Mean age at time of initial shunt placement was 3.2 years (SD 4.5) and mean follow-up time was 3.3 years (SD 2.9); 57% of patients were male. Half of patients underwent shunt revision with mean time to first revision of 8 months. Patients ≤6 months old had a higher shunt revision rate compared with patients >6 months old (P < 0.001). The most common causes of HC requiring VPS were tumors (27.5%), congenital defects (22.5%), and intraventricular hemorrhage (19%). Revision rates in the intraventricular hemorrhage and congenital defects groups were 67% (P = 0.017) and 72% (P = 0.016) compared with 32% in the tumor group. Programmable valves (56%) were more common than nonprogrammable valves, but there was no significant difference in shunt survival (P = 0.632). The mean biparietal measurement change between preoperative and postoperative images was +0.9 mm in the no revision group and +6.6 mm in the revision group (P = 0.003). Half of patients with shunts required revision. Age ≤6 months and intraventricular hemorrhage and congenital defects etiologies of HC were associated with increased risk for shunt revision. Most revisions were done during the first year after the initial VPS.