The effect of institutional learning curve on paediatric ventricular shunt survival: a retrospective cohort study from a new paediatric neurosurgical centre

Abstract

INTRODUCTION: Ventricular shunt procedures and their outcomes are commonly considered a benchmark for the performance of a neurosurgical service. The influence of an institutional learning curve on the outcomes of neurosurgical procedures has not been established; however, it is important when considering the establishment of neurosurgical centres or the introduction of new operating room teams. A benchmark high-volume procedure such as paediatric ventricular shunt insertion represents an ideal model to study this phenomenon. METHODS: We conducted a retrospective review of a prospectively maintained surgical database, covering 10 years of consecutive cases from a new quaternary Australian paediatric neurosurgical centre. Patients undergoing insertion of a new ventricular shunt were included, while all cases of shunt revision were excluded. The 10-year study period was divided into four eras of 2.5-year duration for analysis. RESULTS: A total of 427 new ventricular shunt systems were inserted in the initial 10 years since the hospital's commissioning. There was a significant improvement in overall time to shunt revision from the first era to the second, however no further change thereafter (median overall shunt survival 392 days vs not reached in other eras, p < 0.001). Similarly, time in the operating room improved from the first to the second era and reached a steady state thereafter (mean 129 ± 46 min in years 0-2.5 and 103 ± 44 min in years 2.5-5 (p < 0.001)). CONCLUSION: The institutional learning curve for new ventriculoperitoneal shunt placement appears to plateau within 2.5 years of commissioning a new paediatric neurosurgical hospital.