Stefan Yordanov1, M R Garnett2, T Santarius2, K Holland2, I Jalloh2, M Jawad Naushahi2. 1. Academic Division of Neurosurgery, Addenbrookes' Hospital, Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK. stefan.yordanov@addenbrookes.nhs.uk. 2. Academic Division of Neurosurgery, Addenbrookes' Hospital, Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK.
Abstract
BACKGROUND: Endoscopic third ventriculostomy success score (ETVss) is widely utilised to predict outcomes for ETV. Accurate prediction of success for a procedure is of vital importance both for selecting the optimal management plan and for obtaining informed consent. Existing literature demonstrates a variety of opinions on the accuracy of the currently utilised ETVss and recommends a range of techniques to reduce the number of subsequent ventriculo-peritoneal (VP)-shunt insertions, prompting the present study. METHODS: We retrospectively analysed data for ETV cases since 2007 to review success rate in our regional paediatric neurosurgical centre and if the currently utilised ETVss successfully predicted outcomes. Failed ETV cases were defined as any patient who received a VP-shunt at any time following ETV. Data was analysed with MS ExcelR and RStudioR. RESULTS: 44 ETVs were performed over 13 years with approximately equal distribution between male and female patients; median age 7 years (IQR 4-13 years). Overall, mean ETVss for these 44 procedures was 78%; actual success rate was 70% with no statistically significant difference between them (p = 0.286; Welch two sample t-test). Accuracy of ETVss varied with pathology: tectal gliomas (mean ETVss 75% and actual success 78%); cerebellar tumours (mean ETVss 85% and actual success 81%); other tumours (mean ETVss 75% and actual success 81%); aqueduct stenosis (mean ETVss 71% and actual success 69%); and other pathologies (mean ETVss 70% and actual success 60%). < 1 month and 1-6 months and 1-10 years and > 10 years contributed equally to the accuracy of ETVss. CONCLUSION: Non-telencephalon tumours and obstruction at the level of the mid-brain are most strongly associated with successful ETV outcome. These findings can be used to modify the currently utilised ETVss to further improve accuracy of outcome prediction. We recommend a modified-ETVss (m-ETVss) and a future larger adequately powered prospective study to validate this.
BACKGROUND: Endoscopic third ventriculostomy success score (ETVss) is widely utilised to predict outcomes for ETV. Accurate prediction of success for a procedure is of vital importance both for selecting the optimal management plan and for obtaining informed consent. Existing literature demonstrates a variety of opinions on the accuracy of the currently utilised ETVss and recommends a range of techniques to reduce the number of subsequent ventriculo-peritoneal (VP)-shunt insertions, prompting the present study. METHODS: We retrospectively analysed data for ETV cases since 2007 to review success rate in our regional paediatric neurosurgical centre and if the currently utilised ETVss successfully predicted outcomes. Failed ETV cases were defined as any patient who received a VP-shunt at any time following ETV. Data was analysed with MS ExcelR and RStudioR. RESULTS: 44 ETVs were performed over 13 years with approximately equal distribution between male and female patients; median age 7 years (IQR 4-13 years). Overall, mean ETVss for these 44 procedures was 78%; actual success rate was 70% with no statistically significant difference between them (p = 0.286; Welch two sample t-test). Accuracy of ETVss varied with pathology: tectal gliomas (mean ETVss 75% and actual success 78%); cerebellar tumours (mean ETVss 85% and actual success 81%); other tumours (mean ETVss 75% and actual success 81%); aqueduct stenosis (mean ETVss 71% and actual success 69%); and other pathologies (mean ETVss 70% and actual success 60%). < 1 month and 1-6 months and 1-10 years and > 10 years contributed equally to the accuracy of ETVss. CONCLUSION: Non-telencephalon tumours and obstruction at the level of the mid-brain are most strongly associated with successful ETV outcome. These findings can be used to modify the currently utilised ETVss to further improve accuracy of outcome prediction. We recommend a modified-ETVss (m-ETVss) and a future larger adequately powered prospective study to validate this.
Authors: James Drake; Paul Chumas; John Kestle; Alain Pierre-Kahn; Matthieu Vinchon; Jennifer Brown; Ian F Pollack; Hajime Arai Journal: J Neurosurg Date: 2006-08 Impact factor: 5.115