Year-end Sale % OFF 
Abstract
Communicating hydrocephalus (HCP) in vestibular schwannomas (VS) after gamma knife radiosurgery (GKRS) has been reported in the literature. However, little information about its incidence and risk factors after GKRS for intracranial schwannomas is yet available. The objective of this study was to identify the incidence and risk factors for developing communicating HCP after GKRS for intracranial schwannomas. We retrospectively reviewed a total of 702 patients with intracranial schwannomas who were treated with GKRS between January 2002 and December 2015. We investigated patients’ age, gender, tumor origin, previous surgery history, tumor volume, marginal radiation dose, and presence of tumor control to identify associations with communicating HCP following GKRS. To make predictive models of communicating HCP, we performed Cox regression analyses and constructed a decision tree for risk factors. In total, 29 of the 702 patients (4.1%) developed communicating HCP following GKRS, which required ventriculo‐peritoneal (VP) shunt surgery. Multivariate analyses indicated that age (P = 0.0011), tumor origin (P = 0.0438), and tumor volume (P < 0.0001) were significant predictors of communicating HCP in patients with intracranial schwannoma after GKRS. Using machine‐learning methods, we fit an optimal predictive model. We found that developing communicating HCP following GKRS was most likely if the tumor was vestibular origin and had a volume ≥13.65 cm3. Communicating HCP is not a rare complication of GKRS for intracranial schwannomas. Under specific conditions, communicating HCP following GKRS is warranted for this patient group, and this patient group should be closely followed up.
Highlights
Symptomatic hydrocephalus (HCP) resulting from vestibular schwannoma (VS) is frequently observed, especially among patients with an obstruction in their cerebrospinal fluid (CSF) pathway caused by a large tumor [1, 2], requiring ventriculo-p eritoneal (VP) shunt placement [3]
We believe that our investigations reinforce the value of optimal machine-learning approaches for predictive studies, which helps identify the patients with HCP risk factors before they undergo gamma knife radiosurgery (GKRS)
We evaluated eight classification methods (nearest neighbors classifier, support vector classifier (SVC), decision tree, Random forest classifier (RF), AdaBoost Classifier, Gaussian naive Bayes (GNB), linear discriminant analysis (LDA), and gradient boosting (GB)) for their predictive performance and stability against data perturbation
Summary
Symptomatic hydrocephalus (HCP) resulting from vestibular schwannoma (VS) is frequently observed, especially among patients with an obstruction in their cerebrospinal fluid (CSF) pathway caused by a large tumor [1, 2], requiring ventriculo-p eritoneal (VP) shunt placement [3]. We found that large volume tumors and tumors undergoing temporary size changes were correlated with higher incidence of communicating HCP [10]. We attempted to identify the incidence and possible risk factors of communicating HCP after GKRS using our data registry of patients with intracranial schwannomas. We applied “machine-learning” methods to make more accurate predictions of likelihoods for developing communicating HCP following GKRS for intracranial schwannomas. We explored a large panel of machine- learning approaches for clinicoradiological data from patients who developed communicating HCP. We believe that our investigations reinforce the value of optimal machine-learning approaches for predictive studies, which helps identify the patients with HCP risk factors before they undergo GKRS
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
