Abstract
Objective: Evaluate an ellipsoid model of vestibular schwannoma (VS) tumor volume based on linear dimensions.
Method: MRI scans from 97 patients with VS from 2003 to 2011 were analyzed. Linear measurements of major axes defining an ellipsoid within the cistern and a cone within the internal meatus were made. From those dimensions, geometrically based conformational models of tumor volume were evaluated.
Results: Model estimates of tumor volume were compared to measurements made using software utility in Brainlab. Seventy-two MRI scans were used for model optimization and 25 scans to validate model predictions and performance. Optimal model selection was determined based on the best trade-off between time required to obtain measurements and the model’s ability to capture volumetric variance. Analyses showed the optimal tumor model as one where the cisternal component was modeled as an ellipsoid and the canalicular portion was modeled as a cylinder (r = 0.96, P < .001). This model was robust for a broad range of tumor sizes.
Conclusion: Expected outcomes following treatment of vestibular schwannomas are dependent on tumor size, which has been variably described throughout the literature both by tumor volume and linear dimension of canalicular and cisternal components. Our model can be used to invert volumetric data to linear indices of volume. This tool will enable meta-analysis of disparate data sets.
Get full access to this article
View all access options for this article.