A Consideration of Potential Modifications in Current Grading Scales of Craniopharyngioma in Children

Background There is no classification of craniopharyngiomas that adequately assesses the potential impact of posterior-inferior tumor projection, tumor lateralization, or both, on the clinical outcome following surgical resection. Currently, most classification systems of craniopharyngiomas are based on a linear, vertical projection of tumor growth with respect to the sella turcica, the optic chiasm, and the floor of the third ventricle. We propose a modification to existing classifications to better describe tumor extension, which should allow for a better understanding of morbidity/mortality related to the resection and may clarify the most appropriate surgical approach.

Methods A retrospective chart review was performed on the senior author's series of pediatric craniopharyngioma cases. Only cases with complete demographic, operative, and radiographic data were considered for data analysis. Cases were radiographically classified by a newly developed classification scale that took into consideration posterior-inferior, lateral, or both projections of craniopharyngioma based on MRI imaging. Clinical outcome data with regard to hypothalamic injury, endocrinologic injury, new-onset diabetes insipidus (DI), vision deterioration, and new or worsening weakness were examined.

Results In our single-institution chart review of first-time operations for craniopharyngiomas, we identified 46 pediatric patients from 2004 to 2017. There were 70% males (mean age = 90.6 ±3.9 months). Thirty-three of these cases were operated through expanded skull base approaches (frontotemporal-orbitozygomatic or subfrontal–transbasal approaches). Thirteen were operated through traditional skull base approaches (pterional, transsphenoidal, or interhemispheric approaches). For craniopharyngiomas that were contained to the intrasellar, suprasellar regions, or third ventricle, there was no difference in extent of resection, postoperative visual deficits, or clinical hypothalamic injury (weakness, DI, endocrinologic dysfunction) when comparing the traditional versus expanded skull base approaches. However, in the 17 patients who had posterior or lateral extension of tumors, the expanded skull base approaches were seen to have a higher likelihood of gross total resections (69.2 vs. 0%, p = 0.03), less postoperative visual deficits (30.7 vs. 100%, p = 0.03), less hypothalamic injuries (23.1 vs. 100%, p = 0.02), and less motor deficits (8 vs. 75%, p = 0.02). There was no difference in postsurgical onset of DI (38.4 vs. 75%, p = 0.29).

Conclusion Lateralization and posterior-inferior projection of craniopharyngiomas require an expanded skull base approach to obtain a safer, more thorough resection. For craniopharyngiomas within the sella, suprasellar cistern, or third ventricle, an expanded skull base approach had shown no benefit with regard to achieving a gross total resection or reducing postoperative morbidity. We suggest the addition of subclassifications of lateralization, posterior extent, and both to Yasargil Grade III, IV, and V tumors. Such classifications would more accurately describe tumor extent and complexity, and provide for a better understanding for surgical approach to improve extents of resection and reducing postoperative morbidity when these variables are present.