Craniopharyngioma Classification.
Liangxue Zhou,Chao You
DOI: https://doi.org/10.3171/2009.2.jns081430
IF: 5.408
2009-01-01
Journal of Neurosurgery
Abstract:Craniopharyngiomas (CPs) are among the most surgically complex brain tumors due to their deep location and intimate involvement with critical neural and vascular structures. Furthermore, injury to the adjacent hypothalamus, infundibulum, pituitary gland, optic apparatus, or vasculature can have devastating consequences including hypothalamic obesity, autonomic dysregulation, panhypopituitarism, blindness, and major strokes. Their location, arising from the sella up through the third ventricle, can be accessed via multiple surgical approaches. Broadly, the approaches include variants of the transcallosal, transsylvian, subfrontal, and endonasal approach, all of which are viable options for CP resection. Determining the ideal approach has been a source of lively debate among neurosurgeons for many years, and as such, several classification systems based on surgeon experience have been proposed, including the classification by Fan et al. in the current issue.1 The ideal classification system is one that would predict functional outcomes, extent of resection, and recurrence rates, based on tumor characteristics, patient anatomy, and approach. This would allow the neurosurgeon to select the approach that would provide the best outcome for each patient’s unique tumor. The first classification was put forward in 1990 by Gazi Yaşargil based on his microsurgical experience with 144 CPs (Table 1).2 His classification system was based on the relationship of the tumor with the surrounding anatomical structures, foremost of which were the diaphragma sellae, the tuber cinereum/floor of the third ventricle, and the third ventricle/hypothalamus. The primary challenge during this era was in preoperative diagnostic imaging to understand relationships of the tumor with these structures, because MRI was a relatively new technology at the time. Nevertheless, this anatomical relationship was key to guiding the surgical approach. The pterional approach was used most frequently by Yasargil because it allowed early identification of the stalk, anterior circulation, and protection of the optic chiasm. When tumors extended superiorly in the third ventricle, he would combine the pterional with a transcallosal approach. He used the transcallosal approach for tumors primarily involving the third ventricle and reserved transsphenoidal approaches only for intrasellar infradiaphragmatic (type A) tumors. Additionally, the relationship between the optic chiasm and sella is important for approach selection, whether the chiasm was prefixed (over the tuberculum), over the middle of the sellae (normal), or postfixed (over the dorsum sellae).3 CPs that are considered prechiasmatic (postfixed) push the optic chiasm posteriorly and can be resected via a subfrontal approach, whereas those that are retrochiasmatic push the chiasm anteriorly toward the tuberculum sellae (prefixed), making a subfrontal approach more challenging because most of the tumor is covered by the optic apparatus. During the midto late 1990s and early 2000s, advances in endoscopy and improved instrumentation spurred the development of expanded endonasal approaches.4 The development of the vascularized nasal septal flap provided a much-needed improvement in reconstruction techniques that had been hampered by postoperative CSF leaks,5 and allowed the expanded endonasal approach to be considered for CPs that extended supradiaphragmatically and even up into the third ventricle.6 As the endonasal approach matured and became effective for managing lesions that extended beyond the sella, a new classification system was needed to account for the expanded surgical access. As the expanded endonasal approach developed, it was appreciated that the amount of additional exposure needed to access a tumor was determined by the relationship between the tumor and the infundibulum, which led Kassam and colleagues to develop their classification scheme that may be used to tailor the endonasal approach for resection of CPs (Table 1).7 Because the axis of the endonasal approach parallels the infundibulum, working below the optic chiasm, the position of the chiasm relative to the sella became less important than when coming from a transcranial approach. Tumors with a preinfundibular location (type I) benefit from a transplanum approach. Type II CPs