Challenges in endoscopic third ventriculostomy for patients with achondroplasia: a focus on third ventricle floor anatomy

OBJECTIVE Hydrocephalus is one of the neurological risks occurring in patients with achondroplasia. Ventriculoperitoneal shunt (VPS) insertion is the most common treatment. However, reports of successful endoscopic third ventriculostomy (ETV) suggest that ETV may be a good alternative to VPS insertion in achondroplasia. However, it has been stated that ETV in achondroplasia patients is technically demanding to perform. The current study examined the anatomical variations of the third ventricle and the brainstem in achondroplasia patients and correlated the findings with the difficulty of performing ETV. METHODS A retrospective analysis of 51 patients with achondroplasia and 138 hydrocephalus patients without achondroplasia (48 patients had tumor-related hydrocephalus and 90 patients had hydrocephalus of nontumorous origin) who have visited the authors’ institution since 2012 was performed. Preoperative T2-weighted sagittal MR images were used to measure α (steepness of the third ventricle floor), β (endoscopic angle of incidence), d1 (vertical distance between the dorsum sellae and basilar bifurcation), and d2 (horizontal distance between the dorsum sellae and basilar artery). Each value was compared using the Tukey multicomparison test. RESULTS Achondroplasia patients showed significantly smaller α (p < 0.001) and β (p < 0.001) angles, while there were no significant differences between the control groups (p = 0.947 for α, p = 0.836 for β). The d1 value was significantly larger in achondroplasia patients (p < 0.001), and d2 was smaller (p < 0.001). The control groups showed similar d1 and d2 values (p = 0.415 for d1, p = 0.154 for d2). Smaller α and β values meant that in achondroplasia patients the third ventricle floor stood more vertically than in other patients with hydrocephalus, and the endoscopic contact angles were small, increasing the risk of ventriculostomy devices slipping down into the infundibular recess. Additionally, a large d1 meant that the basilar artery was shifted upward and a small d2 indicated that the basilar artery was located closer to the dorsum sellae, potentially increasing the risk of basilar artery damage. CONCLUSIONS Achondroplasia patients’ skull and brain anatomies were significantly different from those of other hydrocephalus patients, with steeper third ventricle floors and basilar arteries closer to the dorsum sellae. Because these anatomical differences lead to difficulties in performing ETVs in achondroplasia patients, such differences should be considered when ETV is planned for the patients.