Introduction: Transsphenoidal surgery has eclipsed open transcranial surgery as the preferred operative management of large pituitary adenomas. For tumors with large suprasellar components, however, craniotomy may be advantageous in selected patients. Because visual loss due to chiasmal compression is typically the primary reversible condition, surgical decision-making is predicated on which approach will best decompress the optic apparatus. With regard to an endoscopic endonasal approach, the surgeon must assess the probability of diaphragmatic descent. In this study, we aimed to identify which pre-operative neuroimaging features predict failure of intra-operative diaphragmatic descent.
Methods: A retrospective study of all patients who underwent pituitary adenoma surgery by a single neurosurgeon at the UCLA Medical Center between May 2008 and April 2016 was performed. All endonasal surgeries were performed endoscopically with a Neurosurgery-Rhinology team approach. 109 patients with tumors 30 mm or larger were included. We correlated visualization of intra-operative diaphragmatic descent with pre-operative MRI findings. We also tested post hoc variables potentially explaining why the diaphragma failed to drop with tumor debulking.
Results: The diaphragma descended during surgery in 64% of cases; this rate decreased to 28% in tumors with > 20 mm suprasellar extension (p < 0.001). Subarachnoid extension, anterior fossa extension, and dumbbell shape also lowered diaphragma drop rate (p < 0.001). Apoplexy and tumor size were not predictive (p > 0.05). Firm tumors, a factor that could not be identified pre-operatively, was also associated with a lower diaphragma drop rate (p < 0.001). Intra-operative CSF leak did not have an effect (p > 0.05).
Conclusions: Pituitary adenomas with suprasellar extension > 20 mm have a significantly lower probability of achieving chiasmal decompression via an endoscopic endonasal approach. Further study will be required to determine if the risk/benefit ratio favors craniotomy for this subset of patients.
Patient Care: This study provides a framework for estimating the likelihood of diaphragma descent and optic chiasm decompression based on pre-operative MRI features for pituitary adnenomas with large suprasellar extension. It can be used to decide whether craniotomy is necessary while selecting the surgical approach. It is also useful for counseling patients pre-operatively.
Learning Objectives: 1) To identify tumor features on the pre-operative MRI that lowers the probability of endoscopically decompressing the optic chiasm
2) To be able to counsel patients on the probability of optic chiasm decompression and the risk/benefit ratio of using endoscopic endonasal approach versus craniotomy