Predictors of Visual Outcomes in Pituitary Neuroendocrine Tumor Resection

ABSTRACT Objective: We sought to identify clinical and radiologic predictors of visual recovery and explore visual outcomes in patients who underwent transsphenoidal resection of pituitary neuroendocrine tumors (PitNETs). Methods: Retrospective study of patients with PitNETs approaching/contacting the chiasm at a tertiary neuro-ophthalmology practice. Clinical/demographic/MRI data were analyzed. Patients with a final visual acuity (VA) of at least 20/40 and a final visual field mean deviation (VF MD) of −5 dB or better were defined as positive outcomes. Results: 68 patients were analyzed, including 44 patients who underwent surgery and 24 who were observed during the study period. Of those who underwent surgery, patients with positive VA outcomes had better preoperative VA (logMAR 0.3 vs 1.3), VF MD (−8.1 vs −17.9 dB), thicker retinal nerve fiber layer (RNFL) (81.9 vs 69.5 µm) and ganglion cell complexes (GCC) (69.0 vs 50.5 µm) than those without. VF outcomes were predicted by better preoperative VA (logMAR 0.3 vs 0.7), VF MD (−7.4 vs −14.5 dB), thicker RNFL (84.1 vs 70.5 µm) and GCC (71.2 vs 55.6 µm) measurements and smaller tumor volume (8.4 vs 18.2 cm 3 ). Conclusion: Preoperative VA/VF MD/RNFL/GCC thickness may predict visual outcomes after PitNET resection. Neuroradiologic factors failed to predict visual outcomes.