Investigation of predictors of latent visual impairment in patients with sellar lesions.

Objective: Patients with sellar lesions compressing the optic nerve sometimes perceive visual improvement after lesion resection, despite the absence of visual impairment on preoperative ophthalmological examination. This study investigated the indicators of latent visual impairment in patients with sellar lesions.

Methods: Forty-five patients who underwent surgery for sellar lesions compressing the optic nerve with no preoperative visual abnormalities and no change in visual assessment between pre- and postoperative ophthalmological examinations were divided into two groups: 1) patients who perceived recovery of visual function after lesion resection (the improved group), and 2) patients who did not (the unaffected group). Four assessments of optic nerve bending were compared between the groups: 1) coronal optic nerve bending height (CONBH), defined as the vertical distance between the highest apex of the bending optic chiasm and the line connecting the endpoints of the optic nerve on the coronal image; 2) coronal optic nerve bending angle (CONBA), defined as the internal angle formed by the intersection of lines parallel to the left and right optic nerves at the greatest bending section of the optic chiasm on the coronal image; 3) sagittal optic nerve bending height (SONBH), defined as the vertical distance between the highest apex of the bending optic chiasm and the extension line of the optic nerve course in the optic canal on the sagittal image; and 4) sagittal optic nerve bending angle (SONBA), defined as the internal angle formed by the intersection of the line connecting the optic canal entrance level and the highest apex of the bending optic chiasm and the extension line of the optic nerve course in the optic canal on the sagittal image.

Results: Of the 45 patients, 21 were assigned to the improved group and 24 to the unaffected group. CONBH, CONBA, SONBH, and SONBA showed significant differences between the groups (all p < 0.001). Multivariate logistic regression analysis revealed that only SONBA was a significant independent predictor of perceived visual recovery after resection of sellar lesions (OR 2.29, 95% CI 1.03-5.10; p = 0.042). The optimal cutoff point of SONBA for perceiving visual recovery was identified as 30° (specificity 1.000, sensitivity 0.952).

Conclusions: Angle assessment of optic nerve bending due to compression caused by sellar lesions on the sagittal image may be useful in identifying sellar lesions with latent visual impairment.