Using sellar region tumor’s size as a predictor of psychophysical and electrophysiological perimetric visual losses: a logistic regression approach

Background Sellar region tumor growth represents an important cause of visual loss due to mechanical compression of the optic nerve apparatus. Many investigations have used non-invasive tools to evaluate the visual field consequences of this damage, and good associations have been reported between psychophysical and electrophysiological perimetries. Few reports have considered the tumor size as a predictor of visual field loss. Aims In the present study, we evaluated the association between the visual perimetry measured by Humphrey visual field analyzer and multifocal visual evoked cortical potential (mfVECP) and the tumor size. Methods Our sample was composed of 14 patients diagnosed with sellar tumors by magnetic resonance imaging. We accounted the number of sectors with negative visual responses for both methods. A simple logistic regression analysis was used to evaluate the association between the tumor dimensions and the visual field features Results Three patients had preserved visual fields, three patients showed hemianopic defects, and eight patients had generalized visual field losses at both evaluations. We observed that the three maximum diameters of the tumor and total tumor volume had different predictive abilities regarding the extent of visual field loss when using psychophysical and mfVECP data. The maximum craniocaudal diameter of the tumor was the better predictor of the psychophysical measurements, whereas for the mfVECP results, all tumor dimensions and volumes had similar values that predict visual field losses. Conclusion Tumor size as a predictor of visual loss has potential to assist in the clinical intervention and to prevent the irreversible visual impairment caused by tumors of the sellar region.