Pattern electroretinograms for the detection of neural loss in patients with permanent temporal visual field defect from chiasmal compression

Purpose To evaluate the ability of full-field and hemifield pattern electroretinogram (PERG) parameters to differentiate between healthy eyes and eyes with band atrophy (BA) of the optic nerve. Methods Twenty-six eyes from 26 consecutive patients with permanent temporal hemianopic visual field defects and BA of the optic nerve from previous chiasmal compression and 26 healthy subjects were studied prospectively. All patients were submitted to an ophthalmic examination including Humphrey 24-2 SITA Standard automated perimetry. Full-field and hemifield (nasal and temporal) stimulation transient pattern electroretinograms (PERG) were recorded using checkerboard screens. Amplitudes and peak times for the P50 and N95 as well as the overall P50+N95 amplitude were measured. The intraocular N95:P50 amplitude ratio was calculated. Comparisons were made using Student’s t-test. Receiver operating characteristic (ROC) curves were used to describe the ability of PERG parameters to discriminate the groups. Results Full-field P50, N95, and P50+N95 amplitude values were significantly smaller in eyes with BA than in control eyes (P < 0.001). Nasal and temporal hemifield PERG studies revealed significant differences in N95 and P50+N95 amplitudes measurements. No significant difference was observed regarding peak times or N95:P50 amplitude ratios. Nasal and temporal hemifield PERG values did not differ significantly in eyes with BA or in controls. Using the 10th percentile of normals as the lower limit of normal, 16 of 26 eyes were considered abnormal according to the best discriminating parameters. Conclusions Transient PERG amplitude measurements were efficient at differentiating eyes with BA and permanent visual field defects from normal controls. Hemifield stimulation PERG parameters were unable to detect asymmetric hemifield neural loss, but further studies are required to clarify this issue. Study registered on ClinicalTrial.gov number: NCT00553761