To investigate the concordance between subjectively and objectively acquired visual fields in patients with subjectively determined hemianopsia.

Retrospective observational study.

Ten patients, six men and four women, ranging in age from 28 to 68 years, were studied. Goldmann or Humphrey perimeters were used to obtain the subjectively determined visual fields for up to 25 degrees of eccentricity, and the VERIS Scientific System (Electro-Diagnostic Imaging, San Francisco, California, USA) was used to record multifocal visual evoked potential [VEPs] (mfVEPs) to obtain the objective visual fields. Each of the 60 black-and-white segments of the checkerboard stimulus was alternated according to a binary m sequence. The first slices of the second-order kernels were extracted and analyzed.

In five cases, the visual field loci where the mfVEPs were within normal limits corresponded to the scotomatous areas obtained by conventional perimetry. In these discordant cases, the lesions (e.g., arteriovenous malformation) were located in the occipital lobe. Two of these cases had a complete recovery of the subjective visual field. The lesions of the concordant cases were located outside the occipital lobe (e.g., pituitary adenoma). In these cases, no visual field improvement was seen. The temporal crescent syndrome was ruled out in patients with posterior lesions by computed tomography (CT) or magnetic resonance imaging (MRI) findings.

In some patients with occipital lesions, the subjective and objective visual field results are discordant, and some of them will show a recovery of the visual field deficits.