Classification of visual field results

The visual field results were independently reviewed and classified by two masked physician reviewers (Xiaoming Chen & Li Tang). The visual field results were distributed in a randomized fashion, in which the central and peripheral visual field examinations were not grouped in any order. In this way, the reviewers could not compare the examinations of one patient. The interpretations were based on predetermined criteria that placed a field into one of three categories: (1) normal, (2) localized defect, (3) diffuse defect (Tables 1 and ​and2)2) [25, 26]. The masked descriptions by each reviewer were compared for each visual field. A consensus description was obtained if the descriptions from the two reviewers had exactly the same pattern. Otherwise, the third reviewer (Liuzhi Zeng) reassessed the visual field pair in an unmasked fashion.

Criteria used for interpretation of central static visual fields

≥10 dB loss at two or more contiguous points

≥5 dB loss at three or more contiguous points

≥10 dB difference across the nasal horizontal midline at two or more adjacent points

Criteria used for interpretation of peripheral kinetic visual fields