Accuracy of matching optic discs with visual fields: the European Structure and Function Assessment Trial (ESAFAT)

PURPOSE: To determine the ability of ophthalmologists across Europe to match stereoscopic optic disc photographs to visual fields of varying severity.

DESIGN: Evaluation and comparison of 2 diagnostic tests.

PARTICIPANTS: A total of 109 of 260 invited ophthalmologists in 11 European countries. These had participated in the previous European Optic Disc Assessment Trial (EODAT), a trial on glaucoma diagnostic accuracy based on optic discs only.

METHODS: Each participant matched stereo optic disc photographs of 40 healthy and 48 glaucomatous eyes to a visual field chosen from 4 options per disc. The 4 presented visual fields included the corresponding one and 3 other visual fields, varying in severity. The matching accuracy and any inaccuracy per disease severity were calculated. Classification accuracy (as glaucomatous or healthy) was compared with EODAT data. Duplicate slides allowed for the assessment of intraobserver agreement.

MAIN OUTCOME MEASURES: Accuracy of matching optic discs with their corresponding visual field and of classifying them as healthy or glaucomatous; intraobserver agreement (κ).

RESULTS: The overall accuracy of ophthalmologists for correctly matching stereoscopic optic disc photographs to their visual fields was 58.7%. When incorrectly matched, the observers generally overestimated the visual field severity (P<0.001), notably in eyes with early glaucoma. The intraobserver agreement was, on average, moderate (0.52).

CONCLUSIONS: European ophthalmologists correctly matched stereoscopic optic disc photographs to their corresponding visual field in only approximately 59% of cases. In most mismatches, the clinicians overestimated the visual field damage.