Eye movement perimetry in glaucoma.

Present-day computerized perimetry is often inaccurate and unreliable owing to the need to maintain central fixation over long periods while repressing the normal response to presentation of peripheral stimuli. We tested a new method of perimetry that does not require prolonged central fixation. During this test eye movements were encouraged on presentation of a peripheral target. Twenty-three eyes were studied with an Octopus perimeter, with a technician monitoring eye movements. The sensitivity was 100% and the specificity 23%. The low specificity was due to the technician's inability to accurately monitor small eye movements in the central 6 degrees field. If small eye movements are monitored accurately with an eye tracker, eye movement perimetry could become an alternative method to standard perimetry.