Measuring Visual Function Using Involuntary Eye Movements

Abstract

Background: Reliable measurement of visual function in humans is essential for the early diagnosis, treatment and monitoring of eye disease. Methods for clinical assessment of visual function typically require the patient to report what they see, for example, by reading a letter on a chart. Such subjective procedures can be challenging for some patients, which can, in turn, limit the reliability of their assessment. This can especially be the case for children, older people, or people with cognitive impairment. Aims: To develop an automated system for assessing visual function using a type of involuntary eye movement known as optokinetic nystagmus (OKN). Methods: In four studies, we recorded eye movements during OKN stimulation using infrared oculography. First - to determine if OKN, could be the basis of an objective measure of the clarity of functional vision Studies 1 and 2 investigated how OKN changes in response to a change in mean spherical equivalent refractive error (MSE). The recorded OKN signals were quantified either using OKN consistency with the direction of stimulus motion (OKN score), or the ratio between the velocity of eye movements during the slow phase of OKN to the stimulus velocity (OKN gain). In Study 1, we quantified the OKN scores of 20 participants with different MSEs. In Study 2, we quantified the relationship between OKN gain and MSE in 25 participants at five induced powers for each participant. To determine if OKN could provide an objective measure of visual field loss, in Study 3 we quantified the OKN gain of 16 healthy participants viewing stimuli of varying contrasts with different levels of simulated visual field loss (SVFL). Next, in Study 4, we determined if OKN could discriminate asymmetrical visual field loss in 41 participants with glaucoma. Results: Study 1 indicated that estimates of MSE based on OKN scores correlate well with clinical estimates of MSE, r = .878, p < .001, mean difference: 0.00 D, 95% limits of agreement: [-0.85 D to +0.85 D]. Study 2 revealed a correlation across participants of r = .976, p < .001 between clinical measures of MSE and objective MSE estimated based on OKN gain; mean difference: +0.05 D [-0.90 D to +1.01 D]. Study 3 showed that full-field stimulation led to an average OKN gain of 0.92 ± 0.15. This value fell steadily with increasing SVFL to a value of 0.38 ± 0.20 when the periphery was not stimulated at all (i.e. the stimulus was a 5-deg-diameter foveal patch). Study 4 revealed OKN gain was lower in the worse eye—compared to the better eye—of patients with glaucomatous field loss, Friedman’s χ2(1, N = 164) = 37.098, p < .001. The reduction of OKN gain in the worse eye was more prominent at low contrast levels—for example, at 6.25% and 12.5% contrasts compared to other tested contrasts. Monte-Carlo simulation revealed a sensitivity of 90% with a specificity of 83% for the detection of a mean deviation field loss of -10 dB. Conclusion: Studies 1 and 2 indicate that (a) OKN can be used as an objective proxy for patient response to select the best corrective lens and (b) increasing MSE attenuates OKN gain. Studies 3 and 4 indicate that both simulated and glaucomatous field loss weakens the optokinetic response (predominantly to low contrast stimuli). As a means of objectively determining refractive error or quantifying field loss, OKN-tests have the potential to overcome the shortcomings of tasking a patient with making reliable subjective judgements.