Problem: Glaucoma is a leading cause of blindness in the United States. Treatment decisions for glaucoma and other optic nerve diseases are based largely on changes in visual function made by quantification of the visual field. The reliable identification of visual field change is the single most difficult problem in vision testing. With standard perimetry, in areas of moderate visual field damage, retest variability is greater than the physical dynamic range of the instrument and the retinal nerve fiber layer becomes so thin that it is not measurable by current OCT imaging techniques. However, visual function often remains in these damaged areas. We have shown that increasing the stimulus size can lower retest variability while maintaining the same sensitivity to detect new deficits, and extend the useful dynamic range of testing by about 50%. Hypothesis: Using a model of visual receptive field activation by different sized stimuli predicts an improved signal-to-noise ratio using larger perimetric stimuli. We propose that a large portion of perimetric variability and narrow effective dynamic range is due to a poor signal-to-noise ratio associated with using a small stimulus. Using larger perimetric stimuli will enhance the dynamic range of perimetry while lowering retest variability allowing for improved and earlier detection of visual field change. We will test a cohort of 120 glaucoma patients and 60 normals with automated perimetry using stimulus sizes III, V, VI and luminance size threshold perimetry a test that finds a visual threshold by changing stimulus size rather than luminance. Methods:
Specific Aim 1 : Identify the stimulus size with the lowest variability and broadest effective dynamic range. We will define, measure and compare the effective dynamic ranges and retest variabilities of perimetry with stimulus sizes III, V, VI and luminance size threshold perimetry;50 subjects (40 glaucoma patients and 10 healthy participants) will be tested once a week for five weeks.
Specific Aim 2 : Identify the stimulus size that shows visual field change earliest in a patients'course by testing patients longitudinally. We will use linear regression and confidence intervals along with our statistical model (Specific Aim 3) to determine which stimulus size detects visual field change at the earliest time in our 180 subject cohort from our recent Merit Review.
Specific Aim 3 : Develop a statistical model of visual field change. Current techniques incorrectly assume visual field test locations are independent. Using Bayesian and Frequentist analytical tools we will build an exponential spatial correlation structure to model visual field data over time. Summary: We anticipate substantially increasing both the useful dynamic range of perimetry and our ability to detect visual field change. Potential Impact on Veterans Health Care: These methods would impact VA health care by (1) allowing earlier detection of visual field change;(2) minimize misdiagnosis, unnecessary testing and even unnecessary surgery that results from mistakenly interpreting fluctuation of the visual field as change;(3) allow better evaluation of visual function for clinical trials and VA vision rehabilitation programs.
Project Narrative: We will develop methods to substantially lower variability of visual field testing while maintaining the same sensitivity to detect new deficits and to increasing the test's dynamic range. This lower variability will allow better assessment of visual field change over a much broader light sensitivity range. The methods used in the study are practical and would be implemented into clinical practice and clinical trials if proven useful. They would impact VA health care by (1) allowing earlier detection of visual field change;(2) minimize misdiagnosis, unnecessary testing and even unnecessary surgery that results from mistakenly interpreting fluctuation of the visual field as progression or improvement;(3) allow better evaluation of visual function for clinical trials and VA vision rehabilitation programs.
| Wall, Michael (2017) Update on Idiopathic Intracranial Hypertension. Neurol Clin 35:45-57 |
