Review of Poster, AAP SOATT section, 2018
Steven Kane, MD, PhD
Edward S. Harkness Eye Institute, Columbia University, New York City
SOATT Section Poster 2018, November 5. 2018
Title: Automated Brightness Sense Screening for Amblyopia.
Purpose: Notwithstanding standard photo and visual acuity screening, amblyopia remains the leading cause of permanent vision loss in children and young adults. Better methods of early detection are needed to improve treatment outcomes. A child-friendly, self-contained device that detects and quantifies unilateral amblyopia rather than identifies risk factors may better enable early detection of amblyopia with both high sensitivity and high specificity. The brightness sense of the two eyes is closely balanced in humans. Brightness disparity has been demonstrated in unilateral amblyopia using brightness rivalry and the amblyopic defect can be quantified by attenuating the light to the better eye. The depth of brightness disparity is highly correlated with visual acuity disparity between amblyopic and normal eyes.
Methods: 294 children from two schools between 3 and 14 years of age appeared for vision testing; 63.3% were below 9 years of age. One school specializes in special education. The self-testing DiagnosticGame® app on iPads presented rivalrous stimuli according to the Hofeldt Bridge® using two vertically aligned stimulus pairs of reciprocal brightness; the right stimulus is brighter in one pair and the left stimulus is brighter in the other pair. The stimulus pairs fuse to form a top and a bottom impression which are equally bright when the visual system is in balance. When not in balance, the brighter stimulus is dimmed to measure the defect. To achieve separation of binocularity, polarizing filters are used. Standard visual acuities were obtained.
Results: Of 294 children, 3 refused testing and 5 (ages 3 to 7 years) could not understand the rules, leaving 286 tested (97.3%). Of the 286 tested, 10 were disqualified due to error; 96.6% completed the test accurately. All 10 children making testing errors and 4 of the 5 children unable to perform the visual acuity test were from the special education school. There were 284 true negatives and 2 true positives, 0 false negatives and 0 false positives. One of the true negatives was a successfully treated amblyope with 20/20 in each eye. Of 2 true positives detected, 1 was identified during testing and the other was a known amblyope with 20/20 OD and 20/25 OS acuities. Exit scores reviews ranging from 1 (boring) to 10 (fun) were obtained on 136 children, the mean score was a pleasant 9.4.
Conclusion: The children were familiar with the iPad and regarded the experience as a game, approaching it with enthusiasm and remembering a pleasant experience. The normally sighted children were identified by a balanced brightness sense. The children with amblyopia were identified by rivalrous brightness disparity with high sensitivity and specificity. Brightness rivalry may be the most sensitive and specific method to detect amblyopia.