Use of visual acuity to screen for significant refractive errors in preschool children
TL;DR
VA was valid for detecting myopia at the <6/7.5 cutoff for preschool children and Astigmatism and hyperopia, even at high levels, did not consistently impact VA.
Abstract
ABSTRACT Purpose: To determine the sensitivity and specificity of visual acuity (VA) for the detection of refractive errors in preschool children from the Sydney Pediatric Eye Disease Study. Methods: Data from both eyes were pooled for a total of 2332 observations. Children aged 3–6 years had a comprehensive eye examination including VA (monocularly tested using the Electronic Visual Acuity HOTV test) and cycloplegic (cyclopentolate 1%) refraction using the table-mounted autorefractor (RK-F1 Auto Ref-Keratometer; Canon, Tokyo, Japan). Clinically significant refractive error was defined as: Myopia ≤–1.00D, Hyperopia ≥+3.00D (Moderate Hyperopia ≥+3.00D – <+5.00D, High Hyperopia ≥+5.00D) and Astigmatism ≥+1.00D. VA cutoffs used were <6/12, <6/9 and <6/7.5. Ethics approval was obtained by the Human Research Committee, University of Sydney and adheres to the tenants of the Declaration of Helsinki. Results: There was low sensitivity for detection of myopia at a cutoff of <6/12 (41.2%), with moderate sensitivity at <6/9 (58.8%), and high sensitivity (82.4%) and specificity (83.9%) at <6/7.5. Across the three cutoffs, there was low sensitivity for detection of astigmatism (<6/12: 14.7%, <6/9: 22.5%, <6/7.5: 54.3%) and hyperopia (<6/12: 13.8%, <6/9: 21.1%, <6/7.5: 52.3%). This was particularly true for those with moderate levels of hyperopia (sensitivity <6/12: 5.0%, <6/9: 8.8%, <6/7.5: 45.0%). High hyperopia was better detected, particularly at the <6/7.5 cutoff (72.4%); however, had low sensitivity at <6/12 (37.9%) and <6/9 (55.2%). ROC analysis for detection of clinically significant refractive error revealed an area under the curve (AUC) of 0.708 (p < .001) for <6/7.5. This was comparatively lower at <6/9 (AUC = .606) and <6/12 (AUC = .572, all p < .001). Conclusion: VA was valid for detecting myopia at the <6/7.5 cutoff for preschool children. Astigmatism and hyperopia, even at high levels, did not consistently impact VA. This remains a challenge for detecting significant refractive errors when screening in preschool children.