Screening for amblyopia in preverbal children with photoscreening photographs - III. Improved grading criteria for hyperopia

Objective: To examine the ability of the Medical Technology and Innovations (MTI), Inc., Photoscreener (Cedar Falls, IA) to detect hyperopia and to im prove the photograph grading criteria to screen for amblyopiogenic levels o f hyperopia, Design: Cross-sectional study and reanalysis. Participants and Testing In previous work, 392 participants received a comp lete ophthalmologic examination and were photographed using the MTI Photosc reener. For this study, all 209 participants with normal examination findin gs (65 children) or hyperopia without anisometropia (144 children) were sel ected. The data were reanalyzed using modified photograph grading and ophth almologic examination failure criteria. Potential reasons for why many chil dren with hyperopia passed photoscreening were explored. Main Outcome Measurer We determined whether a study participant would pass or fail screening with a given photograph grading and ophthalmologic examin ation failure criteria. Results: Most children with hyperopia of +2.00 to +3.50 diopters (D) passed screening with the MTI instrument, in most cases because their photographs lacked bright crescents. When bright crescents in at least two of the four possible meridians were the grading guideline for screening failure and th e pediatric ophthalmologists' consensus hyperopia failure criteria (> +3.50 D) were adopted, the sensitivity for hyperopia detection was 100% and the specificity was 88%. Identical results were obtained using the American Aca demy of Ophthalmology Preferred Practice Pattern hyperopia failure criteria (greater than or equal to +4.50 D). Conclusions: The MTI photograph grading guidelines can be simplified, and t he ophthalmologic examination failure criteria for hyperopia can be improve d. The presence of a bright crescent in the lower or the left pupillary mar gin indicate hyperopia in an amblyopiogenic range (> +3.50 D). Ophthalmolog y 2000,.107. 1630-1636 (C) 2000 by the American Academy of Ophthalmology.