Résumé
Purpose: To compare the accuracy of noncycloplegic photorefraction [NCP] with that of cycloplegic refraction [CR] for detecting refractive amblyopia risk factors [RARFs] and to determine cutoff points
Methods: In this diagnostic test study, right eyes of 185 children [aged 1 to 14 years] first underwent NCP using the Plusopti Chi SO4 photoscreener followed by CR. Based on CR results, hyperopia [>/=+3.5 D], myopia [>/=-3 D], astigmatism [>/=1.5 D], and anisometropia [>/=1.5 D] were set as diagnostic criteria based on AAPOS guidelines. The difference in the detection of RARFs by the two methods was the main outcome measure
Results: RARFs were present in 57 [30.8%] and 52 [28.1%] of cases by CR and NCP, respectively, with an 89.7% agreement. In contrast to myopia and astigmatism, mean spherical power in hyperopic eyes was significantly different based on the two methods [P<0.001], being higher with CR [+5.96 +/- 2.13 D] as compared to NCP [+2.37 +/- 1.36 D]. Considering CR as the gold standard, specificities for NCP exceeded 93% and sensitivities were also acceptable [>/=83%] for myopia and astigmatism. Nevertheless, sensitivity of NCP for detecting hyperopia was only 45.4%. Using a cutoff point of+1.87 D, instead of+3.5 D, for hyperopia, sensitivity of NCP was increased to 81.8% with specificity of 84%
Conclusion: NCP is a relatively accurate method for detecting RARFs in myopia and astigmatism. Using an alternative cutoff point in this study, NCP may be considered an acceptable device for detecting hyperopia as well