Area-based versus ring-based analysis for detecting subtle corneal curvature changes in a chick model of experimental...

This study compares Area-Based Analysis (ABA) and Ring-Based Analysis (RBA) for detecting corneal changes in a chick model of experimental astigmatism. Both methods were first validated using steel balls with known radii, demonstrating comparable keratometric values (p ≥ 0.14) and good accuracy, with mean deviations of approximately 0.50 D or less across SteepK, FlatK, MeanK, cylindrical power (CYL), and J0 and J45 astigmatic components. The methods were then applied to chicks randomly assigned to monocularly imposed hyperopic with-the-rule (H-WTR, n = 12) or against-the-rule (H-ATR, n = 10) astigmatism groups, with six untreated Control. Compared to RBA, ABA consistently showed significantly higher within-eye repeatability for all parameters (p < 0.001). In treated eyes, ABA detected the CYL differences in both H-ATR (p = 0.017) and H-WTR (p < 0.001) groups versus Control, whereas RBA only identified differences in H-WTR (p = 0.03) versus Control. Furthermore, ABA yielded significantly steeper SteepK, FlatK, and MeanK (p < 0.001), likely due to differences between methods in sampling the radial curvature gradient of the aspheric cornea, whereas the more negative J0 and J45 values (p ≤ 0.001) reflected method-dependent shifts in astigmatic axis. When comparing interocular differences (treated right eye minus untreated left eye), ABA uniquely identified treatment-induced CYL changes in H-WTR compared to H-ATR and Control (p ≤ 0.034). These results underscore that ABA showed higher repeatability and statistical sensitivity for detecting subtle corneal astigmatic changes, establishing it as an effective analytical method for experimental astigmatism research in chicks.