Correlation between corneal topography and subjective refraction in idiopathic and surgery-induced astigmatism

Purpose: To study the correlation between subjective refraction and corneal topography. To compare the topographic analysis of surgically induced asti gmatism (cataract and penetrating keratoplasty) with that of idiopathic ast igmatism. Methods: Subjective astigmatism, subjective spherical equivalent, best spec tacle-corrected visual acuity (LogMAR units), and videokeratoscopy using th e EyeSys 2000(R) device (axial, tangential, and refractive power) were reco rded in 100 eyes with idiopathic astigmatism, 100 eyes after cataract surge ry, and 100 eyes after penetrating keratoplasty. Topographies were classifi ed according to pattern (Bogan classification) and asphericity (shape facto r. prolate or oblate). Results: The asphericity shape distribution was significantly different bet ween the 3 groups (p < 0.001). The shape of idiopathic astigmatism was almo st always prolate (90%) whereas the oblate shape was more frequent in the p enetrating keratoplasty group (75%). There was no significant difference in topographic pattern distribution between the 3 groups (p = 0.11). The asym metric bow tie pattern was the most common topographic pattern. Topography pattern classification was significantly correlated with the subjective ast igmatic cylinder. (r(s) = 0.60, p < 0.001). Unlike the round and oval patte rns, the bow tie pattern was associated with the high subjective cylinder. Correlation between the subjective cylinder, the refractive power cylinder, and the axial power cylinder was strong (r(s) = 0.92 p < 0.001), but it wa s weak for the tangential power cylinder (r(s) = 0.72 p < 0.001). The corre lation between the subjective spherical equivalent and central cornea power was poor (r(s) < 0.37, p < 0.001). Subjective astigmatic cylinder showed t he strongest correlation with best spectacle-corrected visual acuity (r(s) = 0.70, p < 0.001), whereas the predicted corneal acuity, corneal uniformit y index, asphericity, and refractive power symmetry index were poorly corre lated with it (r(s) < 0.54, p < 0.001). Conclusion: Despite the difference in the asphericity shape, the topographi c pattern was similar in the 3 groups. The pattern type was significantly c orrelated with the subjective astigmatic cylinder. Topographic indices fail ed to predict visual acuity, while the subjective cylinder showed a strong correlation with visual acuity.