Videokeratography, keratometry, and refraction after penetrating keratoplasty

PURPOSE: To identify the correlation between videokeratography, autorefract ometry, autokeratometry, and keratometry measurements and the subjective ma nifest refraction and spectacle-corrected visual acuity after penetrating k eratoplasty. METHODS: We studied 100 eyes from 100 patients that had undergone penetrati ng keratoplasty. All eyes were examined by videokeratography (EyeSys 2000) (axial, tangential, and refractive power maps), autorefractometry, autokera tometry, and keratometry. Measurements were made at an an average of 19 +/- 9 months after surgery. Postoperative refractive astigmatism, spherical eq uivalent refraction, and spectacle-corrected visual acuity were studied by regression analysis. RESULTS: Both the total topographic cylinder measured by the refractive pow er map and topographic cylinder measured by the axial power map showed the strongest correlation with the manifest refractive cylinder (r(s) = +0.89, P <.001). The axis of astigmatism determined by keratometry and autokeratom etry showed the strongest correlation with the subjective manifest refracti on axis (r(s) = +0.87, P <.001). The total topographic cylinder showed the strongest correlation with the spectacle-corrected visual acuity (r(s) = +0 .38, P =.001); however the topographic indices of predicted corneal acuity, corneal acuity, corneal uniformity index, asphericity, and refractive powe r symmetry did not correlate with spectacle-corrected visual acuity. CONCLUSION: Measurement of astigmatism after penetrating keratoplasty can b e made more accurate by using videokeratographic measurements to supplement retinoscopic and manifest refraction. Other useful methods for predicting the axis of refractive astigmatism include keratometry, autokeratometry, an d autorefractometry. In this study, indices designed to measure corneal sur face irregularity failed to predict visual acuity after penetrating keratop lasty.