To develop a method to predict the refractive power of the cornea from corneal topography.

We reviewed preoperative and post-operative cycloplegic refraction, keratometry, and corneal topography in 40 eyes of 40 patients who had undergone photorefractive keratectomy, radial keratotomy, myopic keratomileusis in situ, or hyperopic lamellar keratoplasty. For each axial dioptric power map, we calculated the aspheric ellipsoid that best fit that map. Central corneal points were weighted more heavily than peripheral points, based on the Stiles-Crawford effect. The equation of the best-fit ellipsoid yielded the spherical and astigmatic power and axis for each cornea preoperatively and postoperatively.

The preoperative corneal spherical and astigmatic powers measured by the best-fit method were consistent with the spherical and astigmatic powers measured by keratometry and simulated keratometry. The change in corneal spherical power predicted by the best-fit method was significantly (P < .05) more accurate at predicting the change in spherical equivalent refraction than change either in spherical equivalent keratometry or in spherical equivalent simulated keratometry. The prediction of the astigmatic change was less precise than that of the spherical, but the best-fit method was the most accurate.

The best-fit method is more accurate than simulated keratometry and standard keratometry are in evaluating corneal refractive power after refractive surgery. An improved method of calculating corneal refractive power may facilitate subjective refraction after refractive surgery, improve the accuracy of intraocular lens power calculation for eyes that have had previous refractive surgery, and improve ablation profiles for excimer laser refractive surgery.