Computerized videokeratography and keratometry in determining intraocular lens calculations

Abstract

Purpose: To compare the accuracy of standard keratometry and computerized videokeratography (CVK) in determining intraocular lens (IOL) power calculations. Methods: Using the EyeSys Corneal Analysis System^TM, we prospectively obtained CVK maps on 75 eyes of 69 patients scheduled to have phacoemulsification with implantation of a posterior chamber intraocular lens. Using manifest refraction obtained at 6 weeks postoperatively, we optimized the calculations for the Hoffer Q, Holladay, and SRK/T formulas for standard keratometric and the following six CVK values: average curvatures at the 1 mm, 2 mm, and 3 mm zones, the keratometric equivalent at the 3 mm zone, and the Stiles-Crawford weighted averages over the 3 mm and 6 mm zones. The accuracy of these parameters was determined by calculating the mean absolute error and percentage of patients with accuracy within ≤ 0.5 diopter (D), ≤ 1.0 D, and ≤ 2.0 D. Results: Keratometrically derived data were slightly more accurate than the CVK-derived values. The average difference in mean absolute error between the keratometric and CVK values was 0.13 D for the Hoffer Q formula, 0.11 D for the Holladay, and 0.08 D for the SRK/T. Conclusions: In this population of patients, we found the CVK-derived corneal curvature values to be slightly less accurate than standard keratometry in predicting IOL power. However, CVK provides important corneal curvature data for IOL calculations in patients with abnormal or surgically altered corneal surfaces.