Comparison of the accuracy and reproducibility of the keratometer and the EyeSys Corneal Analysis System Model I.

The EyeSys Corneal Analysis System Model I measures corneal topography using digital image analysis of placido rings reflected off the cornea. With three observers, we compared the accuracy and reproducibility (precision) of the Marco Keratometer Model 1 and EyeSys Corneal Analysis System Model I using four poly(methyl methacrylate) spheres (37.50, 42.51, 47.54, and 55.06 diopters [D]), three steel spheres (40.50, 42.50, and 44.75 D), and 20 normal human eyes (41.50 to 46.00 D). For the spheres, the standard deviations of intra-observer and overall reproducibility for both devices were less than 0.12 D; the absolute mean differences between the measurements of the seven spheres and the known values were 0.25 D or more for two spheres as measured by the keratometer and none as measured by the EyeSys. For the normal corneas, the standard deviations of intra-observer and overall reproducibility for dioptric measurements were 0.07 D and 0.14 D for the keratometer and 0.13 D and 0.19 D for the EyeSys. The EyeSys Corneal Analysis System Model I exceeds keratometer accuracy in reading calibrated spheres and approaches keratometer reproducibility in measuring the 3-mm zone of normal human corneas.