Evaluation of refractive error measurements of the WaveScan WaveFront system and the Tracey wavefront aberrometer

Purpose. To evaluate the accuracy and repeatability of the WaveScan WavePrint system and the Tracey wavefront aberrometer in measuring refractive errors in phakic eyes. Setting: Cullen Eye Institute, Baylor College of Medicine, Houston, Texas, USA. Methods. Using subjective manifest refraction (MR) as the standard, the spherical equivalent (SE), sphere, and cylinder were compared to values measured by WaveScan and Tracey devices in virgin eyes and eyes that had had corneal refractive surgery. Astigmatism was evaluated using vector analysis. The accuracy of the WaveScan and Tracey devices was assessed by 95% limits of agreement (95% LA), and repeatability was analyzed by 2 standard deviations (SDs) and intraclass correlation coefficients (ICCs). Results: The mean differences in SE, sphere, and cylinder between MR and WaveScan were -0.26 diopter (D), -0.12 D, and -0.28 D, respectively, and between MR and Tracey, -0.21 D, -0.01 D, and -0.40 D, respectively. The 95% LA for SE, sphere, and cylinder were -1.09 to 0.57 D, -1.14 to 0.89 D, and -0.95 to 0.40 D, respectively, for WaveSGan and -1.37 to 0.95 D, -1.27 to 1.26 D, and -1.16 to 035 D, respectively, for Tracey. Vector analysis revealed mean differences of -0.47 + 0.07 x 9degrees between MR and WaveScan and of -0.53 +0.27 X 12 between MR and Tracey. The 2 SDs for SE, sphere, and cylinder were 0.26 D, 0.29 D, and 0.16 D, respectively, for WaveScan and 0.31 D, 0.36 D, and 0.33 D, respectively, for Tracey. The ICCs for SE, sphere, and cylinder were 0.993, 0.992, and 0.902, respectively, for WaveScan and 0.994, 0.992, and 0.764, respectively, for Tracey. The Tracey device measured all eyes evaluated; the WaveScan could not measure 14% of normal eyes and 50% of post laser in situ keratomileusis eyes. Conclusions: Using MR as the standard, refractive errors measured by the WaveScan and Tracey devices were reliable and reproducible. However, the Tracey device was more robust in its ability to obtain measurements in normal and postoperative eyes. (C) 2003 ASCRS and ESCRS.