Treatment-induced shifts of ocular reference axes used for measurement centration.

PURPOSE To determine the shifts of the main corneal reference points in dependence of the chosen centration axis for the treatment. SETTING Federal Institute of Technology Zurich, Institute of Biomedical Engineering, Zurich, Switzerland. METHODS Computer simulations were performed on several variants of the Gullstrand-Emsley schematic eye, which was modified by an off-axis fovea. Refractive corrections were simulated by centering Munnerlyn's formula on each of the 4 corneal reference points determined in the preoperative eye: the optical axis, the line of sight, the visual axis, and the first corneal reflex. Subsequently, the postoperative locations of these axes were determined and compared with the preoperative values. RESULTS The postoperative line of sight was found to depend least on the choice of the preoperative centration axis for both myopic and hyperopic treatments. It undergoes a maximum movement of 0.040 mm when centering a +5 diopter correction on the preoperative line of sight, whereas the corneal reflex, which is used for centering most topography systems, can move by more than 0.10 mm. CONCLUSIONS Centration of the correction on the preoperative line of sight enabled good comparability between preoperative and postoperative measurements that use the line of sight as a reference axis. Yet, centration of the treatment on the preoperative line of sight does not ensure comparability between preoperative and postoperative measurements that use the corneal reflex as a reference axis such as most corneal topography systems. Axis shifts might lead to misinterpretation of data such as a wrong diagnosis of a decentered ablation or changes in the Zernike representation.