Repeatability and Cross-Domain Agreement of Pyramidal-Sensor Ocular Wavefront Measurements in Myopic Eyes: A Registry Validation of OSIRIS

Purpose: The purpose of this study was to quantify within-session repeatability of ocular wavefront metrics from a pyramidal wavefront-sensor aberrometer (OSIRIS; CSO), to define a pupil-support and zero-code gate for 6-mm analysis, and to assess agreement between OSIRIS whole-eye and MS-39 total-corneal wavefront. Methods: Retrospective single-center registry analysis of myopic and myopic astigmatic eyes. Within-subject standard deviation (SD), repeatability coefficient (RC = 2.77 × SD), and intraclass correlation coefficient ICC (1.1) were computed across within-session scan clusters with cluster-bootstrap confidence intervals (CIs). Agreement used Bland-Altman limits and Lin concordance correlation coefficient (CCC); and astigmatism used Thibos power vectors. Results: Across 10,416 sessions from 2533 eyes, single-scan repeatability was excellent: 4-mm higher-order aberration (HOA) root mean square (RMS), ICC = 0.987 (RC = 0.033 µm); pupil-supported 6-mm HOA RMS ICC = 0.992 (RC = 0.050 µm); and 6-mm spherical aberration ICC = 0.997. Preoperative eyes were the noisier stratum (6-mm HOA RC = 0.065 vs. 0.039 postoperatively). Diameters below 6 mm gave entirely zero-coded 6-mm exports in 96.8% of sessions. Omitting scan-selection worsened 6-mm HOA RC to 0.257 µm. Whole-eye and total-corneal values correlated but were not interchangeable (6-mm HOA RMS CCC = 0.623). Objective and manifest refraction were consistent (4-mm sphere CCC = 0.965), a rounding-limited consistency check. Conclusions: In myopic eyes, OSIRIS shows excellent within-session repeatability under the described acquisition and quality-control (QC) workflow. Six-millimeter analysis requires pupil-diameter and zero-code gating. Whole-eye and total-corneal wavefront are not interchangeable. Translational Relevance: Within-session precision measured in routine high-volume practice, with a pupil-diameter and zero-code quality rule, defines the smallest same-session change distinguishable from measurement variability, informing wavefront-guided planning, lens selection, and post-refractive follow-up.