Refractive Outcomes and Ocular Biometry of Two Swept-Source Optical Coherence Tomography-Based Biometers With Segmental or Equivalent Refractive Indices (ORION)

Overview

This prospective, single-centre, randomized, parallel-group comparative clinical study evaluated the refractive prediction accuracy of two swept-source optical coherence tomography (SS-OCT) biometers in a Thai cataract surgery cohort. The ARGOS biometer applies segmental refractive indices to individual ocular segments, whereas the IOLMaster 700 applies an equivalent refractive index to the whole eye. Each device is conventionally paired with a different keratometry-formula combination: ARGOS with the standard Barrett Universal II (Barrett UII K), and IOLMaster 700 with the Barrett Universal II using total keratometry (Barrett UII TK). The study compared these two device-formula pairings as they are used in routine clinical practice. Both biometers were performed on every eye on the same day, enabling a within-subject paired comparison of biometric agreement and refractive prediction error at one month after phacoemulsification cataract surgery with implantation of a TECNIS Monofocal one-piece IOL (model GCB00V)

One hundred and forty-four eyes of 144 consecutive Thai cataract surgery patients aged 40-90 years presenting with visually significant age-related cataract were enrolled at the Ophthalmology Service, Walailak University Hospital between May 2024 and December 2024. Participants underwent preoperative biometric measurement on the same day with both the ARGOS (software version A2.4; Movu, Inc., a Santec company; distributed by Alcon Laboratories) and the IOLMaster 700 (software version 1.90.10; Carl Zeiss Meditec AG) by a single experienced operator. Participants were randomized 1:1 between two IOL-selection arms using a computer-generated allocation sequence (Research Randomizer, blocks of 12): Group 1 used the IOLMaster 700 with the Barrett Universal II TK formula for IOL power selection, and Group 2 used the ARGOS with the standard Barrett Universal II K formula. Allocation was concealed from the enrolling surgeon until completion of preoperative biometry. Participants were masked to the device-formula arm. All cataract surgeries were performed by four experienced surgeons (in approximately equal numbers) using a standardized phacoemulsification technique with in-the-bag implantation of the TECNIS Monofocal one-piece IOL (model GCB00V; Johnson \& Johnson Vision). The primary outcome was mean absolute prediction error (MAE) at one month postoperatively, assessed by a masked refractionist using subjective manifest refraction. Secondary outcomes were median absolute prediction error and the proportion of eyes within ±0.25 D and ±0.50 D of target refraction. One eye per patient was enrolled. The study was registered prospectively as a non-inferiority trial; the implemented analysis is reported as a within-subject paired comparative analysis with descriptive accuracy thresholds rather than as a formal non-inferiority hypothesis test (CONSORT item 3b). The total enrolled sample was 144 patients; within-subject paired analyses included 138 eyes with complete data for both devices.