Repeatability of a new swept-source optical coherence tomographer and agreement with other three optical biometers.

PURPOSE: To investigate the repeatability of Anterion and compare the agreement of ocular biometric measurements and predicted intraocular lens (IOL) powers with other three optical biometers. METHODS: Flat keratometry (Kf), steep keratometry (Ks), J0 and 45 vectors, central cornea thickness (CCT), anterior chamber depth (ACD), and axial length (AL) from the Anterion, IOLMaster 700, Lenstar LS 900, and OA-2000 were recorded. The IOL powers were calculated with the Hoffer Q, Holladay 1, SRK/T, and Haigis formulas. The repeatability was evaluated using the within-subject standard deviation (Sw), repeatability coefficient (RC), coefficient of variation (COV), and intraclass correlation coefficient (ICC). Inter-device agreement between the four biometers was assessed with the 95% limits of agreement. RESULTS: In total, 101 right eyes of 101 participants were enrolled. The Anterion showed good repeatability for all the included biometric parameters with all the CoV ≤ 0.30% and ICC ≥ 0.930 except for J45 with moderate repeatability (ICC was 0.849). Good agreement was found among the four devices for Kf, Ks, J0, J45, ACD, and AL. Generally, wide 95% LoA was found for the predicted IOL powers with the four IOL calculation formulas between the four devices. CONCLUSIONS: The Anterion showed good repeatability of biometric measurements for most parameters. Good agreement among the four optical biometers was achieved for all the parameters except for CCT and the predicted IOL power. The AL values exhibited the best repeatability with Anterion and the best agreement among the biometers in our study.

Comparison of Refractive Prediction Accuracy With Three Optical Devices.

PURPOSE: To investigate refractive prediction accuracy with the OA-2000 (Tomey), Anterion (Heidelberg Engineering), and IOLMaster 500 (Carl Zeiss Meditec AG) in patients with cataract. METHODS: Patients with cataract referred for phacoemulsification were enrolled and scanned with the OA-2000, Anterion, and IOLMaster 500 in random order. The success rate of axial length (AL) measurements per device was calculated and a chi-square test was used to identify the differences in acquisition rate between the three devices. The Bland-Altman method was used to appraise the agreement of biometric parameters between the three devices. Four different formulas (Barrett Universal II [BUII], Haigis, Holladay 1, and SRK/T) were included in the study. The parameters of refractive prediction accuracy comprised predictive error (PE), absolute PE (AE), and percentages of eyes with a PE within ±0.50, ±0.75, and ±1.00 diopters (D). RESULTS: The acquisition rates of AL measurements with the OA-2000 and Anterion were 97.35% and 94.70%, respectively (chi-square = 3.82, P > .05). A significantly lower acquisition rate of 84.82% was obtained with the IOLMaster 500 compared with the other two devices (P < .05). Bland-Altman analysis identified good agreement between the three biometers with narrow 95% limits of agreement for flat and steep keratometry (K1 and K2), anterior chamber depth (ACD), and AL. For PE, only the differences between the Anterion and IOLMaster 500 with the Barrett UII and Haigis formulas were statistically significant (P < .05). The three devices revealed no statistically significant differences in MAE, MedAE, and the proportion of eyes with a PE within ±0.50, ±0.75, and ±1.00 D (P > .05). CONCLUSIONS: The OA-2000 and Anterion showed a similarly higher acquisition rate of AL measurements than the IOLMaster 500 in patients with cataract. Good agreement for K1, K2, ACD, and AL was found between the three biometers. Regarding refractive prediction accuracy, the Anterion did not significantly outperform both the OA-2000 and IOLMaster 500. [J Refract Surg. 2023;39(1):48-55.].