Precision, Agreement, And Accuracy Of Argos® And Eyestar 900® Swept-Source Optical Coherence Tomography Biometers In Eyes Undergoing Cataract Surgery.

Biometers based on swept-source optical coherence tomography (SS-OCT) technology are being increasingly used to determine axial length (AL) in patients undergoing cataract surgery due to its precision, fast acquisition rate and non-contact nature. New devices have been recently available to the clinician, requiring assessment of their performance. We compared the measurements obtained with the Argos^® (Alcon Healthcare, US) and the Eyestar 900^® (Haag-Streit, Switzerland) SS-OCT biometers in eyes with cataracts undergoing phacoemulsification.

OMIQ (Oftalmologia Mèdica i Quirúrgica) Research; Barcelona, Spain. The study was conducted from October to December 2023.

Prospective study comparing the precision, agreement, and accuracy of Argos^® and Eyestar 900^® SS-OCT biometers. One eye of patients of either sex aged ≥50 years undergoing cataract surgery had 3 consecutive AL measurements with each biometer in random order. Refraction was performed one month after surgery. The primary endpoint, median AL, was compared between biometers using the Mann-Whitney test. For secondary endpoints, the between-device agreement was established using Bland-Altman plots, the within-subject standard deviation (S_w) was measured, and the % of failed AL measurements and eyes within +/-0.50 and +/-1.00 determined after intraocular lens power calculation using the AL derived from Argos^® using Barrett Universal II formula.

We included 107 eyes of 107 patients, 60.8% of whom were female with a mean age of 73.1 years (standard deviation 7.8). The median difference in AL measurements (Argos^® - Eyestar 900^®) was -0.01 mm (interquartile range [IQR] 0.06), p=0.01. Bland-Altman plots showed 95% limits of agreement of -0.11 to +0.08 mm and a trend towards less extreme AL measurements for Argos^®. The median (IQR) S_w were 0.0058 (0.0058) and 0.0000 (0.0058) for Argos^® and Eyestar 900^®, respectively (p<0.0001). There were no failed AL measurements (0%; 95% confidence interval 0% to 3.4%), with 96.1% of eyes within +/-0.50D and 100% within +/-1.00D.

Argos^® and Eyestar 900^® SS-OCT biometers provided statistically significant but clinically irrelevant differences in AL, with very small differences between repeated measurements. Nonetheless the devices are not interchangeable, particularly in very long or very short eyes due to the segmented index approach used in the Argos^® device.