Abstract
Purpose :
The Heidelberg Anterion (Heidelberg Engineering, Heidelberg, Germany) is a new swept-source AS-OCT device that provides fast, high resolution anterior segment imaging. However, the intra- and inter-device reproducibility of biometric measurements produced by this device are unknown. This study compares biometric measurements obtained with the Anterion and CASIA SS-1000 (Tomey Corporation, Nagoya, Japan), and examines the benefit of image registration by the Anterion.
Methods :
Subjects 18 years of age or older were recruited. 3 consecutive scans were performed on CASIA & Anterion, or vice versa, in a dark-room environment after 5 minutes of dark adaptation. 1 image per eye was analyzed along the horizontal (temporal-nasal) meridian. Pupillary diameter (PD), anterior chamber width (ACW), lens vault (LV), angle opening distance (AOD), trabecular iris space area (TISA), and scleral spur angle (SSAngle) were measured in each image using manufacturer-provided image analysis software. Inclusion criteria for the study included patients who had PD within 15% of each other for all analyzed scans on both devices. On the Anterion, the Heyex software was closed after the first scan, which disabled its image registration function. ICC values were computed for both devices to assess inter-device and intra-device as well as the effect of image registration by the Heyex software.
Results :
32 AS-OCT images from 32 eyes of 21subjects (10 female, 11 male; mean age 59 ± 19.92 years) were analyzed. PD was similar between the CASIA and Anterion groups (4.75 ± 1.70 mm and 4.84 ± 1.75 mm, respectively). Intra-device reproducibility of measurement values was excellent for both the CASIA (ICC range 0.94 for TISA to 0.97 for LV) and Anterion (ICC range from 0.96 for ACW, LV, TISA to 0.97 for AOD, SSA). Inter-device reproducibility of measurement values was also excellent (ICC range 0.85 for TISA to 0.96 for LV). Image registration did not a have significant effect on the reproducibility of biometric measurements: with registrations (ICC range 0.86 for TISA to 0.99 for LV), without registration (ICC range 0.89 for TISA to 0.99 for LV).
Conclusions :
Our results suggest that biometric measurements by swept-source AS-OCT devices are highly reproducible and may be interchangeable between devices. Also, image registration, at least in the short term, has minimal benefit on measurement reproducibility.
This is a 2020 ARVO Annual Meeting abstract.