Clinical evaluation of a new fully automated biometer integrating optical low-coherence reflectometry and Placido-disk topography
摘要
To investigate the precision (intraobserver repeatability and interobserver reproducibility) of the AL550, an automated biometer that integrates optical low-coherence reflectometry with Placido-disk topography, and to assess the inter-device agreement with the IOLMaster 700.
MethodsThis prospective study enrolled 112 healthy participants. Each participant underwent three consecutive right-eye measurements by two operators with different levels of experience using the AL550 to assess intraobserver and interobserver variability. Additionally, three measurements were obtained by a skilled operator using the IOLMaster 700 for inter-device comparison. The parameters assessed included axial length (AL), central corneal thickness (CCT), anterior chamber depth (ACD), lens thickness (LT), flattest, steepest, and mean keratometry (Kf, Ks, and Km), astigmatism, and corneal diameter (CD). Measurement precision was analyzed using the within-subject standard deviation (Sw), test–retest repeatability (TRT), coefficient of variation (CoV), and intraclass correlation coefficient (ICC). Inter-device agreement was assessed using paired t-tests and Bland–Altman analysis.
ResultsThe AL550 demonstrated excellent repeatability and reproducibility across all parameters, with low Sw, TRT, and CoV values, and ICCs exceeding 0.939. Inter-device agreement with the IOLMaster 700 was excellent for most parameters, with narrow 95% limits of agreement (LoA) (AL: − 0.09 to 0.08 mm, CCT: − 16.55 to 5.46 μm, ACD: − 0.04 to 0.18 mm, LT: − 0.27 to 0.07 mm, Km: − 0.26 to 0.26 D). However, CD exhibited a wider 95% LoA (− 0.82 to − 0.15 mm).
ConclusionThe AL550 provides highly repeatable and reproducible biometric measurements with minimal operator dependence. Inter-device agreement with the IOLMaster 700 was excellent for all parameters except CD, suggesting that the two devices may be used interchangeably in clinical practice.