The Repeatability and Agreement of Two SS-OCT Optical Biometers in Adult Eyes Undergoing Cataract Surgery.

Purpose

To compare the repeatability of and describe the agreement between the IOLMaster 700 (Carl Zeiss Meditec AG) and Eyestar 900 (Haag-Streit) under identical test conditions.

Methods

This was a comparative cross-sectional study conducted at a single-surgeon ophthalmology clinic in Australia. Adults presenting for cataract surgery underwent two scans each on both the IOLMaster and Eyestar in random order. Bland and Altman's methods were used to assess repeatability and agreement.

Results

A total of 160 left eyes satisfied inclusion. For anterior keratometry, the 95% repeatability (test-retest) errors for the Thibos power components were lower with the Eyestar (M = 0.23, J0 = 0.20, J45 = 0.15 diopters [D]) than the IOLMaster (0.28, 0.21, and 0.15 D). For posterior keratometry, the errors were also lower with the Eyestar (M = 0.05, J0 = 0.02, J45 = 0.03 D) than the IOLMaster (0.09, 0.07, 0.07 D). The segmental axial variables had comparable repeatability. White-to-white distance was more repeatable with the IOLMaster (0.14 mm) than the Eyestar (0.26 mm). The Eyestar had notably inferior repeatability for the chord of angle alpha (ICX = 0.51, ICY = 0.57 mm) compared to the IOLMaster (0.14, 0.24 mm). For the chord of angle kappa, the repeatability of the Eyestar (PCX = 0.48, PCY = 0.49 mm) was also inferior to that of the IOLMaster (0.17, 0.15 mm). The Eyestar consistently measured a lower posterior corneal power; the mean difference in the posterior flat curvature (PK1) was -0.40 D (95% CI: -0.42 to 0.40). The biometers had close agreement for the segmental axial variables.

Conclusions

In eyes undergoing cataract surgery, the Eye-star demonstrated superior repeatability to the IOLMaster for keratometry but notably inferior repeatability for most en face variables.