Abstract
Purpose:
Accurate biometric measurements are vital to the process of intraocular lens (IOL) power selection for cataract surgery and the optimization of refractive outcomes. The Haag-Streit LenStar LS 900 makes corneal measurements based on 32 data points as compared to 6 used by the Carl Zeiss IOL-Master V5. This study aimed to discern whether the LenStar consequently provides more accurate measurements and predictions of post-operative refractive outcomes.
Methods:
This retrospective study used data on 28 eyes from 28 patients who underwent phacoemulsification cataract surgery performed by a single surgeon. The average corneal power (K), cylinder (AST), and axis values measured by each biometer were compared to those of the Carl Zeiss Atlas 9000 Corneal Topography System. Axial length (AL) measurements were also compared. Predictions of post-operative spherical equivalent (SE) produced by both machines using the Holladay I and SRK/T formulas were compared to the actual post-operative SE at 6 weeks. The threshold for concluding clinical significance was set at 0.50 D of difference in final refractive error.
Results:
The mean absolute errors of the LenStar and IOL-Master respectively as compared to the Atlas were as follows: K errors of 0.27 ± 0.11 D and 0.38 ± 0.11 D, AST errors of 0.27 ± 0.12 D and 0.30 ± 0.11 D, and axis errors of 21.36 ± 8.82 degrees and 20.18 ± 7.96 degrees (95% CI). Although a statistically significant difference was found between K errors (p<0.05), equivalence testing demonstrated no clinically significant difference (within 0.05 D). The other variables demonstrated no significant difference (p>0.05). Equivalence testing showed good agreement of AL measurements (within 0.04 mm). The mean absolute SE prediction errors for the LenStar and IOL-Master respectively were 0.42 ± 0.11 D and 0.39 ± 0.10 D using the Holladay I formula, and 0.40 ± 0.11 D and 0.36 ± 0.10 D using the SRK/T formula (95% CI). These results showed no significant difference (p>0.05), and equivalence testing demonstrated good agreement between refractive outcomes (within 0.09 D).
Conclusions:
The results of this study demonstrated no significant difference between K, AST, axis, or AL measurements, nor between SE predictions produced by the two biometers. This suggests that the two devices have comparable utility in IOL power selection for cataract surgery.
Keywords: 465 clinical (human) or epidemiologic studies: systems/equipment/techniques •
479 cornea: clinical science •
445 cataract