Comparison of ocular biometry and refractive outcome between a new and classic optical biometer
Haowen Lin1, Xiaohang Xie, Ruoxi Gao
1From the State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangdong Provincial Key Laboratory of Ophthalmology and Visual Science, Guangzhou, China (Lin, Xie, Gao, Jing, Chen, Jinhong Zhang, Qiu, Jiaqing Zhang, Luo); Guangdong Provincial Clinical Research Center for Ocular Diseases, Guangzhou, China (Lin, Xie, Gao, Jing, Chen, Jinhong Zhang, Qiu, Jiaqing Zhang, Luo).
Purpose:
To compare the ocular biometry and refractive outcome between 2 optical biometers.
Setting:
Zhongshan Ophthalmic Center, Guangzhou, China.
Design:
Prospective observational study.
Methods:
953 patients with cataract underwent preoperative biometry including ZW-30 sum-of-segments (SOS) method (ZW SM ), ZW-30 composite method (ZW CM ), and IOLMaster 700. Agreement of axial length (AL), with or without Cooke-modified AL (CMAL) adjustment, was analyzed using Bland-Altman 95% limits of agreement (LoA). Subgroup analysis was used based on ALs (short eyes: AL <22 mm; normal eyes: 22 mm ≤ AL <26 mm; long eyes: AL ≥26 mm). Refractive prediction accuracy was evaluated using the Emmetropia verifying optical (EVO) 2.0 formula and its SOS-optimized version (EVO 2.0 SOS ).
Results:
In short and normal eyes, narrow 95% LoAs (<0.2 mm) were identified among 3 ALs. However, AL obtained by ZW SM was lower compared with this obtained by ZW CM and IOLMaster 700 (95% LoA -0.39 to 0.01 mm; -0.38 to 0.04 mm) in long eyes. CMAL adjustment enhanced the agreement of AL between ZW SM and ZW CM (95% LoA -0.01 mm to 0.02 mm), ZW SM and IOLMaster 700 (95% LoA -0.10 mm to 0.07 mm) in long eyes. Myopic prediction errors (PE) have been identified in the use of ocular biometric parameters obtained from ZW SM (mean PE [ME]: EVO 2.0, -0.19 diopters [D]; EVO 2.0 SOS , -0.18 D). After adjusting ME to zero, no difference was observed in PE calculated using any combination of formulas based on biometric measurements from 3 devices.
Conclusions:
This new segmented biometer demonstrated excellent agreement with IOLMaster 700 in short and normal eyes. However, ALs obtained by IOLMaster 700 are not interchangeable with the SOS method and require CMAL adjustment in long eyes. The application of the SOS method's ocular biometric parameters in refractive prediction led to myopic errors, which suggests constant optimization.
