J Refract Surg. 2021 Aug;37(8):538-544. doi: 10.3928/1081597X-20210506-01. Epub 2021 Aug 1.
To analyze the results of new intraocular lens (IOL) formulas (Emmetropia Verifying Optical [EVO], Kane, Olsen, and Barrett Universal II), traditional formulas (Haigis and SRK/T), and modified Wang-Koch axial length adjustment formulas with the SRK/T and Holladay 1 (SRK/T and H1) in Chinese patients with long eyes.
In this retrospective case series, patients with an axial length of 26 mm or greater having uneventful femtosecond laser-assisted cataract surgery with one trifocal IOL model were enrolled. The actual postoperative spherical equivalent of the manifest refraction was compared with the formula-predicted refraction based on the implanted IOL power. A subgroup analysis was performed based on the axial length.
A total of 113 eyes was enrolled. Using User Group for Laser Interference Biometry constants, the modified Wang-Koch formulas had the lowest percentage of eyes with hyperopic outcomes. The Barrett Universal II, Olsen, Kane, and EVO 2.0 formulas produced a statistically lower median absolute error than the SRK/T and SRK/T formulas ( < .05). The Barrett Universal II formula produced higher percentages of eyes within ±0.50 diopters (D) of the prediction error than the SRK/T formula ( < .05). In eyes with axial lengths of less than 28 mm, there were no significant differences in the prediction accuracy of the eight formulas. In eyes with axial lengths of 28 mm or greater, the new IOL formulas yielded the lowest median absolute error, followed by the H1 and Haigis formulas. The SRK/T formula had the highest mean absolute error and the lowest percentages of eyes within ±0.25 and ±0.50 D of endpoint. The traditional formulas yielded the highest risk of refractive surprise.
All formulas achieved good results in eyes with axial lengths of less than 28 mm with trifocal IOL implanted. The newer formulas tend to produce better outcomes for eyes with high myopia. The SRK/T formula provided improved accuracy only in eyes with axial lengths of 30 mm or greater. .
分析新的人工晶状体（IOL）公式（Emmetropia Verifying Optical [EVO]、Kane、Olsen 和 Barrett Universal II）、传统公式（Haigis 和 SRK/T）和经 Wang-Koch 轴向长度调整公式改良的 SRK/T 和 Holladay 1（SRK/T 和 H1）在中国人长眼患者中的结果。
在这项回顾性病例系列研究中，纳入了轴向长度为 26 毫米或更长、无并发症的飞秒激光辅助白内障手术后植入一种三焦点 IOL 模型的患者。实际术后的屈光球镜等效值与基于植入 IOL 屈光力的公式预测屈光值进行比较。根据轴向长度进行了亚组分析。
共纳入 113 只眼。使用 User Group for Laser Interference Biometry 常数，改良的 Wang-Koch 公式使远视结果的眼比例最低。Barrett Universal II、Olsen、Kane 和 EVO 2.0 公式的中位绝对误差均低于 SRK/T 和 SRK/T 公式（<0.05）。Barrett Universal II 公式产生的预测误差在±0.50 屈光度（D）内的眼比例高于 SRK/T 公式（<0.05）。在轴向长度小于 28 毫米的眼中，这 8 种公式的预测准确性没有显著差异。在轴向长度为 28 毫米或更长的眼中，新的 IOL 公式产生的中位绝对误差最低，其次是 H1 和 Haigis 公式。SRK/T 公式的平均绝对误差最高，预测误差在±0.25 和±0.50 D 内的眼比例最低。传统公式的屈光意外风险最高。
所有公式在植入三焦点 IOL 的轴向长度小于 28 毫米的眼中均取得良好结果。对于高度近视的患者，较新的公式往往会产生更好的结果。只有在轴向长度为 30 毫米或更长的眼中，SRK/T 公式才会提高准确性。
