高角膜屈光力白内障患者人工晶状体屈光力的计算

目的①探讨角膜屈光力异常增大对人工晶状体（intraocular lens，IOL）屈光力计算准确性的影响。②比较不同计算公式对高角膜屈光力患者IOL屈光力计算的准确性。方法对33例（43眼）角膜屈光力高于47D的白内障患者行超声乳化白内障吸除联合IOL植入手术。术前采用Orbscan-Ⅱ眼前节分析系统测量角膜屈光力，根据高斯光学理论和临床资料推导得出IOL屈光力理论计算公式（本方法）。采用本方法、Holladay2、Holladay、Hoffer Q、SRFUT与SRK2公式对IOL屈光力进行计算，植入IOL术后人工晶状体眼实际屈光状态与术前预测值的差异为预测误差（predictive error，PE），预测误差的绝对值为绝对预测误差（absolute predictive error，AE），分别计算不同计算公式的PE与AE。用SPSS11.0软件进行如下分析：①比较不同公式AE的差异。②回归分析不同公式AE与角膜屈光力的相关性。结果①本方法、。Holladay2、Holladay、Hoffer Q、SRK／T与SRK2公式所产生的AE分别为（0.28±0.17）D、（0.24±0.18）D、（0.27±0.17）D、（0.25±0.23）D、（0.59±0.35）D及（0.83±0.48）D，SRK／T与SRK2公式产生的AE高于所研究的其他计算公式，差异具有显著性。②SRK2公式与SRK／T公式的AE分别与角膜屈光力呈正相关（SRK2公式：r^2=0.522，F=44.82，P〈0.01；SRK／T公式：r^2=0.443，F=32.63，P〈0.01），本方法、Holladay2、Holladay及Hoffer Q公式产生的AE与角膜屈光力不相关。结论对于角膜屈光力异常增大的白内障患者，SRFUT与SRK2公式的计算准确性较差，两者的计算误差与角膜屈光力大小相关，本方法、Holladay2、Holladay与Hoffer Q计算的准确性不受角膜屈光力的影响，其计算结果较为可靠。

Intraocular lens power calculation for cataract patients with high corneal refraction

Objective To compare methods for calculating intraocular lens （IOL） power for cataract surgery on eyes which have high corneal refraction, Methods Thirty-three patients （43 eyes） who had corneal refraction over 47 D were analyzed, Orbscan-Ⅱ Z was used to measure the total axial power of the cornea and the horizontal corneal white-to-white distance. A-scan was used to measure axial length, anterior chamber depth and lens thickness. IOL power was calculated by SRK2, SRK/T, Hoffer Q, Holladay, Holladay2 formulas and a formula developed by the authors （our formula）. Phacoemulsification and IOL implantation were performed. The accuracy of each formula was judged by its comparison to the refraction after IOL implantation. The relationship between the absolute predictive error and corneal refraction was also analyzed by SPSS11.0 software. Results The absolute predictive errors of our formula, Holladay2, Holladay, Hoffer Q, the SRK/T and the SRK2 formulas were （0.28±0.17）D, （0.24±0.18）D, （0.27±0.17）D, （0.25±0.23）D, （0.59±0.35）D and （0.83±0.48）D, respectively. The absolute predictive errors of the SRK2 and the SRK/T formulas were higher than the other formulas. The linear regression trend line reflected the high correlation between the absolute predictive errors of the the SRK2 （r^2=0.522, F=44.82, P〈0.001） and SRK/T （r^2=0.443, F=32.63, P〈0.01） formulas and corneal refraction, No correlation was built between absolute predictive error of the other formulas and corneal refraction. Conclusion High corneal refraction may cause large errors in the calculation of IOL power. Hoffer Q, Holladay, Holladay2 formulas and our formula can provide more highly accurate IOL power calculations than the SRK2 and SRK/T formulas in cataract patients with high cornea refraction.