目的：SRK-Ⅱ法和SRK-T法在计算白内障患者人工晶状体(intraocular lens，IOL)影响的临床研究，为临床不同眼轴的白内障患者术前选择IOL测量公式及预测合适的IOL屈光度数提供依据。

方法：随机选择2013-04/2015-11本院收治的不同眼轴的白内障患者160例200眼。男92例120眼，女68例80眼，平均年龄66.2±4.36岁。术前使用A型超声仪测量研究对象眼轴长度(axial length，AL)，根据长度划分成四组，短眼轴为A组，正常眼轴为B组，长眼轴为C组，超长眼轴为D组，然后分别用SRK-Ⅱ和SRK-T计算A、B、C和D组的预测术后IOL屈光度数，植入相应的IOL(美国AMO人工晶状体)，术后1wk，1、3mo分别通过验光、检影相结合的方法测量最佳矫正视力(best corrected visual acuity，BCVA)的实际屈光度数。计算平均绝对屈光误差值(mean absolute refractive error，MAE)。

结果：SRK-Ⅱ和SRK-T两种方法术后1wk和术后1mo、术后1wk和术后3mo的MAE差异均有统计学意义(P<0.05)，术后1mo和术后3mo对比无统计学差异(P>0.05)。SRK-Ⅱ和SRK-T在A组的对比，差异无统计学意义(P>0.05)，而在B、C和D组的对比差异都有统计学意义(P<0.05)。A组患者术后1mo比术后1wk更趋于近视，差异有统计学意义(P=0.035)； B组和C组中，患者在术后1mo和术后1wk两个时间段均出现了近视漂移现象，差异无统计学意义(P=0.84、0.88)； D组患者术后1mo比术后1wk更趋于远视，差异有统计学意义(P=0.041)。

结论：本研究表明，术后1mo屈光状态基本稳定，为配镜提高依据； 两种方法在A、B组中预测准确性接近，在C、D组预测IOL屈光度数误差对比中，SRK-Ⅱ高于SRK-T，SRK-T方法在计算不同眼轴的白内障患者IOL优于SRK-Ⅱ。

AIM: To study the SRK-II and SRK-T in clinic for calculating intraocular lens(IOL)in cataract patients, and to provide the basis for preoperative selection of IOL measurement formula and prediction of appropriate IOL diopter in cataract patients with different axial.

METHODS: Randomized selection of 160 cataract patients of 200 eyes with different axial from April 2013 to November 2015 admitted to the hospital were taken. There were 92 males with 120 eyes, 68 females with 80 eyes, the average age of 66.2 ± 4.36 years old. The axial length(AL)was measured by type A ultrasonography. They were divided into four groups according to AL. Patients with shorten AL were Group A, with normal AL were Group B, with lengthening AL were Group C, with extremely AL were Group D. The IOL diopter of the four groups were calculated by SRK-Ⅱ and SRK-T, and the corresponding IOL(American AMO intraocular lens)was implanted. The actual diopter at best corrected visual acuity(best corrected visual acuity, BCVA)was measured by optometry and retinoscopy at 1wk, 1, 3mo after operations. The mean absolute refractive error(MAE)was calculated.

RESULTS: The MAE of the SRK-Ⅱ and SRK-T at 1wk after operations was different with that at 1mo and at 3mo(P<0.05), that at 1mo was not significantly different with that at 3mo(P>0.05). there was no difference between SRK-Ⅱ and SRK-T in Group A(P>0.05), but there were significant differences in Group B, Group C and Group D(P<0.05). Patients of Group A more tended to become myopia at 1mo than at 1wk(P=0.035). Patients of Group B and C both got myopia shift at 1wk and 1mo after operation(P =0.84, 0.88). Patients of Group D tended to become hyperopia at 1mo than at 1wk(P=0.041).

CONCLUSION: This study shows that refraction become stable at 1mo after operations; the accuracy of the two methods are nearly same in Group A and B, while in Group C and D, SRK-Ⅱ is better than SRK-T on the comparison of MAE. SRK-T is better than SRK-Ⅱ on IOL calculation in patients with different AL.