角膜曲率对人工晶状体屈光度计算公式在高度近视伴后巩膜葡萄肿白内障中的影响研究

The Effect of Corneal Curvature on Intraocular Lens Power Calculation Formulas in Cataract of High Myopia with Posterior Staphyloma

摘要: 目的：分析角膜曲率对Haigis，SRK-II，Hoffer-Q，Hollday-1及SRK-T 5种人工晶状体屈光度计算公式在高度近视伴后巩膜葡萄肿合并白内障患者中的影响，从而为眼科临床工作提供理论依据。方法：选取我院行白内障手术的高度近视并后巩膜葡萄肿的患者59例(72眼)，术前分别用IOL Master测量眼轴长度，角膜曲率，前房深度等参数，然后选择眼轴长度为26.0 mm ≤ AL ≤ 30.0 mm的患者，按照角膜曲率分为A组(K ≤ 43.0 D)，B组(43.0 D < K ≤ 45.0 D)，C组(K > 45.0 D)。分别对每组患者采用IOL Master500系统带的Haigis，SRK-II，Hoffer-Q，Hollday-1，SRK-T 5种人工晶状体屈光度计算公式预测术后屈光，术后3个月联合显然验光及电脑验光的实际屈光值，与各公式预测的实际平均屈光误差(MNE)及平均绝对屈光误差(MAE)进行比较。从而评价各种人工晶状体计算公式在高度近视伴后巩膜葡萄肿患者的优劣。结果：Haigis，SRK-II，Hoffer-Q，Hollday-1，SRK-T公式在A组计算的MNE各个公式之间比较无显著差异(F = 0.242, P = 0.913)；B组公式之间比较有显著差异(F = 1.040, P = 0.393)；C组间比较出现了差异(F = 2.561, P = 0.047)，SRK-II与其余各组比较均有意义(P < 0.05)。在MAE的比较中，A组中各公式之间比较无明显差异(F = 0.128, P = 0.971)；B组各组之间的比较有明显意义(F = 3.407, P = 0.013)；C组组间比较出现了差异(F = 4.762, P < 0.05)；SRK-II与其余各组比较均有意义(P < 0.05)；Haigis与其余各组之间均出现了差异(P < 0.05)。结论：在高度近视伴后巩膜葡萄肿的患眼中，当K ≤ 43.0 D及K > 45.0 D时，Haigis最准确，SRK-T，Hoffer-Q次之，SRK-II准确性最差；43.0 D < K ≤ 45.0 D，SRK-T更准确，其次是Haigis，Hollday-1，Hoffer-Q，SRK-II。

Abstract: AIM: To analyze the effect of Haigis, SRK-II, Hoffer-Q , Hollday-1, SRK-T formulas for calculating the intraocular lens (IOL) power for eyes of high myopia with posterior staphyloma, and make guidance for the practical application of clinical work. METHODS: Prospective clinical study. Seventy-two eyes of fifty-nine cases of high myopia with posterior staphyloma were treated with cataract surgery in our hospital. IOL Master was used to measure axial length (AL), cornea curvature (K), anterior chamber depth (ACD) and other parameters before operation, and the patients with 26.0 mm ≤ AL ≤ 30.0 mm were selected, which were then divided into 3 groups: group A (K ≤ 43.0 D), group B (43.0 D < K ≤ 45.0 D), group C (K > 45.0 D). They all underwent regular phacoemulsification and posterior chamber IOL implantation. The actual postoperative refraction was measured with the methods of phoropter and subjective optometry 3 mo after surgery. Then we took MNE (mean numerical error) and MAE (mean absolute error) as comparison parameters to compare the differences of the predicted and actual postoperative refraction of the five formulas in each group. RESULTS: MNE: There was no significant difference in group A (F = 0.242, P = 0.913) and B (F = 1.040, P = 0.393); There was a difference between group C (F = 2.561, P = 0.047), and the compari-son between SRK-II and other groups was significant (P < 0.05). MAE: There was no significant dif-ference among the formulas in group A (F = 0.128, P = 0.971), but group B had obvious significance (F = 3.407, P = 0.013); and comparing the differences between group C (F = 4.762, P < 0.05), SRK-II and Haigis, all showed difference and the data were statistically significant (P < 0.05). CONCLUSION: When the cornea curvature was too large or too small (K ≤ 43.0 or K > 45.0 D), Haigis was the most accurate, then the SRK-T and Hoffer-Q, SRK-II the last. But when cornea curvature was normal (43.0 D < K ≤ 45.0 D), the order was SRK-T, Haigis, Hollday-1, Hoffer-Q, SRK-II.