准分子激光原位角膜磨镶术后屈光回退危险因素的Logistic回归分析  

Conditional Logistic regression analysis of risk factors for refractive regression after laser in situ keratomileusis

作　　者：汪明红[1] 刘磊[2] 李新宇[2] 

机构地区：[1]荆门市第一人民医院眼科,湖北荆门448000 [2]华中科技大学同济医学院附属同济医院眼屈光治疗中心,湖北武汉430030

出　　处：《中华眼视光学与视觉科学杂志》2010年第1期11-14,共4页Chinese Journal Of Optometry Ophthalmology And Visual Science

摘　　要：目的探讨准分子激光原位角膜磨镶术（LASIK）后屈光回退的危险因素。方法回顾分析2001至2004年在武汉同济医院眼屈光治疗中心接受LASIK手术的7537例患者资料．收集其巾屈光回退的病例298例（484眼），按1：1病例对照研究方法选取相等例数非屈光回退病例作为对照组，两组的性别、手术时间等条件配比。收集两组病例的一般资料，如年龄、性别、家族史和服部相关检查结果（等效球镜度、最佳矫正视力、中央角膜厚度、切削直径等）等相关因素。应用SPSS13．0软件对两组的一般情况行）（。检验和t检验，对LASIK术后屈光回退的危险因素进行单因素和多因素条件Logistic回归分析。结果分析结果提示，有12个变量纳入多因素条件Logistic回归模型，有4个因素进入Logistic回归预测模型，其中年龄（OR=1．062，P=0．004）和等效球镜度（SE）（OR=3．747，P=0．001）是屈光回退的危险因素，而中央角膜厚度（CCT）（OR=0．475，P=0．002）和切削直径（LAD）（OR=0．982，P=0．000）是屈光回退的保护阁素。Logistic回归预测模型为：P=1／1＋Exp∑（-7．582—0．060AGE-1．321SE＋0．744LAD＋0．018CCT），表示P值越大．术后越易发生屈光回退。判断术后屈光回退的灵敏度为79．54％，特异度为76．78％。结论年龄越大，屈光度越高和中央角膜厚度越薄者，术后越容易出现屈光回退。术中采用较大的激光切削直径将有助于提高LASIK的预测性，减少屈光回退的发生。本研究建立的Logistic回归模型能较好地预测术后屈光回退的发生概率。Objective To investigate the risk factors for refractive regression after laser in situ keratomileusis （LASIK）. Methods This study was a 1：1 case-control study with 298 cases of refractive regression and 298 controls who were matched by gender and date of surgery. Participants were selected from 7537 patients at Tongji hospital. Data for all cases included age, family history of myopia and ocular region examination results like SE and LAD were collected. Univariate analysis and multivariate conditional Logistic regression were used to analyze the associations between the exposure factors and refractive regression with SPSS 13.0 software. Results Refractive regression was closely related to 4 factors： age （OR=1.062, P=0.004）, spherical equivalent （OR=3.747, P=0.001）, central corneal thickness （OR=0.475, P=0.002）, and laser ablation diameter （OR=0.982, P=0.000）. The first two factors are risk factors and the latter two arm protective factors. The Logistic regression equation was as follows： P=1/1＋Exp∑ （-7.582-0.060AGE-1.321SE＋0.744LAD＋0.018CCT）. The sensitivity and specificity of the Logistic model were 79.54% and 76.78%, respectively. Conclusion In cases with older age, thinner preoperative corneal thickness and higher myopic diopter, refractive regression occurs easily after surgery. A large laser ablation diameter will help to improve predictability and reduce regression. The Logistic regression model based on these factors is reliable in predicting refractive regression after LASIK.

关 键 词：角膜磨镶术 激光原位 屈光回退 LOGISTIC回归分析 

分 类 号：R77[医药卫生—眼科]