目的：比较飞秒激光微小切口基质透镜取出术(SMILE)术后角膜基质实际切削量与术前预测量间差别，评估SMILE基质切削的可预测性与精确性。方法：前瞻性研究。选取2019-01/07在云南大学附属医院行SMILE术近视矫正患者113例220眼，术前，术后1、3mo行常规检查，包括视力、非接触式眼压、等值球镜度(SE)、角膜平均曲率、角膜前表面球面系数、Pentacam眼前节分析。随机选取研究对象中102眼于术前及术后3mo运用A超角膜测厚仪测量角膜中央厚度(CCT)。术后实际切削量为术前角膜最薄点厚度与术后角膜最薄点厚度的差值，误差量为术前预计切削量与术后实际切削量的差值。观察切削误差大小，并对误差量与术前各生理参数进行相关性分析。结果：SMILE术后效果良好，术后1、3mo角膜形态和视力相对稳定。A超角膜厚度测厚仪测量数据与Pentacam眼前节分析系统中角膜最薄点数据一致性较好。术前预计切削量101.36±18.91μm，术后1、3mo实际切削量88.89±18.69、84.95±18.64μm比较有差异(F=334.65，P<0.01)，术后1、3mo时切削误差量为12.59±9.78、16.50±9.21μm，两者间比较有差异(t=-8.370，P<0.01)。术后1、3mo切削误差量均只与术前SE具有相关性(r=0.299，P<0.01； r=0.305，P<0.01)。术后1、3mo SE与同时间点切削误差量具有相关性(r=-0.275，P<0.01； r=-0.306，P<0.01)，随着切削量误差增加，术后等值球镜向负镜度偏移。结论：SMILE术后角膜基质实际切削量小于术前预测切削量，随着术前屈光度增大，预测切削误差亦增大； 随着误差量增大，术后屈光度逐渐呈负向偏移，但此误差在术后早期并未影响目标视力。

AIM: To compare the difference between the actual cutting amount and the preoperative predicted amount of corneal stroma after the small incision lenticule extraction(SMILE), and evaluate the predictability and accuracy of SMILE for corneal stroma. METHODS: Prospective study. A total of 113 myopic patients(220 eyes)who had taken SMILE in the Affiliated Hospital of Yunnan University were selected, and routine examinations were carried out before and 1,3mo after operation, including visual acuity, non-contact tonometer(NCT), spherical equivalents(SE), mean corneal curvature, spherical coefficient of anterior corneal surface and Pentacam anterior segment analysis. All the 102 eyes in the research objects were randomly selected to measure the central corneal thickness(CCT)with the A-supersonic cornea thickness gauge before and 3mo after operation. The actual cutting amount after operation is the difference between the thickness of the thinnest spot of the cornea before and after operation, and the error amount is the difference between the predicted cutting amount before operation and the actual cutting amount after operation. The cutting error amount was observed and its correlation with physiological parameters before operation was analyzed. RESULTS: SMILE had a good performance and the corneal morphology and visual acuity were relatively stable 1 and 3mo after operation. The consistency was good between the data measured by the A-supersonic cornea thickness gauge and the data of the thinnest spot of the cornea in the Pentacam anterior segment analysis, where the difference had no statistical significance(t= -1.877, P=0.063). The difference between the predicted cutting amount before operation(101.36±18.91)μm, and the actual cutting amount 1mo after operation(88.89±18.69)μm and 3mo after operation(84.95±18.64)μm(F=334.65, P<0.01)had statistical significance; There was statistical difference between the cutting amount 1 and 3mo after operation, and the predicted errors before operation [(12.59±9.78)μm and(16.50±9.21)μm]. The cutting amount errors were only correlated with the preoperative equivalent diopter(r=0.299, P<0.01)and(r=0.305, P<0.01). The equivalent diopter at 1 and 3mo after operation was correlated with the cutting amount error at the same time(r=-0.275, P<0.01)(r= -0.306, P<0.01). With the increase of the cutting amount error, the postoperative spherical equivalent shifted to negative.CONCLUSION: The actual cutting amount of corneal stroma after SMILE is smaller than the predicted preoperative cutting amount, and the predicted cutting amount error increases with the increase of preoperative diopter. As the cutting amount error increases, postoperative diopter gradually shifted to negative. The error, however, does not influence the target's visual acuity in the early postoperative period.

国家自然科学基金项目(No.82160204)