Analysis of astigmatism after laser in situ keratomileusis for spherical myopia

Objective:To study the rules and causes of astigmatic change after laser in situ keratomileusis(LASIK) for spherical myopia and the measures to prevent it.Methods:137 patients(188 eyes) with spherical myopia who underwent myopic LASIK were included in this study. Average age of patients was 24.38 years(18～50 years). Of the 137 patients,69 were men and 68 were women. Average refraction before surgery was -6.03D(-1.00～-13.75D). Average astigmatism was 0.10D(-0.25～+0.25D). The surgery merely treated myopia in spite of minor astigmatism. Preoperative examination included distant and near visual acuity with the naked eye,best spectacle corrected visual acuity(BSCVA),refraction,slit lamp microscope examination,fundus examination,intraocular pressure,central corneal thickness and computerized corneal topography. Some of these examinations, such as visual acuity and refraction,were performed postoperatively at 1 day,1 week and 1,3 and 6 months. An automated refractor TOPCON 8100 was used for refractions. Both preoperative and postoperative astigmatism were determined in this way. One surgeon performed all surgeries. A 9.0 mm diameter superior hinged or nasal hinged corneal flap measuring approximately 160 μm in thickness was created using a Moria microkeratome. This was followed by a spherical midstromal ablation using the Visx Star S2 excimer laser. Laser parameters included a repetition rate of 10 Hz,radiant exposure at the corneal plane of 160 M j/cm 2 ablating diameter of 6.0 mm. A photorefractive keratectomy ablation algorithm with multizone pattern was programmed into the laser computer. After stromal ablation,the flap's posterior surface and corneal stromal bed were irrigated with normal saline. The flap was repositioned to its original location. Antibiotic and corticosteriod drops were administered postoperatively. Information on all patients was input into a database. The data was calculated with SPSS software,P0.05 was considered significant.Results:A patient with astigmatism ≤0.5 D was assigned to the no astigmatism group;0.5 D was assigned to the astigmatism group. Myopic astigmatism with 0°～22.5° or 157.5°～180° axis and hyperopic astigmatism with 67.5°～112.5° axis were regarded as with the rule astigmatism;hyperopic astigmatism with 0°～22.5° or 157.5°～180° axis and myopic astigmatism with 67.5°～112.5° axis were regarded as against the rule astigmatism;astigmatism with 22.5°～67.5° or 112.5°～157.5° axis was regarded as oblique astigmatism. The number and percentage of patients in the astigmatism group and no astigmatism group was assessed at different times after surgery. It was found that the astigmatism group had 78 eyes(41.49%) and the no astigmatism group had 110 eyes(58.51%) at 1 day after surgery. As time passed,the number of eyes in the astigmatism group decreased gradually and the converse was true for the no astigmatism group. At 3～6 months after surgery,the astigmatism group had 52 eyes(27.66%) and the no astigmatism group had 136 eyes(72.34%). In the astigmatism group,there was no dominant axis;with the rule astigmatism tended to decrease,and there were 30 eyes(15.96%) and 9 eyes(4.79%),at 1 day and 3～6 months,respectively,after surgery. The amplitude of astigmatism after surgery was much higher than the preoperative amplitude(P0.01). It reached a peak 1 day postoperatively(0.64±0.52D),then gradually decreased. The differences among the amplitudes of astigmatism at 1 week,1 month and 3～6 months postoperatively were not significant(P0.10). There was a preoperative range of -0.25～0.25D for astigmatism. It changed remarkably after surgery. Postoperatively,it was -2.25～2.00D,-2.00～2.00D,-1.75～1.75D and -1.50～1.00D at 1 day,1 week-1 month and 3～6 months,respectively. The number of eyes with astigmatism of 1 D was 29 at 1 day,16 at 1 week,11 at 1 month and 7 at 3～6 months postoperatively. The mean and SD of preoperative best spectacle corrected visual acuity(BSCVA) and postoperative uncorrected visual acuity(UCVA) were calculated and compared using a pai