Long-term Visual Quality after Small Incision Lenticule Extraction (SMILE) and Laser Epithelial Keratomileusis (LASEK) for Low Myopia

Background: Few studies have reported the visual outcomes of small-incision lenticule extraction (SMILE) and laser-assisted epithelial keratomileusis (LASEK) for myopia correction. This study aims to compare the visual quality and corneal wavefront aberrations after SMILE and LASEK for low-myopia correction. Methods: In this prospective study, we included 29 eyes of 29 patients who received SMILE, and 23 eyes of 23 patients who received LASEK between June 2018 and January 2019. The following measurements were assessed: uncorrected (UDVA) and corrected (CDVA) distance visual acuity, manifest refraction, corneal wavefront aberrations, and subjective visual quality. All patients were followed up for two years. Results: All procedures were uneventful. An ecacy index of 1.19 ± 0.17 was established in the SMILE group and 1.23 ± 0.20 in the LASEK group. No eyes lost more than two lines of CDVA. We found that 93% (27/29) of the treated eyes in the SMILE group and 91% (21/23) in the LASEK group had spherical equivalent (SE) within ± 0.25D. The increases in the total corneal spherical aberration and the corneal front spherical aberration were lower in the SMILE group than in the LASEK group (P < 0.01). In contrast, the increases in the total corneal vertical coma and the corneal front vertical coma in the SMILE group were greater than those in the LASEK group (P < 0.01). Conclusion: Both SMILE and LASEK have good safety, stability, and patient-reported satisfaction for low myopia. SMILE induced less corneal spherical aberration but greater vertical coma than LASEK. lenticule extraction, LASEK laser-assisted sub-epithelial keratomileusis, UDVA uncorrected distance visual acuity, CDVA corrected distance visual acuity, SE spherical equivalent, D diopters, CCT central corneal thickness, IOP intraocular pressure, AL axial


Background
Corneal refractive surgeries have experienced rapid development in the past few decades. Procedures, such as PRK, LASEK, LASIK, femtosecond-laser-assisted LASIK, and small-incision lenticule extraction (SMILE), have shown favorable results in terms of e cacy, safety, stability, and predictability in the correction of myopia. LASEK has gained world-wide acceptance for low-myopia correction. [1][2][3][4][5] First reported in 2011 by Sekundo and Shah[6,7], SMILE is a novel all-in-one procedure without the creation of a corneal ap. So far, SMILE has had good results in the correction of low myopia in several reported studies. [8][9][10] However, to the best of our knowledge, no evidence exists of comparisons of visual outcome and visual quality between SMILE and LASEK applied for low-myopia correction.
In the present study, we aimed to the compare the 2-year visual outcomes, objective and subjective visual quality of SMILE and LASEK applied for low-myopia and myopic astigmatism correction.

Methods
The study was conducted in compliance with the principles of the Helsinki Declaration and was approved by the Ethics Committee of Fudan University's EENT Hospital Review Board. Each patient signed an informed consent form after a detailed explanation of the risks and bene ts of the study.

Inclusion and exclusion criteria
The following inclusion criteria were applied: (1) Age between 20 and 40 years; (2) Spherical error between − 3.00D and − 0.50D, cylinder up to 1.50D, corrected distance visual acuity (CDVA) less than or equal to 0.0 (log MAR); (3) Stable refractive error (annual change of refractive error less than 0.50D in the past two years); (4) Thickness of the residual corneal stromal bed greater than 280 µm; (5) Patients who wore contact lenses were required to discontinue wearing soft ones for one week and hard ones for more than two weeks. Exclusion criteria: (1) Suspicious keratoconus; (2) A history of eye trauma or eye surgery; (3) Other eye diseases or systemic diseases affecting the eyes.
In this study, we included a total number of 52 patients (52 eyes) with low myopia who underwent SMILE or LASEK surgery between June, 2018 and January, 2019 at the Fudan University Eye and ENT Hospital (Shanghai, China). The SMILE group was composed of 29 patients (29 eyes; male: female ratio, 8:21; age: 26.8 ± 5.2 years), and 23 patients (23 eyes) constituted the LASEK group (male: female ratio, 11:12; age: 29.3 ± 5.1 years). All patients were enrolled with one eye, and if both eyes met the criteria, the right eye was selected (Table 1).

Subjective visual quality
Questionnaires were used to obtain information from patients about their subjective visual quality. The common visual complaints after refractive surgery were selected, including glare, halo, starburst, hazy vision, blurred vision, and vision uctuation. The rst ve items were provided with corresponding pictures to reduce the possibility of irrelevant answer. In addition, the patients were asked how much their visual quality improved post-operatively, about their overall satisfaction with the procedure, and whether they would like to recommend the procedure to others.

Surgical techniques
SMILE was performed through the following procedure steps: The Visumax femtosecond laser system (Carl Zeiss Meditec AG, Jena, Germany) with a frequency of 500kHz and expert mode was used. The following settings were applied: pulse energy of 130 nJ, thickness of the cornea cap of 120 µm, optical zone diameter of 6.5 to 6.8 mm, and base thickness of 10 µm. The upper side incision was set at 90° (12 o'clock), and the width was 2.0 mm. Point spacing of 2.5 µm was used for lenticule cutting and cap cutting, and point spacing of 2.0 µm was adopted for lenticule side-cutting and cap-side cutting. Then, the lenticule was separated and removed.
LASEK was performed in the following steps. The Triple-A model of the Carl Zeiss MEL 90 excimer laser system (Carl Zeiss Meditec AG, Jena, Germany) was used. First, the corneal epithelium was soaked in 20% alcohol for 12 seconds, and the upper corneal epithelium ap was separated by the corneal epithelium shovel. The corneal stromal bed was exposed, and excimer laser was applied, with a laser frequency of 500 Hz and an optical zoon diameter of 6.5 mm. The corneal stromal bed was rinsed with Ringer's uid, the upper corneal ap was repositioned, and a corneal bandage lens was placed.

Postoperative medication
Eye drops were used in the SMILE group as follows: 0.5% levo oxacin (Cravit; Santen, Osaka, Japan), four times daily for seven days; 0.1% uorometholone (Fluorometholone; Santen) eight times daily and tapered to one time daily for over 24 days; and arti cial tears (Hyalein, 0.1% hyaluronic acid, Santen) four times daily for one months.
Eye drops were used in the LASEK group as follows: 0.5% levo oxacin (Cravit; Santen) four times daily for seven days. The contact lenses were removed 5-7 days post-operatively. 0.1% uorometholone (Fluorometholone; Santen) eight times daily for one week, seven times daily for one week and then tapered to one time daily for over 14 weeks. Arti cial tears (hyalein, 0.1% hyaluronic acid; Santen) were applied four times daily for three months.

Statistical analysis
Statistical analysis was performed using R version 3.6.2 (R Project for Statistical Computing, http://cran.rproject.org). Continuous variables were expressed as mean ± standard deviation (SD), whereas categorical variables were represented as frequency and percent. Shapiro-Wilk test was executed to examine the data for normal distribution. For normally distributed variables, independent t-tests were performed, whereas the Wilcoxon test was applied for variables which were not normally distributed. The chi-square test was used to assess the statistical signi cance of differences in percentages. P-values less than 0.05 were considered to indicate statistically signi cant differences.

Visual outcomes
All surgical operations were uneventful without any complications such as infection ( Table 2) 3A) and 26% (6/23) of LASEK-treated eyes (Fig. 3B) gained two-line. In either group, no eyes lost two or more lines.
The predictability is represented in Fig. 4A  In terms of stability, the postoperative SE of the SMILE and LASEK groups were − 0.07 ± 0.18D and − 0.02 ± 0.21D, correspondingly. The postoperative astigmatism in the SMILE (Fig. 5A) and LASEK (Fig. 5B) groups was 0.11 ± 0.18D and 0.11 ± 0.22D, respectively.

Corneal wavefront aberrations (6 mm)
Total corneal HOAs and corneal front surface HOAs were increased in both groups post-operatively, but the increases in HOAs were not statistically signi cant (P-values of 1.00 and 0.68, respectively). The postoperative increase of whole corneal spherical aberration and corneal front surface spherical aberration in the SMILE group (Supplemental Fig. 1A) were less than that in the LASEK group (P < 0.01), and the increase of total corneal vertical coma and corneal front surface vertical coma were greater than that of LASEK (P < 0.01) (Supplemental Fig. 1B).

Postoperative subjective visual quality and patient satisfaction
The most common postoperative visual complaints in both groups were starburst, halo and vision uctuation (Table 3). We observed starburst in 38% (11/29) of the SMILE-treated eyes and in 30% (7/23) of the LASEK-treated eyes. Additionally, we established that 35% (10/29) of the SMILE-treated eyes and 17% (4/23) of the LASEK-treated eyes experienced halo. Only 10% (3/29) of the SMILE-treated eyes and 22% (5/23) of the LASEK-treated eyes suffered from blurred vision, and 10% (3/29) of the SMILE-treated eyes and 35% (8/23) of the LASEK-treated eyes experienced visual uctuation. Most of the visual symptoms reported were mild, without daily-life disturbance. It is noteworthy that 93% (27/29) of the SMILE group patients and 87% (20/23) of the LASEK group ones were satis ed with the treatment. Moreover, 97% (28/29) of the SMILE-treated patients and 96% (22/23) of the LASEK-treated ones felt that their visual quality was signi cantly improved as compared with the preoperative levels. Notably, 86% (25/29) of the patients in the SMILE group (Supplemental Fig. 2A) and 91% (21/23) of patients in the LASEK group (Supplemental Fig. 2B) were willing to recommend surgery to myopic patients with similar conditions. Discussion SMILE has been widely accepted for the correction of myopia less than − 12.0 D; LASEK has also been shown to have good visual outcomes in the correction of low myopia. [2,9,11] In the present study, we compared for the rst time the visual quality and corneal aberrations obtained by the two procedures applied for low myopia correction.
Both SMILE and LASEK showed good safety, e cacy, and predictability in the correction of low myopia. The e cacy indexes of the SMILE group and LASEK group were 1.19 ± 0.17 and 1.23 ± 0.20, respectively; the safety indexes were 1.24 ± 0.17 and 1.28 ± 0.18, correspondingly, which was consistent with the results of previous studies. [9,10] Reinstein et al [10] reported that after the application of SMILE (mean SE: -2.61 ± − 0.54 D) for low-myopia with 1-year follow-up, 96% of the patients had UDVA of 20/20 or better and UDVA of 20/25 or better with mean residual SE of -0.05 ± 0.36D. In terms of safety, no patient lost two or more lines. The authors concluded that SMILE had safety and e cacy similar to those of LASIK used for low-myopia. In another study, Autrata et al [12] performed LASEK for low to moderate myopia with 20% alcohol inside the alcohol solution cone for 25-30s. LASEK had faster vision recovery, milder pain, and lower incidence of haze than PRK. At 2-year follow-up visit, the safety and e cacy indexes were 1.04 and 0.98, respectively; 62% of the patients had a postoperative SE within ± 0.5D and 92% within ± 1.0D. Spadea et al [13] used ap-preserved LASEK without alcohol to correct low to moderate myopia and obtained good outcomes with an e cacy index of 0.87 and a safety index of 1.25 after a follow-up of 60 months.
In addition to safety and e cacy, other considerations also should be considered when selecting surgical methods for patients. SMILE and LASEK have their own advantages and disadvantages. For example, SMILE maintains the integrity of the corneal epithelium and Bowman's layer and therefore mild postoperative ocular discomfort, faster recovery and free of ap-related complications. However, SMILE required 10-30 µm of additional base thickness in corneal stroma, and thus more corneal tissue was removed using this technique than the application of excimer laser surgery. Furthermore, the cooperation of patients was critical during the femtosecond laser scanning. In addition, the design of the parameters in the SMILE set for low myopia treatment should be more carefully determined since the outcomes were affected by the adjustment of the nomograms and the applied laser energy and femtosecond laser scanning quality. [14] LASEK was also a apless procedure, which prevented ap-related complications.
Compared with SMILE, less corneal tissue was removed in LASEK, and therefore it was more suitable for patients with relatively thin cornea. Additionally, the excimer laser machine had an eye-tracking system that was valuable in the process of excimer laser scanning, and was hence more appropriate for patients with poor cooperation or large angle kappa. Previous studies had shown that postoperative pain of LASEK might be related to the time and concentration of alcohol used in the procedure. [1,12,15] In this study, 20% alcohol in ltration was applied for 12s, after which the patients reported mild postoperative pain or discomfort. In addition, the medication time after LASEK is longer, and more frequent follow-up is needed.
We found that the increment of total corneal spherical aberrations (SA) and SA of the anterior cornea surface after SMILE were less than those after LASEK. These results were in accord with previous studies indicating that the SA induced by SMILE was less while the coma error was higher than that of FS-LASIK.
[16, 17] The increase of spherical aberration would augment the post-operative occurrence of halo in the dark. Zhu et al [5] and Yu et al[8] also reported signi cantly lower HOAs and SA after SMILE than after LASEK, with no signi cant differences in the coma and trefoil aberrations between groups. In the current study, the vertical coma was signi cantly increased after SMILE, which was in line with the results of previous studies.
[18] Lack of an eye-tracking system as well as involuntary Bell phenomenon in SMILE procedure may contribute to the increase of vertical coma.
In clinic, many patients who complain of visual disturbances after refractive surgeries have visual acuities of 1.0 or better and refractions close to plano, therefore, it is inadequate to assess their symptoms with conventional measures of acuity or refraction. Subjective visual quality after laser surgery should be given full attention. In this study, the main visual disturbances were starburst, halo, and vision uctuation two years after the surgery. These symptoms were not reported to disturb patients' daily life except for inconvenience when driving at night postoperatively, which is consistent with the ndings of Wei et al. [19] Our study has some limitations. First, we observed only the changes in the corneal aberrations, whereas the effect of total ocular aberrations on visual quality was not analyzed. Second, we included only patients with spherical equal to -3.00D or less and astigmatism equal to 1.50D or less, and thus the results could not be extrapolated to patients with spherical − 3.00D and astigmatism − 1.50D or more.
Third, the sample size was small, and thus larger samples of observation with a longer follow-up period should be implemented in the future.

Conclusions
In conclusion, SMILE and LASEK used for correction of low myopia provided good safety, stability, and visual quality. Nonetheless, the underlying mechanism leading to greater vertical coma in SMILE than in LASEK remains to be further explored. -Consent for publication Written informed consent for this article was obtained from individual or guardian participants.

List Of Abbreviations
-Availability of data and materials The data used during the current study are available from the corresponding author on reasonable request.