Small incision lenticule extraction (SMILE) is a relatively new laser surgical technique used for the correction of refractive errors. It was developed from the femtosecond lenticule extraction procedure, and was first introduced in 2008 for the correction of myopia.1 Several studies have already reported on the visual and refractive outcomes following SMILE, which has proven to be a safe, predictable, and effective procedure for correcting myopia.2–5 It has been reported that femtosecond lenticule extraction induces fewer fourth-order ocular aberrations than wavefront-guided-LASIK.6 On the other hand, SMILE has been demonstrated to produce a better quality of vision than femtosecond laser-assisted LASIK (FS-LASIK), with a more precise correction of myopia.2 This is due to the fact that SMILE is a flapless, minimally invasive, and all-femtosecond laser refractive procedure, which leads to smaller corneal biomechanical changes and subsequently the induction of fewer higher-order aberrations (HOAs).7 As a result, SMILE is becoming increasingly popular among myopic patients and refractive surgeons alike.
Laser-assisted subepithelial keratomileusis (LASEK) was first introduced in 1998 and is highly successful in correcting mild to moderate myopia. LASEK does not require the creation of a corneal flap and has been shown to induce fewer HOAs than LASIK.8 SMILE and LASEK are both flapless refractive procedures, avoiding the complication of inducing HOAs due to the creation of a corneal flap.
To our knowledge, there is no study comparing the quality of vision following SMILE and LASEK for the correction of mild to moderate myopia. The aim of the current study was to compare the visual quality in patients following refractive surgery using SMILE and LASEK for the correction of myopia, with analysis of wavefront aberrations, contrast sensitivity, and subjective quality of vision.
Patients and Methods
This prospective, non-randomized, comparative study involved 65 eyes of 33 patients (18 men and 15 women) who underwent SMILE and 50 eyes of 25 patients (11 men and 14 women) who underwent LASEK between January and July 2014 at the Eye and ENT Hospital of Fudan University, Shanghai, People's Republic of China.
The main inclusion criteria were: age between 18 and 40 years, a stable refractive error for at least 2 years (<−0.25 diopters [D] change), spherical equivalent (SE) from −1.25 to −6.00 D, corrected distance visual acuity (CDVA) of 20/25 or better, minimum corneal thickness of 490 µm, and no systemic or localized ocular disease. This study was approved prospectively by the institutional review board of The Ethics Committee of the Eye and ENT Hospital of Fudan University. Written informed consent was obtained from all patients before surgery. All patients were treated in accordance with the tenets of the Declaration of Helsinki.
All patients underwent routine preoperative ophthalmologic examination, which included measurements of uncorrected distance visual acuity (UDVA), CDVA, slit-lamp biomicroscopic examination, intra-ocular pressure with non-contact tonometry, refraction (manifest and cycloplegic), dilated fundus examination, corneal topography (Pentacam; Oculus Optikgeräte, Wetzlar, Germany), contrast sensitivity (Takagi Contrast Glare Tester CGT-1000; Takagi Seiko Co. Ltd., NaganoKen, Japan), and wavefront aberrations (WASCA wavefront analyzer; Carl Zeiss Meditec AG, Jena, Germany).
All surgical procedures were performed under topical anesthesia with 0.4% oxybuprocaine hydrochloride (Santen Pharmaceutical Co., Ltd., Osaka, Japan).
For patients in the SMILE group, the surgical procedure was performed using the VisuMax femtosecond laser system (Carl Zeiss Meditec AG). A repetition rate of 500 kHz with a pulse energy of 130 nJ was used to create the posterior and anterior surfaces of the refractive lenticule and the lenticule border. The thickness of the cap was intended to range from 110 to 120 µm, the ablation zone diameter was set to 6.5 or 6.6 mm, and the diameter of the cap was set to 7.5 mm. The refractive lenticule of the intrastromal corneal tissue was extracted through a superior lateral incision opening, 2 mm in length, using surgical forceps.2
For patients in the LASEK group, the surgical procedure was performed using 20% ethanol-aqueous solution to create a corneal epithelial flap, 9 mm in diameter, which was then peeled back with a crescent blade (Model 52424A; 66vision Tech Co., Ltd., SuZhou, China), leaving a hinge at the 12-o'clock position. Corneal stromal tissue ablation was performed using the Mel 80 excimer laser (Carl Zeiss Meditec AG) over an optical zone diameter ranging from 6.25 to 6.75 mm and transition zone diameter ranging from 9.25 to 9.75 mm, at a repetition rate of 250 kHz with a pulse energy of 150 nJ. The epithelial flap was then repositioned and a bandage contact lens (ACUVE OASYS; Johnson & Johnson, New brunswick, NJ) was inserted for 7 days.
Postoperatively, topical steroids (fluorometholone 0.1%; Santen Pharmaceutical Co., Ltd.) were used initially six times daily and tapered for a period of 30 days for SMILE and 60 days for LASEK. Topical antibiotics (ofloxacin ophthalmic solutions 0.5%; Santen Pharmaceutical Co., Ltd.) were used four times daily for 7 days. Tears Naturale (hypromellose 2910, dextran 70, glycerol eye drops; Alcon Laboratories, Inc., Fort Worth, TX) were used four times daily for 60 days postoperatively.
Postoperative Ophthalmologic Examinations
Follow-up examinations, including measurements of CDVA, corneal topography, and wavefront aberrations, were scheduled at 3 months postoperatively. Contrast sensitivity was scheduled 1 and 3 months postoperatively. A quality of vision questionnaire was also conducted at 3 months postoperatively. The questionnaire contained 10 items and was derived from a previously published study.9 Patients were asked to assess the extent of their visual symptoms at 3 months postoperatively and provide a score from 0 to 10, where 0 represented no symptoms and 10 corresponded with a disabling problem. The questionnaire was not included if more than one of the 10 questions were not answered.
The ocular wavefront aberrations were analyzed for a standardized pupil diameter of 6 mm. The Zernike coefficients of vertical trefoil (Z−33), horizontal trefoil (Z33), vertical coma (Z−13), horizontal coma (Z13), and spherical aberration (SA) (|Z04|) were analyzed.10 In addition, the root mean square (RMS), expressed as coma
, HOAs, and total aberrations were calculated because they are clinically significant in determining visual quality.
Contrast sensitivity was measured at six target sizes with and without the presence of glare: 6.3°, 4.0°, 2.5°, 1.6°, 1.0°, and 0.7°.11 There were 13 contrast levels (0.01 to 0.64 contrast or 2.00 to 0.34 log10CS) with an average step size of 0.15 log10CS. Contrast levels for the Takagi CGT-1000 were converted into log contrast sensitivity by taking −log10 (contrast level) to linearize the measurement data, thus facilitating statistical analyses.12
We enrolled the data from the left eye for statistical analysis. All data were analyzed using SPSS 19.0 software (SPSS, Inc., Chicago, IL) and reported as mean ± standard deviation. Graphs were partly plotted by Microsoft Excel 2010 (Microsoft Corporation, Redmond, WA). The Student's t test was used to compare normally distributed data between the two groups, and the paired Student's t test was used to compare data before and after each treatment. The non-normally distributed data were analyzed by the Mann–Whitney rank-sum test and Wilcoxon signed-rank test. A P value less than .05 was considered statistically significant.
All surgeries were uneventful and no specific intraoperative or serious postoperative complications were observed. Sixty-five eyes of 33 patients in the SMILE group and 50 eyes of 25 patients in the LASEK group completed all of the preoperative and postoperative examinations. No difference was found between the two groups in mean age (P = .30, t = −1.046). Both groups were comparable in the results of the preoperative examinations and no patient complained of glare or halos in either group preoperatively. Detailed preoperative clinical data are shown in Table A (available in the online version of this article).
Preoperative Demographic Data
There was no significant difference in UDVA between SMILE and LASEK after treatment (SMILE group = 0.016 ± 0.047 logMAR [20/20 Snellen], LASEK group = 0.042 ± 0.049 logMAR [20/20 Snellen]; P = .05, t = −2.02).
The preoperative RMS of aberrations, including total aberrations, HOA, SA, and coma, was not significantly different between the SMILE and LASEK groups (Table A). Data analysis that compared these parameters before and after treatment demonstrated that HOA and coma increased significantly (paired t test; SMILE group: HOA, t = −3.934, P < .001; coma, t = −3.576, P = .001; LASEK group: HOA, t =−5.491, P < .001; coma t = −2.151, P = .04) in both groups, whereas SA increased significantly (paired t test; t = −5.461, P < .001) in the LASEK group only. Figure 1 displays a graph showing the comparison of mean RMS aberrations between the SMILE and LASEK groups 3 months postoperatively. At 3 months postoperatively, total aberrations HOAs and SA were lower (P < .05) in the SMILE group than that in the LASEK group. The detailed postoperative RMS data are shown in Table B (available in the online version of this article).
Comparison of the wavefront aberrations between the small incision lenticule extraction (SMILE) and laser-assisted sub-epithelial keratomileusis (LASEK) groups 3 months after treatment. *P < .05 = statistically significant. **P < .01 = statistically significant. HOA = higher-order aberrations; SA = spherical aberration; RMS = root mean square
Mean RMS Postoperatively (6 mm Diameter)
Figure A (available in the online version of this article) shows graphs for the photopic and scotopic contrast sensitivity in the SMILE and LASEK groups at six spatial frequencies (6.3° to 0.7°) preoperatively and postoperatively. Preoperatively, there was no significant difference in contrast sensitivity between the two groups. At 1 month after surgery, contrast sensitivity was reduced from baseline in both groups at the spatial frequencies of 0.7° (paired t test; light off contrast sensitivity – SMILE group: t = 3.492, P = .002; light on contrast sensitivity – SMILE group: t = 2.531, P = .02; LASEK group: t = 2.240, P = .04). Contrast sensitivity at this spatial frequency recovered at 3 months postoperatively in both groups. At 1 month after surgery, there was no significant difference in contrast sensitivity between the two groups across all spatial frequencies.
Comparison of (A) photopic and (B) scotopic contrast sensitivity between the small incision lenticule extraction (SMILE) and laser-assisted subepithelial keratomileusis (LASEK) groups at all spatial frequencies (6.3° to 0.7°), preoperatively and postoperatively. Horizontal axis corresponds to visual angle of target size (degree). Vertical axis corresponds to log10 contrast sensitivity.
Quality of Vision Questionnaire
Figure 2 shows the average scores of each question item in the quality of vision questionnaire 3 months postoperatively. No significant difference was observed between the two groups on the items of visual clarity, haze, diurnal fluctuations of vision, and dry eye. For night glare and halos, patients in the SMILE group scored lower (fewer symptoms) than patients in the LASEK group (night glare: SMILE group = 1.25 ± 1.22, LASEK group = 2.32 ± 1.99, P = .03; halos: SMILE group = 0.97 ± 1.12, LASEK group = 1.96 ± 1.77, P = .02), indicating that patients in the SMILE group may have experienced a better quality of vision than patients in the LASEK group. In addition, patients in the SMILE group did not report any trouble with monocular double vision, which was reported by two patients in the LASEK group.
Comparison of the quality of vision questionnaire between the small incision lenticule extraction (SMILE) and laser-assisted subepithelial keratomileusis (LASEK) groups postoperatively. *P < .05 = statistically significant. Haze means the hazy feeling after treatments, vision is the patient's evaluation of his or her visual quality, dry severe is the severity of dry eye, gritty is a foreign body sensation.
Our results demonstrated that SMILE and LASEK are both safe and effective procedures for correcting mild to moderate myopia. A previous study showed that LASEK and LASIK resulted in increases in SA; however, LASIK alone induced aberrations associated with flap creation, whereas LASEK did not.13 Both SMILE and LASEK are flapless refractive procedures. However, there has still been no reported clinical study comparing the quality of vision in patients after SMILE and LASEK.
Sekundo et al. reported that following SMILE, 83.5% of the treated eyes attained UCVA of 1.0 (20/20 Snellen) or better, 53% of the treated eyes attained UCVA equal to their preoperative CDVA, 32.3% gained one line of vision, and 3.3% gained two lines of CDVA.2 According to Vestergaard et al., 95% of the treated eyes achieved UCVA of 20/40 or better at 3 months after SMILE, with all of the patients achieving a visual acuity of 20/40 or better at 6 months postoperatively.14 In a comparative study between SMILE and FS-LASIK, Ganesh and Gupta found that 96% of eyes in the SMILE group attained UCVA of 20/20 or better, with only 88% of eyes in the LASIK group achieving this benchmark at 3 months postoperatively.15 In the current study, the two surgical treatments proved effective with no significant difference observed between the CDVA before treatments and the UCVA achieved after the surgeries in both the SMILE and LASEK groups. In addition, all of the patients attained UCVA of 20/25 or better at 3 months postoperatively, with 93.8% and 94% of the treated eyes in the SMILE and LASEK groups, respectively, achieving UCVA of 20/20 or better. Our findings demonstrated better visual success with both SMILE and LASEK, compared to the results of Sekundo et al.2 and Vestergaard et al.6
The induction of HOAs can cause significant night vision problems such as glare, haze, and halos. In a comparative study of SMILE and FS-LASIK, Ganesh and Gupta15 found that HOAs were significantly fewer in the SMILE group than in the FS-LASIK group postoperatively. Agca et al.16 reported that HOAs, SA, coma, and trefoil increased postoperatively in SMILE group, and Arbelaez et al.17 found a significant postoperative increase of SA in the LASEK group, but there is no study comparing the HOAs after LASEK and SMILE. We found that HOAs, especially SA, increased significantly after LASEK, a finding also observed by McAlinden and Moore.13 The current study showed that SMILE induced fewer HOAs than LASEK, especially SA, which is a significant factor in determining visual quality after refractive surgery.
With regard to contrast sensitivity, our measurements at 6 and 12 cycles/degree demonstrated that both SMILE and LASEK caused a slight reduction in high spatial frequency contrast sensitivity 1 month after surgery, which recovered 3 months postoperatively, comparable to the findings with SMILE reported by Sekundo et al.3
Townley et al.18 found that contrast sensitivity at medium to high spatial frequencies had no significant difference under photopic or scotopic conditions before and 3 and 6 months after LASEK, and our results showed that contrast sensitivity at medium spatial frequencies remained stable after both LASEK and SMILE. A study by Reinstein et al.5 looking at outcomes following SMILE in low myopia showed that contrast sensitivity at 1 year was slightly increased across all spatial frequencies, which our results also demonstrated, with results showing recovery of contrast sensitivity after the surgery.
The Takagi Glare Tester CGT-1000 is still in use clinically, and the results produced by it have been shown to be similar to those produced by other devices, as recently demonstrated by Luger et al., who analyzed contrast sensitivity in patients both before and after LASIK,19 and by Yamauchi et al., who assessed contrast sensitivity in patients both before and after intraocular lens implantation.20 Nonetheless, the device has limitations, such as ceiling and floor effects, which may impair the ability of the test to distinguish between groups. Furthermore, doubling of the screen illumination has been shown in some cases to cause a drop in the median contrast sensitivity.11 As a result of these limitations, the difference in contrast sensitivity between SMILE and LASEK may not be as clearly distinguishable. In a previous study by Ganesh and Gupta, it was found that contrast sensitivity in the SMILE group was better than contrast sensitivity in the LASIK group across all spatial frequencies by 3 months postoperatively.15 Because there has been no study to date comparing the contrast sensitivity between SMILE and LASEK using a more advanced method of contrast sensitivity measurement, such as the Vector Vision CSV 1000E test, there is potential for further investigation into this area.
With regard to ablation depth, differences in the two procedures showed that the lenticule thickness in the SMILE group was on average thicker than the ablation depth in the LASEK group, despite no significant difference in the spherical equivalent between the SMILE and LASEK groups. This is expected, with previous studies having shown that when correcting 1.00 D of myopia, the VisuMax femtosecond laser removes a greater amount of corneal tissue than the MEL-80 excimer laser.21,22
The quality of vision questionnaire used in our study was derived from a previously published study designed to assess visual quality after refractive surgery, with particular emphasis on evaluating symptoms such as glare and halos. Our results showed that patients in the SMILE group achieved better subjective visual function than patients in the LASEK group. Regarding the other items in the questionnaire, such as dry eye, haze, and gritty sensation, there was no significant difference between SMILE and LASEK. Ganesh and Gupta15 assessed quality of vision using a different questionnaire, and also found that the scores for glare complaints were lower following SMILE compared to LASIK. Another study evaluating subjective visual quality using yet another questionnaire observed that most patients reported an improvement in their vision, with no patients complaining of glare after SMILE.2
One limitation of the study is that the questionnaire was not presented preoperatively in the study. However, considering that the quality of vision was associated with the scotopic pupil diameter and refractive error, and seeing as how there were no significant differences in preoperative pupil diameter or refractive error between the two groups, this suggests that the preoperative quality of vision scores are likely to have been comparable between the two groups. This protocol has been previously implemented in a study by Ganesh and Gupta,15 who gave a questionnaire on glare and satisfaction to their patients postoperatively and demonstrated that glare had a more significant impact following LASIK.
Overall, when correlating the aberration results and the quality of vision questionnaires, we can speculate that the disturbance in night vision was almost entirely due to SA rather than coma, which was also reported by Oshika et al.23 The fact that SMILE induced less SA than LASIK may also be supported by Kamiya et al.,7 who observed that the Q value changed less following femtosecond lenticule extraction than it did following wavefront-guided LASIK. This observation may suggest that SMILE is capable of maintaining the normal aspheric system of the cornea, whereas LASIK modifies it, changing the physiological curvature of the anterior corneal surface10 and making the cornea more oblate and flat. It also suggests that SMILE may largely maintain the original shape of the cornea, which provides excellent biomechanical stability and creates good visual quality for patients.24 However, further research is still needed in this area.
The current 3-month prospective study indicates that both LASEK and SMILE are excellent surgical options for the correction of mild to moderate myopia. When considering both objective and subjective visual quality measurements, SMILE may be a superior option over LASEK in providing a painless procedure with high refractive accuracy and subsequently producing a better quality of vision in the early stages after surgery. Considering visual quality after LASEK was not as stable as that after SMILE 3 months postoperatively, long-term observations of the efficacy and stability of the two procedures is needed to determine the long-term visual outcomes following SMILE and LASEK.