Effective attenuation of astigmatism has been a significant challenge for refractive surgery. An accurate astigmatic correction is essential for achieving an optimal uncorrected distance visual acuity (UDVA) after refractive surgery. Several studies reported that the small incision lenticule extraction (SMILE) was effective and safe in correcting low to moderate myopic astigmatism in the short term.1–3 However, an undercorrection by nearly 13% per diopter (D) for low and moderate astigmatism (< 2.00 D) and nearly 16% per diopter for high astigmatism (> 2.00 D) was reported.4,5 The undercorrection of astigmatism was primarily attributed to cyclotorsion of the astigmatism.6 Consequently, some scholars recommended the manual compensation of cyclotorsion during SMILE, for which the short-term refractive outcomes after SMILE for correcting myopic astigmatism are satisfying.5,7,8 However, there is a dearth of research reporting the long-term vector outcomes of SMILE in correcting the moderate to high myopic astigmatism. Consequently, this prospective, observational study assessed the long-term refractive outcomes and vector analyses after SMILE for eyes with moderate and high astigmatism using the horizontal slit-beam marking method.
Patients and Methods
A total of 71 patients (71 eyes) were enrolled in this prospective study. The SMILE procedures were conducted in the refractive surgery unit of Zhongshan Ophthalmic Center, Sun Yatsen University, Guangzhou, People's Republic of China, from June 2015 to February 2016. The current study was performed in accordance with the tenets of the Declaration of Helsinki. The protocols of this single-center, prospective study were approved by the Research Ethics Board of the Zhongshan Ophthalmic Center of Sun Yatsen University (Certificate No. 2013MEKY036).
The study's inclusion criteria were a minimum age of 18 years, stable myopia for at least 1 year, moderate to high refractive astigmatism (cylinder refraction ≥ 1.50 D), emmetropia as the attempted correction (eyes targeted for monovision were excluded), and a calculated residual stromal thickness exceeding 250 µm. The exclusion criteria were a history of keratoconus, previous corneal lesions, prior corneal surgery, severe cataracts, glaucoma, or posterior abnormalities (eg, choroidal neovascularization, retinoschisis, retinal detachment, or macular holes). The patients who met the inclusion criteria were counselled to explain the known side effects and complications of SMILE surgery. Written informed consent was obtained from the patients.
To preclude surgeon bias, all SMILE surgeries were performed by a single, highly experienced SMILE surgeon (QL). The VisuMax 500-kHz laser system (Carl Zeiss Meditec AG) had a repetition rate of 500 kHz, pulse energy of 140 nJ, and spot diameter less than 3 µm.
The astigmatic axis marking method included three steps, which have been described in our previous work.9,10 First, the horizontal axis between 0° and 180° could be marked accurately on the cornea with a horizontal slit beam using a standard corneal marking pen. Second, centration on the desired corneal vertex location was achieved during docking and before vacuum suction. Third, the manual compensation of cyclotorsion during SMILE was performed before the laser application proceeded. In theory, this astigmatic axis marking method could help to prevent decentration and misalignment.
The intended optical zone was set from 6.5 to 7.1 mm to keep the residual bed thickness within the safety limit and minimize the risk of postoperative glare. The cap diameter varied from 7.5 to 7.9 mm, and the intended cap thickness was between 120 and 130 µm. The added magnitude of −0.25 D sphere refraction was used for patients younger than 40 years.
Manifest refraction was measured preoperatively and at 1, 3, 6, and 30 months postoperatively, including uncorrected and corrected distance visual acuity (UDVA and CDVA) measured with the Snellen chart. The efficacy index was calculated by dividing the postoperative UDVA (in decimal) by the preoperative UDVA (in decimal). The safety index was calculated by dividing the postoperative CDVA (in decimal) at 30 months by the preoperative CDVA (in decimal). The attempted spherical equivalent (SE) was SE planned for correction by SMILE. The achieved SE was the actual SE corrected by SMILE, which was calculated by subtracting the postoperative SE from the preoperative SE, based on manifest refraction.
The Oculyzer (Oculus Optikgeräte GmbH) was used preoperatively and at 30 months postoperatively. Only scans graded as “OK” by the instrument were used for further analyses. Posterior corneal curvature was analyzed as axial power, presented as traditional keratometric values, including mean (Km) and astigmatism power (steep [K1] – flat [K2]).
The evaluation of astigmatism was based primarily on the definitions and formulas given by Alpins and Goggin.11 Only right eyes were included in this study. Further, the refraction data were analyzed in the negative cylinder form. The target induced astigmatism (TIA) was defined as the vector between the preoperative astigmatism and target astigmatism. The TIA value was equal to the preoperative cylinder in this study because the target refraction was emmetropia. Accordingly, the difference vector (DV) was equal to the postoperative cylinder.
The surgically induced astigmatism (SIA) was defined as the vector difference between the preoperative cylinder and achieved postoperative cylinder. The index of success (IOS) was the ratio of the magnitude of DV and the magnitude of TIA. The IOS indicated whether there was residual astigmatism after SMILE. The correction index (CI) was the ratio of the magnitude of SIA and TIA, indicating whether the intended correction was successfully achieved (CI = |SIA| / |TIA|).
All statistical analyses were performed using IBM SPSS Statistics software version 21.0 (SPSS, Inc). For multiple comparisons of preoperative and postoperative quantitative results, repeated measures of general linear model were applied. The mean ± standard deviation (range) was used for all variables. A P value of less than .05 was considered statistically significant.
A total of 71 eyes from 71 patients with a mean ± standard deviation (SD) age of 24.37 ± 6.52 years (range: 18 to 40 years) underwent SMILE. The pre-operative SE was −6.18 ± 0.70 D (range: −2.38 to −9.25 D), and the preoperative cylinder refraction was −2.07 ± 0.69 D (range: −1.50 to −4.75 D). The baseline characteristics are summarized in Table 1 and Table A (available in the online version of this article). In Table A, the eyes were subdivided into two groups based on a lower and higher cylinder power. The eyes of the moderate astigmatism group had moderate cylinder refraction ranging from −1.50 to −2.00 D, whereas the eyes of the high astigmatism group had high cylinder refraction ranging from −2.25 to −4.75 D preoperatively. But these two groups had similar preoperative SE (P = .081).
Preoperative Patient Demographics in Eyes Undergoing SMILE
Preoperative and Postoperative Refractive Outcomes Stratified by Preoperative Cylinder
At 30 months postoperatively, the efficacy index was 1.04 ± 0.18; 93% of eyes had a postoperative UDVA of 20/20 or better (Figure 1A). As shown in Figure 1B, 90% of the eyes had a postoperative UDVA the same or better than the preoperative CDVA. No eyes had UDVA three lines worse than the preoperative CDVA.
Nine standard graphs for reporting refractive surgery showing the visual and refractive outcomes for 71 eyes at 30 months after SMILE. SMILE = small incision lenticule extraction; UDVA = uncorrected distance visual acuity; CDVA = corrected distance visual acuity; D = diopters; Postop = postoperative; Preop = preoperative; SEQ = spherical equivalent refraction; TIA = target induced astigmatism; SIA = surgically induced astigmatism
Additionally, no significant difference was found in efficacy index between the moderate and high astigmatism groups (P = .177, Table A). However, the UDVA (logMAR) at 30 months postoperatively had a significant decrease from that at the early stage (P < .05, Table B, available in the online version of this article).
CDVA and UDVA Outcomes
The safety index was 1.15 ± 0.17 for all eyes enrolled in this study. Meanwhile, the safety index of the moderate astigmatism group was higher than that of the high astigmatism group (P = .005, Table A). A total of 99% of the eyes had no change or gain of one line on CDVA (Figure 1C) at 30 months postoperatively. A significant increase of postoperative CDVA at 1, 3, 6, and 30 months postoperatively was observed from the preoperative values (P < .05, Table B).
All surgeries were completed without intraoperative or postoperative complications.
A scatter plot of the attempted versus the 30-month postoperative achieved SE correction is shown in Figure 1D. The difference between the attempted versus the postoperative achieved SE was −0.009 ± 0.013 D at 1 month (P > .999), −0.007 ± 0.019 D at 3 months (P > .999), −0.024 ± 0.022 D at 6 months (P > .999), and −0.099 ± 0.028 D at 30 months (P = .008). At 30 months postoperatively, 96% of eyes were within ±0.50 D and 100% of eyes were within ±1.00 D of the attempted correction (Figure 1E).
As shown in Figure 1F, the SE of 4% of eyes changed more than 0.50 D from 3 to 30 months postoperatively. The mean postoperative SE was −0.01 ± 0.11 D at 1 month, −0.01 ± 0.16 D at 3 months, −0.03 ± 0.19 D at 6 months, and −0.10 ± 0.23 D at 30 months. Only change in SE from 3 to 30 months (P30m-3m = .020) and from 1 to 30 months (P30m-1m = .015) of follow-up were statistically significant. The stability of postoperative sphere refraction and cylinder refraction are shown in Figure A (available in the online version of the article). The postoperative sphere refraction was 0.028 ± 0.015 D at 1 month, 0.028 ± 0.021 D at 3 months, 0.012 ± 0.022 D at 6 months, and −0.016 ± 0.028 D at 30 months (P30m-6m > .999, P30m-3m = .800, P30m-1m = .731, P6m-3m > .999, P6m-1m > .999, P3m-1m > .999). Significant changes were found in the 30-month postoperative cylinder refraction (−0.169 ± 0.030 D) when compared with the early stage cylinder refraction, including at 1 month (−0.074 ± 0.021 D), 3 months (−0.074 ± 0.020 D), and 6 months (−0.077 ± 0.030 D) (P30m-6m = .016, P30m-3m = .024, P30m-1m = .008, P6m-3m > .999, P6m-1m > .999, P3m-1m > .999).
The stability of sphere and cylinder refraction. SEQ = spherical equivalent refraction; D = diopters
Refractive Astigmatism and Vector Analysis
The refractive astigmatism results are shown in Figures 1G–1I and Figure B (available in the online version of this article). At 30 months postoperatively, the percentage of eyes with myopia cylinder refraction less than 0.50 D and less than 0.75 D was 94.4% and 98.6%, respectively. The magnitude of TIA was 2.07 ± 0.69 D. The magnitude of postoperative SIA was 2.13 ± 0.71 D at 1 month (P1m-pre = .018), 2.12 ± 0.72 D at 3 months (P3m-pre = .006), 2.13 ± 0.69 D at 6 months (P6m-pre = .010), and 2.19 ± 0.72 D at 30 months (P30m-pre < .001). The vector distributions of TIA and postoperative SIA are shown in Figure C (available in the online version of the article). The x-coordinate of the centroid of TIA was −0.04 ± 0.49 D, and that of SIA was −0.08 ± 0.51 D at 1 month (P1m-pre = .093), −0.09 ± 0.51 D at 3 months (P3m-pre = .029), −0.07 ± 0.49 D at 6 months (P6m-pre = .087), and −0.09 ± 0.53 D at 30 months (P30m-pre = .585). The y-coordinate of the centroid of TIA was −2.00 ± 0.75 D and that of SIA was −2.04 ± 0.77 D at 1 month (P1m-pre = .033), −2.04 ± 0.78 D at 3 months (P3m-pre = .011), −2.05 ± 0.75 D at 6 months (P6m-pre = .012), and −2.10 ± 0.78 D at 30 months (P30m-pre < .001).
Postoperative cylinder refraction distribution at 1, 3, 6, and 30 months.
Surgically induced astigmatism (SIA) vector distribution at 1, 3, 6, and 30 months. TIA = target induced astigmatism; SIA = surgically induced astigmatism; D = diopters
The centroid coordinates of the cylinder refractive vector were −0.74 ± 2.01 D, 0.33 ± 0.36 D preoperatively, 0.04 ± 0.12 D, 0.05 ± 0.13 D at 1 month, 0.05 ± 0.14 D, 0.04 ± 0.10 D at 3 months, 0.03 ± 0.09 D, 0.06 ± 0.15 D at 6 months, and 0.06 ± 0.21 D, 0.11 ± 0.19 D at 30 months. Although the x-coordinate did not change significantly from preoperatively to postoperatively at all time points, the y-coordinate of the centroid of the cylinder refractive vector had a significant change only from 3 to 30 months postoperatively (P30m-3m = .036). The CI was 1.03 ± 0.07 at 1 month, 1.02 ± 0.05 at 3 months, 1.03 ± 0.08 at 6 months, and 1.06 ± 0.10 at 30 months (P30m-6m = .281, P30m-3m = .010, P30m-1m = .074, P6m-3m > .999, P6m-1m > .999, P3m-1m > .999). The IOS was 0.04 ± 0.09 at 1 month, 0.03 ± 0.07 at 3 months, 0.04 ± 0.10 at 6 months, and 0.09 ± 0.13 at 30 months (P30m-6m = .041, P30m-3m = .005, P30m-1m = .008, P6m-3m > .999, P6m-1m > .999, P3m-1m > .999).
In this study, we share our long-term experience with the correction of moderate to high with-the-rule astigmatism using SMILE under astigmatic axis marked condition. To our best knowledge, this is the first article focusing on the vector analyses of long-term astigmatism outcomes after SMILE.
The method we used to align the astigmatic axis has been described in our previous work.9 A well-known limitation of SMILE is the lack of a pupil-tracking and evaluation system for accurate docking of the eyes.8 The cyclotorsion control during surgery is demanding due to the rotation of an eye resulting from a change in the body position. Some studies reported different methods of marking of the horizontal meridian: slit-lamp–assisted marking with a horizontal slit-beam, bubble-marking method, slit-lamp–assisted marking with a pendulum-attached marker, and nonpendular marker.12,13 In the bubble-marking/pendular-marking/tonometer-marking method similar to the triple centration technique, the eye was marked with a marker that had a water level bubble for horizontal positioning of the instrument.14 After the marker was dipped in a blue inkpad, the limbus of the eye was touched to mark two points of the horizontal meridian and the inferior vertical meridian. The only disadvantage of such a method used was on the limbus, but not on the peripheral cornea. Most of the marking methods gave accurate results, except for the tonometer marker. Second, intraoperative cyclotorsion compensation following suction was performed by gently rotating the cone and aligning the 0° to 180° peripheral corneal marks with the horizontal axis of the laser microscope eye piece. Consequently, the alignment method of marking the axis of cylinder under the sitting position was used, with slit-lamp–assisted marking, due to its convenience and ease of implementation. The axis of 0° to 180° was marked on the peripheral cornea, but not along the limbus. This was because the surgical field of view was limited to the 8-mm diameter of the cornea centered at the corneal apex. The slit was placed horizontal and the marking was performed by making gentle scratches using a thin needle and staining them with a dye. Further study to simplify and improve the accuracy of alignment with the astigmatism axis would be beneficial.
A triple centration technique was proposed to solve both the alignment of astigmatism axis and the localization for the corneal vertex. Under this method, the efficacy at 6 months postoperatively (UDVA of −0.05 ± 0.07 decimal) was inferior to our results (−0.103 ± 0.007 decimal).14 It is possible that greater preoperative cylinder was responsible for the difference between the earlier study14 and this study. Also, in our study, the 30-month postoperative efficacy index was 1.04 ± 0.18, which was slightly higher than 1.00 and implied that most of the eyes had a 30-month postoperative UDVA better than the preoperative CDVA. Moreover, no significant difference was found in the efficacy index when correcting moderate cylinder refraction using SMILE versus correcting high cylinder refraction (Table A). Meanwhile, the long-term efficacy observed in our study was better than the results from other recent studies.15−17 Additionally, in this study, the difference between the attempted versus achieved SE was −0.099 ± 0.028 D at 30 months, mainly due to the regression of nearly −0.10 D in cylinder power. However, in others' research at 5 years of follow-up, a significant regression of −0.375 D was observed, which might be due to the absence of both cylinder axis alignment and added overcorrection of spherical power.18 Overall, the slight regression in cylindrical power was clinically insignificant. At 30 months postoperatively, 96% of eyes were within ±0.50 D and 100% of eyes were within ±1.00 D of the attempted correction. All of these results confirmed the long-term efficacy and predictability of SMILE in correcting moderate to high myopic astigmatism with astigmatic axis marked condition.
Postoperative CDVA increased significantly compared to the preoperative CDVA (Table B). All surgeries were performed successfully. No intraoperative complications were observed during the surgery, such as black areas, unintended posterior plane dissection, or tearing of the lenticule. Thus, it was unlikely that the blue mark on the peripheral cornea affected the laser ablation. Also, no eye had any ectasia after SMILE. Consequently, safety of SMILE in correcting with-the-rule-astigmatism with the astigmatic axis marking method was demonstrated in this study.
Better than the results of Kunert et al2 without astigmatic axis alignment, the magnitude of SIA was slightly higher than the magnitude of TIA in our study, which was comparable to LASIK results.15,16 Further, only the y-coordinate of the centroid changed significantly over time. Consequently, the adjustment of the cylinder refraction nomogram should focus on the y-coordinate of TIA. Meanwhile, significant changes in both CI and IOS at 30 months postoperatively were found when compared with early follow-up time points. Also, the postoperative UDVA decreased at 30 months when compared with at 3 months. This could indicate that when CI exceeds 1.06, the efficacy of SMILE would be affected. Interestingly, the high astigmatism group had a better CI outcome than the moderate astigmatism group. It could be implied that the overcorrection of myopic astigmatism with SMILE is more obvious in the eyes with moderate astigmatism than in the eyes with high astigmatism. This needs to be confirmed with a larger sample size in future studies. As a result, more exploration needs to be done in terms of adjustment of the cylinder refraction nomogram when using SMILE to correct moderate to high with-the-rule astigmatism under astigmatism axis marked condition.
Otherwise, slight overcorrection of cylinder refraction was observed in this study. This is somewhat contrary to the general observation and results of some studies published, where generally an undercorrection of astigmatism correction is observed.5,19 However, in our previous study focusing on another technique for SMILE cyclotorsion compensation, a CI of 1.03 ± 0.30 was observed.8 It could be implied that both the results and the procedures by an optometrist might have an effect on the clinical outcomes after refractive surgery. Thus, the worldwide consensus of optometry in refractive surgery might be recommended. In addition, during the long-term follow-up, only the y-coordinate of the SIA vector had a significant overcorrection compared with that of the TIA vector. Meanwhile, in our research focusing on the long-term clinical results of SMILE for eyes without astigmatism, cylinder refraction of −0.11 ± 0.21 D was observed at 30 months postoperatively (unpublished data). Thus, it could be implied that the eyelid tension would affect the postoperative astigmatic refraction.20 In further research, the nomogram for SMILE might be more proper to take into consideration the vector adjustment on the x- and y-coordinates.
This study has limitations. The comparison of the vector results between laser in situ keratomileusis (LASIK) and SMILE for correcting moderate to high astigmatism was not evaluated in our study. Theoretically, compared with LASIK, SMILE with lack of excimer ablation and 360° flap cut and less induced proliferation and inflammation should lead to better results when correcting moderate to high astigmatism under astigmatic axis alignment condition with the proper nomogram.21 Consequently, further study would focus on the adjustment of the SMILE astigmatic nomogram and on the comparison of the vector outcomes between SMILE and LASIK when correcting moderate to high myopic astigmatism under astigmatic axis marked condition. In future studies, the vector analyses focusing on against-the-rule astigmatism and oblique astigmatism corrected by SMILE should be evaluated.
- Sekundo W, Kunert KS, Blum M. Small incision corneal refractive surgery using the small incision lenticule extraction (SMILE) procedure for the correction of myopia and myopic astigmatism: results of a 6 month prospective study. Br J Ophthalmol. 2011;95(3):335–339. doi:10.1136/bjo.2009.174284 [CrossRef]
- Kunert KS, Russmann C, Blum M, Sluyterman VLG. Vector analysis of myopic astigmatism corrected by femtosecond refractive lenticule extraction. J Cataract Refract Surg. 2013;39(5):759–769. doi:10.1016/j.jcrs.2012.11.033 [CrossRef]
- Yildiz BK, Urdem U, Goksel Ulas M, et al. Correction of myopic astigmatism by small incision lenticule extraction: does laterality matter?Lasers Med Sci. 2019;34(2):311–316. doi:10.1007/s10103-018-2591-9 [CrossRef]
- Ivarsen A, Hjortdal J. Correction of myopic astigmatism with small incision lenticule extraction. J Refract Surg. 2014;30:240–247. doi:10.3928/1081597X-20140320-02 [CrossRef]
- Ganesh S, Brar S, Pawar A. Results of intraoperative manual cyclotorsion compensation for myopic astigmatism in patients undergoing small incision lenticule extraction (SMILE). J Refract Surg. 2017;33:506–512.
- Chernyak DA. Cyclotorsional eye motion occurring between wavefront measurement and refractive surgery. J Cataract Refract Surg. 2004;30(3):633–638. doi:10.1016/j.jcrs.2003.08.022 [CrossRef]
- Chen P, Ye Y, Yu N, Zhang X, Zhuang J, Yu K. Correction of astigmatism with SMILE with axis alignment: 6-month results from 622 eyes. J Refract Surg. 2019;35:138–145.
- Xu J, Liu F, Liu M, et al. Effect of cyclotorsion compensation with a novel technique in small incision lenticule extraction surgery for the correction of myopic astigmatism. J Refract Surg. 2019;35:301–308. doi:10.3928/1081597X-20190402-01 [CrossRef]
- Yang X, Liu F, Liu M, Liu Q, Weng S, Lin H. 15-month visual outcomes and corneal power changes of SMILE in treating high myopia with maximum myopic meridian exceeding 10.00 D. J Refract Surg. 2019;35:31–39.
- Liu Q, Yang X, Lin L, et al. Review on centration, astigmatic axis alignment, pupil size and optical zone in SMILE. Asia Pac J Ophthalmol (Phila). 2019;8(5):385–390. doi:10.1097/01.APO.0000580144.22353.46 [CrossRef]
- Alpins NA, Goggin M. Practical astigmatism analysis for refractive outcomes in cataract and refractive surgery. Surv Ophthalmol. 2004;49(1):109–122. doi:10.1016/j.survophthal.2003.10.010 [CrossRef]
- Popp N, Hirnschall N, Maedel S, Findl O. Evaluation of 4 corneal astigmatic marking methods. J Cataract Refract Surg. 2012;38(12):2094–2099. doi:10.1016/j.jcrs.2012.07.039 [CrossRef]
- Elhofi AH, Helaly HA. Comparison between digital and manual marking for toric intraocular lenses: a randomized trial. Medicine (Baltimore). 2015;94(38):e1618. doi:10.1097/MD.0000000000001618 [CrossRef]
- Jun I, Kang DSY, Reinstein DZ, et al. Clinical outcomes of SMILE with a triple centration technique and corneal wavefront-guided transepithelial PRK in high astigmatism. J Refract Surg. 2018;34:156–163. doi:10.3928/1081597X-20180104-03 [CrossRef]
- Chan TCY, Wang Y, Ng ALK, et al. Vector analysis of high (≥3 diopters) astigmatism correction using small-incision lenticule extraction and laser in situ keratomileusis. J Cataract Refract Surg. 2018;44(7):802–810. doi:10.1016/j.jcrs.2018.04.038 [CrossRef]
- Chan TC, Ng AL, Cheng GP, et al. Vector analysis of astigmatic correction after small-incision lenticule extraction and femtosecond-assisted LASIK for low to moderate myopic astigmatism. Br J Ophthalmol. 2016;100(4):553–559. doi:10.1136/bjophthalmol-2015-307238 [CrossRef]
- Zhang J, Wang Y, Wu W, Xu L, Li X, Dou R. Vector analysis of low to moderate astigmatism with small incision lenticule extraction (SMILE): results of a 1-year follow-up. BMC Ophthalmol. 2015;15(1):8. doi:10.1186/1471-2415-15-8 [CrossRef]
- Blum M, Täubig K, Gruhn C, Sekundo W, Kunert KS. Five-year results of small incision lenticule extraction (ReLEx SMILE). Br J Ophthalmol. 2016;100(9):1192–1195. doi:10.1136/bjophthalmol-2015-306822 [CrossRef]
- Taneri S, Kiessler S, Rost A, et al. Small-incision lenticule extraction for the correction of myopic astigmatism. J Cataract Refract Surg. 2019;45(1):62–71. doi:10.1016/j.jcrs.2018.08.030 [CrossRef]
- Ho RW, Fang PC, Chang CH, Liu YP, Kuo MT. A review of periocular botulinum neurotoxin on the tear film homeostasis and the ocular surface change. Toxins (Basel). 2019;11(2):11. doi:10.3390/toxins11020066 [CrossRef]
- Karmona L, Mimouni M, Vainer I, Sela T, Munzer G, Kaiserman I. Induced de novo astigmatism after hyperopic LASIK versus myopic LASIK surgery in nonastigmatic eyes. Cornea. 2017;36(9):1040–1043. doi:10.1097/ICO.0000000000001253 [CrossRef]
Preoperative Patient Demographics in Eyes Undergoing SMILE
|Characteristic||Mean ± SD (Range)|
|Age (y)||24.37 ± 6.52 (18 to 40)|
|Male sex (%)||43.7%|
|Sphere (D)||−5.14 ± 1.80 (−1.00 to −8.50)|
|Cylinder(D)||−2.07 ± 0.69 (−1.50 to −4.75)|
|Spherical equivalent (D)||−6.18 ± 0.70 (−2.38 to −9.25)|
|CDVA (logMAR)||−0.05 ± 0.05 (0.05 to −0.18)|
|Anterior corneal astigmatism (D)||2.27 ± 0.70 (0.80 to 4.20)|
Preoperative and Postoperative Refractive Outcomes Stratified by Preoperative Cylindera
|Parameter||Moderate Astigmatism||High Astigmatism||P|
| Sphere (D)||−5.55 ± 1.77 (−1.50 to −8.50)||−4.23 ± 1.53 (−1.00 to −7.50)||.003|
| Cylinder (D)||−1.72 ± 0.21 (−1.50 to −2.00)||−2.85 ± 0.75 (−2.25 to −4.75)||< .001|
| SE (D)||−5.93 ± 1.54 (−2.375 to −9.25)||−5.65 ± 1.49 (−2.875 to −8.625)||.081|
| CDVA (logMAR)||−0.05 ± 0.05 (−0.18 to 0.05)||−0.06 ± 0.05 (−0.18 to 0.00)||.211|
|30 months postoperatively|
| Sphere (D)||−0.01 ± 0.24 (−0.50 to 0.75)||−0.02 ± 0.23 (−1.00 to 0.25)||.870|
| Cylinder (D)||−0.20 ± 0.27 (−1.00 to 0.00)||−0.10 ± 0.20 (−0.50 to 0.00)||.142|
| SE (D)||−0.11 ± 0.25 (−0.88 to 0.50)||−0.07 ± 0.22 (−1.00 to 0.00)||.530|
| Efficacy index||1.06 ± 0.19 (0.58 to 1.50)||1.00 ± 0.16 (0.67 to 1.25)||.177|
| Safety index||1.18 ± 0.17 (1.00 to 1.50)||1.06 ± 0.16 (0.80 to 1.25)||.005|
| SE regression||−0.10 ± 0.26 (−0.88 to 0.38)||−0.06 ± 0.25 (−1.00 to 0.25)||.523|
| CI||1.08 ± 0.13 (1.00 to 1.52)||1.02 ± 0.05 (1.00 to 1.19)||.041|
| TIA||1.72 ± 0.21 (1.50 to 2.00)||2.85 ± 0.75 (2.25 to 4.75)||< .001|
| SIA||1.85 ± 0.24 (1.50 to 2.36)||2.91 ± 0.78 (2.25 to 4.91)||< .001|
| AOE||1.97 ± 33.85 (−159.49 to 166.30)||8.17 ± 35.60 (−2.23 to 167.28)||.485|
| MOE||0.13 ± 0.20 (0.00 to 0.78)||0.06 ± 0.14 (0.00 to 0.50)||.147|
| IOS||0.12 ± 0.17 (0.00 to 0.67)||0.04 ± 0.08 (0.00 to 0.22)||.009|
CDVA and UDVA Outcomes
|Visit||CDVA (logMAR)||UDVA (logMAR)|
|Preoperative||−0.052 ± 0.050||–|
|1 month postoperative||0.124 ± 0.064, P1M-PRE < .001||−0.100 ± 0.009|
|3 months postoperative||−0.141 ± 0.047, P3M-PRE < .001, P3M-1M = .311||−0.118 ± 0.007, P3M-1M = .202|
|6 months postoperative||−0.128 ± 0.056, P6M-PRE < .001, P6M-1M > .999, P6M-3M = .704||−0.103 ± 0.007, P6M-3M = .197, P6M-1M > .999|
|30 months postoperative||0.108 ± 0.061, P30M-PRE < .001, P30M-1M > .999, P30M-3M < .001, P30M-6M = .196||−0.064 ± 0.008, P30M-1M = .023, P30M-3M < .001, P30M-6M < .001|