To date, studies on SMILE have not determined the standard size or location for CSCs. Most studies13–15 have determined a 90° direction for CSCs of 1.95 mm length. For patients with superior corneal pannus or extremely deep sockets, the surgeon may adjust the CSC direction to 120° to avoid hemorrhage or ease the procedure. Dan et al. reported the design of two CSCs in the superior nasal and superior temporal directions in SMILE surgery for low myopia16. We performed this research to investigate the effects of the two CSC types on astigmatism correction and visual performance.
In this study, all eyes in both the two groups had a postoperative UDVA ≥ 20/20 at the 3-month visit. The safety and efficacy index were 1.01 ± 0.06 and 1.00 ± 0.08 in group 1 and 1.04 ± 0.14 and 1.02 ± 0.14 in group 2, respectively. The safety index was close to that reported in previous studies17,18, which reported a median safety index of 1 and 1.1, respectively, at the 3-month follow-up after SMILE surgery. Another study19 reported an efficacy index of 1.18 ± 0.21, with a mean preoperative astigmatism of -0.68 ± 0.46 D, which was − 0.82. ±0.52 D in the current study. There were no differences in UDVA, CDVA, SE, safety, or efficacy index between the two groups. Our findings showed no difference in refraction outcomes between the two CSC groups.
Vector analysis revealed an average CI of 1.028 and 1.043 in groups 1 and 2, respectively (P = 0.971), which indicated a slight overcorrection of astigmatism in both groups. The average CI in a previous study20 showed a similar value of 1.05 using SMILE surgery and 0.95 using laser epithelial keratomileusis. Although a lower IOS was found in group 1 (0.365 vs. 0.548, P = 0.09), the difference was not significant, and the difference in ME between the two groups was not statistically significant (P = 0.447), indicating no difference in residual astigmatism between the two groups. Chan et al. also concluded that there was no difference in astigmatism correction between two opening incisions21. In addition to the magnitude, it is important to correct astigmatism with a precise axis. Alpins noted that the deficiency of the correction effect is 13.4% when treatment is skewed by 15° and is almost 50% when treatment is skewed by 30°22. Xia et al. conducted that 77% of the eyes had AE values between − 15° and 15°23. In this study, 85.71% and 83.33% of the eyes in group 1 and group 2, separately, had AE values between − 15 and 15°, which showed a slightly more accurate correction of astigmatism in group 1; however, no statistical significance was found between the two groups. The AE in group 1 was − 0.630 ± 9.64 and in group 2 was − 2.710 ± 13.99; the less favorable range (9.64° and 13.94°) occurred in both groups, suggesting some variable factors, such as healing or alignment, which require further investigation. No significant difference was detected in other variables by using vector analysis.
Corneal anterior surface aberrations account for 80% of aberrations in the eye and are the main factors influencing visual quality24,25. In the Zernike HOA chart, SA and coma are the most common parameters used to analyze the impact on visual quality26. Corneal anterior astigmatism and aberration variability were used in the present study. The change in corneal anterior astigmatism was greater in group 1 (0.588) than in group 2 (0.520), but the difference was not statistically significant. Corneal aberrations, including RMS HOA, SA, HC, and VC increased after SMILE surgery. Several studies27–29 have reported increased HOA in both SMILE and laser-assisted in situ keratomileusis (LASIK). The induced VC was considerably higher than the induced HC in both groups (P = 0.447), and Chalita et al. reported that increased HC after LASIK surgery was correlated with double vision but not with increased VC30. No significant difference was discovered in the induced RMS HOA, SA, HC, or VC between the two groups.
A limitation of our investigation is that the 3-month follow-up may be not adequate to evaluate astigmatic correction and vision quality in the two different directions of CSCs. Studies with longer follow-up durations and more directions of CSCs are needed to confirm the results of the present study.