Small Incision Lenticule Extraction (SMILE) Versus Laser Assisted Stromal In Situ Keratomileusis (LASIK) for Astigmatism Corrections: A Systematic Review and Meta-analysis.

PURPOSE: To compare small incision lenticule extraction (SMILE) and laser assisted stromal in situ keratomileusis (LASIK) for astigmatism correction.

DESIGN

Systematic review and meta-analysis

METHODS

We reviewed published studies comparing outcomes after LASIK and SMILE for astigmatism correction by querying PubMed, EMBASE, Cochrane, and Web of Science, with a cut-off date of September 3, 2022. We also compared the changes in visual acuity, refraction, and high-order aberrations between the surgeries. Astigmatism correction outcomes in the low-to-moderate group (less than or equal to -2.00 D) and high group (greater than -2.00 D) were evaluated using vector analysis. The Cochrane risk of bias tool in RevMan software was used for randomized studies (RCT), and Risk Of Bias In Nonrandomized Studies - of Interventions (ROBINS-I) was used for the nonrandomized studies (NRSs).

RESULTS

There were 17 studies (5 randomized studies and 12 cohort studies), including 1,985 eyes. A statistically significant difference was found in the correction index (mean difference [MD] = -0.02, 95% confidence interval [CI] = -0.04 to -0.01, P =0.01), although there was no significant difference in the index of success (MD = 0.01, 95% CI = -0.03 to 0.05, P =0.51), different vector (MD = 0.07, 95% CI = 0.00 to 0.13, P =0.04), and angle of error (MD = 0.56, 95% CI = -0.34 to 1.45, P =0.22) between SMILE and LASIK. However, for low-to-moderate astigmatism correction, SMILE exhibited a smaller correction index (MD = -0.08, 95% CI= -0.13 to -0.02, P =0.008) and a larger difference vector (MD = 0.18, 95% CI = 0.09 to 0.27, P <0.0001) than LASIK. There was no significant difference between the different procedures in visual acuity and refraction (spherical equivalent: MD = -0.04, 95% CI = -0.08 to 0.01, P =0.15) or high-order aberration (MD = -0.01, 95% CI = -0.07 to 0.04, P =0.67), except spherical aberration (MD = -0.12, 95% CI = -0.23 to -0.01, P =0.04). The risk of bias was moderate in most studies because of poor reporting of several bias domains for RCTs, and because of confounding and selective outcome reporting for NRSs.

CONCLUSIONS

When used to treat severe astigmatism, both SMILE and LASIK provide effective and predictable results and generally have equivalent outcomes. However, evidence reveals a tendency toward undercorrection in the SMILE groups for astigmatism correction. In addition, LASIK has a greater probability of causing postoperative spherical aberration.