Lower Femtosecond Laser Energy Levels Lead to Better Visual Recovery After Small-incision Lenticule Extraction for Myopia

This prospective randomized comparative study was approved by the Institutional Review Board, Yonsei University College of Medicine, Seoul, South Korea (IRB No. 4-2016-0840). The study adhered to the tenets of the Declaration of Helsinki and followed good clinical practices. All patients provided informed consent after a detailed explanation of the possible risks and benefits of the study. The right eye of each patient was included in the study unless contraindicated, in which case the left eye was used. The patients were randomized to one of two groups (L-SMILE or C-SMILE) by means of permuted block-randomization with the Clinstat software (Available from: http://www-users.york.ac.uk/~mb55/soft/soft.htm). Thereafter, the patients were randomly allocated a laser energy level within their SMILE group by means of minimization with the Minim software (Available from: http://www.users.york.ac.uk/~mb55/guide/minim.htm) due to the small sample size (< 200).

Patients were recruited from November 2016 to December 2017. All patients underwent a baseline preoperative assessment including anterior and posterior segment examinations. Inclusion criteria were: corneal thickness of more than 500 μm, manifest refractive sphere of -3.00 to -6.00 diopters (D), manifest refractive cylinder less than 6.00 D, stable refractive error with less than a 0.50 D change in sphere and cylinder in the previous year, corrected distance visual acuity (CDVA) of 20/20 or better in both eyes, and age of 20 years or older. Exclusion criteria were: severe ocular surface disease, any corneal disease, cataract, glaucoma, macular disease, or previous history of intraocular or corneal surgery. Patients with suspicion of keratoconus on corneal topography were also excluded.

We reviewed thoroughly literatures which included 239 online-available papers about SMILE, and, of them, we selected 116 original articles on clinical outcome or studies involving human lenticules after SMILE surgery and identified all the energy levels used in each article, as possible. Finally, we found almost all previous researches provided surgical results by using a 200 or 500 kHz VisuMax (Carl Zeiss Meditec AG, Jena, Germany) with energy levels of 115 to 190 nJ except only three studies: two studies by same researchers obscured the laser energy level (approximately 110 nJ) used in SMILE, however the other latest study provided definitely their laser energy 100 nJ (Figure 1 and supplemental table 1). There were 5 studies on SMILE using 115 nJ of laser energy in the literature.

Based on the review of literature, we regarded SMILE using energy levels of 115 nJ or higher as 'conventional' energy-SMILE (C-SMILE). Since 500kHz VisuMax femtosecond laser used in this study had a threshold photodisruption energy level of 100 nJ, lowest energy level of L-SMILE was set at 100 nJ. Because the spot distance must be changed by approximately 1 μm when laser energy changed by 50 nJ, we could set the maximum laser energy to 150 (100 + 50) nJ as highest one of C-SMILE while keeping the spot distance constant. Finally, we established two groups divided by different laser energies: L-SMILE group using lower energy levels of 100, 105, and 110 nJ, and C-SMILE group using 'conventional', not higher, energy levels of 115 to 150 nJ.

Based on our preliminary results (n=6 per group) showing the significant mean difference of postoperative visual acuity between C-SMILE and L-SMILE, a priori power analysis was performed. Group sample sizes of 54 and 81 achieved 81% power to detect a difference of - 0.1 between the null hypothesis that both group means are 0.0 and the alternative hypothesis that the mean of group 2 is 0.1 with estimated group standard deviations of 0.1 and 0.1 and with a significance level (alpha) of 0.05 using a two-sided two-sample t-test. Therefore, we confirmed that sample size of L-SMILE is 60 and one of C-SMILE is 90 as considering 10 % reduction.