The demand of improvement for visual quality has gained importance in step with the advances in refractive surgery technology. However, laser refractive surgery to eliminate low refractive errors could cause irregular changes of the corneal surface, with the induction of HOAs, which adversely affects the postoperative visual function [22,23,24]. Several studies [25,26,27] have demonstrated that total HOAs, coma and spherical aberration increase significantly after SMILE procedures. For this reason, accurate HOAs measurement is essential for evaluating the potential impact of surgery in visual quality. Previous studies [5,6,7] have already reported good to excellent intrasession repeatability and reproducibility of anterior segment measurements obtained with the Sirius system in healthy, post-refractive surgery (PRK/LASIK) and keratoconus patients. Our study enrolled 75 eyes that had undergone laser corneal refractive surgery with another type of technique, SMILE surgery. To explore the precision of corneal morphology measurements in more detail and comprehensively, this study was designed to evaluate the aberrations of the anterior and posterior surface as well as the total cornea with Sirius Scheimpflug-Placido topographer.
In this study, the repeatability of anterior corneal aberrations was excellent, with ICCs for all types of aberrations being more than 0.931 and Sw values being lower than 0.07 μm, except for trefoil. In any case, the level of repeatability for trefoil components was also acceptable, with ICC of 0.768 for the first operator and 0.763 for the second operator. Bayhan et al.  found the similar results with the Sirius device, obtaining good repeatability for anterior corneal aberrations in normal eyes, with ICCs ranging from 0.678 to 0.976 for astigmatism II Z (4, ± 2), trefoil Z (3, ± 3), coma Z (3, ± 1), spherical aberration Z (4, 0), higher-order RMS and total RMS. The only difference was that Sw values of Zernike coefficients corresponding to astigmatism II were lower (0.02 μm) in our study. As with previous studies [13, 14, 28] assessing the consistency of corneal spherical aberration measurements in healthy eyes, the repeatability associated to this Zernike coefficient was higher than that corresponding to other aberrometric components. The mean value of corneal high-order RMS was 0.63 ± 0.18 μm here, which was a higher magnitude than that reported in a previous study for healthy eyes (0.41 ± 0.06 μm) using the same Scheimpflug-Placido topographer . As expected according to previous literature, HOAs of the anterior corneal surface are increased after SMILE.
The difference in refractive index between the cornea and the aqueous humor is approximately 10% of the difference of indexes between cornea and air. It is, therefore, likely that the anterior corneal surface makes a significant contribution to ocular aberrations [29,30,31]. Because of this, it was expected that ICCs of total corneal aberrations were similar to those corresponding to the anterior surface (Table 3). The Sw of spherical aberration was the lowest (0.02 μm) in our study for both anterior and total corneal calculations. Then, it can be considered as an optical effective reference, especially for those surgeons that implant aspheric IOLs . The trefoil Z (3, ± 3) components showed the worst ICC value (0.722) and a high Sw (0.14 μm). Similar results were reported by Aramberri et al.  and Cervino et al. , who investigated the repeatability and reproducibility of the Scheimpflug imaging-based Pentacam and Galilei systems in healthy eyes. They found moderate to high precision in coma Z (3, ± 1) and spherical aberration Z (4, 0) coefficients (ICC = 0.74–0.915), with the worst results for trefoil Z (3, ± 3) (ICCs of 0.478 and 0.552). Cervino et al.  indicated that the reason for this variability might be the incomplete eye opening during the measurement procedure, causing the outliers of superior periphery, which especially affected the 90-degree coma coefficient and both trefoil coefficients. Our results also demonstrated that the total RMS had the highest ICC value associated (0.979). The mean total corneal high-order RMS obtained with the Scheimpflug-Placido device was 0.63 ± 0.18 μm in our study, a value close to that reported by Ventura et al.  (0.63 ± 0.77 μm) using Placido topographer and lower than that obtained (0.93 ± 0.37 μm) using the dual Scheimpflug device in post-refractive surgery (PRK/LASIK) eyes.
Compared to the previous studies in healthy eyes [17, 18], our results showed better repeatability. The increase within acceptable levels of aberrations and irregularity of the cornea after the laser refractive surgery procedure may account for this better repeatability. These trends were in agreement with other studies [13, 14, 34], in which the results also showed better repeatability in irregular corneas compared to healthy eyes. In one study performing the repeatability analysis using keratoconus eyes, the authors concluded that the magnitude of HOAs were higher and could be clearly defined, being this a possible cause of more repeatable and useful measurements for the disease classification .
A previous study by Li et al.  indicated that the posterior corneal surface played an important role in compensating for the spherical aberration of the anterior surface of the cornea. There were few reports on repeatability of posterior corneal aberrometric measurements to this date [19, 34]. None of them have enrolled patients after SMILE. In our study, the Sw corresponding to each aberration of the posterior corneal surface was below 0.03 μm. The ICCs ranged from 0.432 to 0.957 for the first operator and from 0.305 to 0.953 for the second operator. Good precision values were only found for total RMS, coma Z (3, ± 1) and spherical aberration Z (4, 0). However, for high-order RMS, trefoil Z (3, ± 3) and astigmatism II Z (4, ± 2), more limited ICCs were obtained. In general, the central coefficients of the Zernike pyramidal diagram showed better intraobserver repeatability. The ICCs for coma (0.892–0.927) were better than those corresponding to trefoil (0.305–0.432) in the 3rd order and ICCs for spherical aberration (0.810–0.853) were better than those corresponding to astigmatism II (0.585–0.634) in the 4th order. Similar outcomes were also presented in former studies. In a study conducted by Bayhan et al. , the ICCs for posterior corneal coma (0.824) and spherical aberration (0.822) were better than those corresponding to trefoil (0.802) and astigmatism II (0.691) in healthy eyes. Similar results were also found in the study by Piñero et al. , in which the repeatability of Zernike coefficients was better for the aberrometric defects of the center of the Zernike pyramid than for those corresponding to the periphery of the Zernike pyramid at the 3rd and 4th order when using the Pentacam Scheimpflug imaging system in normal eyes (ICC > 0.943 for coma and spherical aberration, ICC < 0.887 for astigmatism II and trefoil). De Jong et al.  also reported that the Galilei G2 and Pentacam HR instruments provided similar and good repeatability in measuring the posterior corneal shape, except for oblique astigmatism and the two trefoil terms in the healthy eye. Furthermore, these results suggested that the ICCs of the Zernike coefficients describing the posterior corneal aberrations were lower compared with their anterior counterparts when comparing with the anterior corneal aberrations in our current study. Although the Sw and TRT values were lower for posterior corneal aberrometric data than those corresponding to the anterior surface, differences may be of little clinical relevance due to the small magnitude of such variability. In any case, one reason for this higher variability of posterior corneal aberrometric measurements may be in relation to the significantly lower magnitude of these aberrations compared to the anterior corneal surface, representing any small change that resulting in a significant level of variability. Additionally, the reason for the poor repeatability of posterior corneal aberrometric data might be the inadequate characterization of posterior corneal aberrations with Zernike polynomial expansion due to subtle movements during scanning . More studies are still needed to overcome these potential limitations.
Regarding the reproducibility results of the corneal aberrometric components analyzed by both observers in our study, all Sw values for all types of aberrations were equal to or below 0.04 μm. The reliability analysis showed ICCs ranging from 0.834 to 0.989 for anterior and total corneal aberrations, with the lowest ICC value for the trefoil component Z (3, ± 3), suggesting excellent interobserver reproducibility except for trefoil. Concerning posterior corneal aberrometric data, ICCs of total RMS and coma Z (3, ± 1) were 0.976 and 0.967, also indicating excellent interobserver reproducibility. These results were better than those from a previous study conducted by Sideroudi et al.  evaluating the reproducibility of Pentacam-derived posterior aberrations measurements in both normal and ectatic corneas, with coma, coma-like and HOA RMS showing acceptable reliability. Moreover, the excellent interobserver reproducibility observed in our series was consistent with that previously reported for other anterior segment parameters [7, 36]. Hernandez et al.  analyzed the cornea and anterior segment using the Sirius system, obtaining ICC values of more than 0.9 for all of the measured variables. Bao et al.  who evaluated the reproducibility of posterior corneal surface measurement using the Sirius system also reported high reproducibility for posterior corneal surface measurement in normal eyes. To our knowledge, this is the first report that evaluates the reproducibility of Sirius measurements of corneal aberrations in post-SMILE eyes.
It should be noted that there are some factors that may account for the significantly decreased precision, such as misalignment or movement during scanning, short acquisition time, pupil translation , and tear film instability. Current results indicate that the repeatability and reproducibility of trefoil were the worst. The reason might be incomplete eye opening during the measurement procedure, which limited acquisition of the superior cornea. Certain instruments with different principles show a similar phenomenon. Previous studies found similar results with the Scheimpflug imaging-based topographer , new pyramid wavefront sensor , and Hartmann-Shack aberrometer . We also acknowledge that patients included in this study were young and well-coordinated, with a successful recovery after SMILE without complications. Therefore, further research may include patients with a wider age range, and evaluate which device can more precisely determine the aberrations of the posterior surface of the cornea.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.