With the increasing prevalence of myopia in recent years, refractive surgeries have become popular for myopia correction.1,2 However, the occurrence of iatrogenic ectasia following refractive surgery underscores the importance of pre-operative prediction and long-term ectasia risk monitoring.3 Small-incision lenticule extraction (SMILE) and laser-assisted subepithelial keratomileusis (LASEK) are stromal flap-free procedures, showing a lower association with corneal ectasia compared to laser in situ keratomileusis (LASIK).4 This reduced risk may be attributed to factors such as the absence of a corneal flap and considerations like residual stromal bed thickness.5 Consequently, eyes post-SMILE and post-LASEK serve as valuable models for comprehending the impact of tissue removal on the cornea.

Corneal biomechanical properties, crucial for predicting keratoectasia pre- and post-refractive surgery, are traditionally assessed using devices like the ocular response analyzer (ORA) and corneal visualization Scheimpflug technology (Corvis ST).6,7 While these studies consistently demonstrate overall biomechanical weakening post-operatively, there remains ongoing debate regarding which surgical techniques (e.g., flap-based vs. cap-based procedures) better preserve corneal integrity. This controversy stems from fundamental differences in how these techniques modify corneal structure—particularly regarding anterior stromal ablation depth versus preservation of the stronger anterior lamellae.8–10 However, these deformation-based methods are susceptible to confounding factors such as central corneal thickness (CCT) and intraocular pressure (IOP), complicating their interpretation in post-surgical or thin corneas when comparing with normal and pre-operative results.5,9,11 The newly developed Corvis ST biomechanical index-laser vision correction (CBI-LVC) metric by Corvis ST, associated with post-LVC, lacks validated indices for diagnosing post-LVC ectasia at various stages, and its stability requires further long-term research.11

In contrast, Brillouin microscopy offers a transformative approach as a non-contact, all-optical technique that probes intrinsic biomechanical properties via the Brillouin frequency shift and modulus (BM).12 By quantifying the cornea’s longitudinal modulus at the molecular level—a property dictated by collagen organization and hydration—it is less affected by CCT and IOP.13–16,33 This technique has found application in evaluating corneal diseases and assessing of outcomes in corneal refractive and cross-linking surgeries, with capability to distinguishes depth-dependent variations within different corneal layers, from the epithelium to the endothelial layer.17,18 While showing promise in these areas, its capacity to identify post-operative biomechanical changes in refractive surgery, especially those with subtle effects, and the factors influencing such changes, remains unknown.19–22 Consequently, Brillouin microscopy is suitable for both pre- and post-operative measurements of corneal biomechanical properties.

This study’s primary objective is to discern differences in corneal biomechanics between normal corneas and those that underwent flapless corneal refractive surgeries, specifically SMILE and LASEK, using Brillouin microscopy. This study aims to contribute to a more comprehensive understanding of the disparities in corneal biomechanics between normal corneas and those that have undergone specific corneal refractive surgeries.

In this study, through cross-sectional and within-subjects comparison analyses, we confirmed that Central, Mean, and Max BM were not significantly influenced by SMILE or LASEK. However, Min BM decreased and Max-Min BM increased after both procedures. These findings led us to hypothesize that while the two stromal flap-free procedures not significantly impact the overall corneal biomechanical characteristics, they do result in a reduction in the stiffness of the cornea's weakest points. This was quite discrepant with previous contact-based researches, which suggested that the entire corneal deformation was more observed after corneal ablation induced by refractive surgeries, as represented by various Corvis ST and ORA metrics.10,27 However, it worth noting that entire corneal thickness and residual stromal bed are predominant determinant of stiffness decreases after surgeries.8 Still, our results align with previous research on subjects undergoing LASIK and photorefractive keratectomy (PRK), which observed a similar increase in Max-Min BM and Std BM, and a decrease in Min BM.19 Therefore, we posit that refractive surgery may induce regional lesions and focal decreases in corneal biomechanical properties while maintaining consistent stiffness across the entire cornea.

Our findings collectively illustrate that Brillouin microscopy parameters are largely independent of IOP, are comparable to the patient's preoperative measurements, and can serve as a valuable supplement or standalone tool for monitoring post-operative corneal biomechanical changes. However, attention should be paid to the influence of corneal curvature, CCT, and follow-up time. Overall, the mean BM and maximum BM emerged as the most stable indicators and were unaffected by the other studied factors.

As we have emphasized, Brillouin measurements are less influenced by CCT and IOP due to the fundamental principles of Brillouin light scattering. The underlying Brillouin light scattering process involves interactions between incident photons and thermodynamic phonons within the corneal tissue. This interaction is primarily governed by the local microstructure, particularly the relative composition of extracellular matrix (fluid phase) and protein content (solid phase), as well as the thermodynamic effect.16,28 More hydrated, thicker corneas (e.g., one hour after wakeup) exhibit weaker Brillouin biomechanics (∼25 MHz Brillouin frequency shift, equivalent to 0.03 GPa BM).23 Moreover, increased collagen fibril density (solid phase content) enhances Brillouin biomechanical properties, rather than the entire cornea.13 This approach fundamentally differs from conventional deformation-based techniques, which rely on: the application of external stress (e.g., air-puff or contact), measurement of the resulting corneal strain, and interpretation based on the balance between external forces and intraocular pressure.

To investigate the factors influencing Brillouin metrics in assessing post-operative corneal biomechanical changes and guiding their clinical application, we performed rigorous variable selection and multiple linear regression analyses. In post-SMILE corneas, the Central BM was negatively associated with CCT (P = 0.028). Generally, a thicker cornea is associated with better biomechanical properties (positive correlation). However, in this study, we report a negative correlation, which may be attributed to the differences in pre-operative corneal status. Specifically, we hypothesize that for higher myopic eyes undergoing SMILE surgery, these eyes exhibit both worse biomechanical pre- and post-outcomes due to corneal structure, greater tissue ablation and lower residual CCT post-operatively.29,30 For post-LASEK corneas, where the stromal layer retains integrity, the Max-Min BM showed a positive correlation with Kmean. This indicates that as the post-LASEK cornea becomes steeper (usually with less corneal ablation), the Max-Min BM tends to increase. Considering that LASEK reduces Min BM and that ablation might induce spatial discrepancy within the cornea compared with normal eyes, our results suggest that the steeper the cornea, the more likely it is to decrease Min BM (coefficient=−0.018, P = 0.082) and consequently increase Max-Min BM.

In the within-subjects comparison analysis, ΔCentral BM was negatively related to the time after SMILE surgery. Our findings suggest that the examination time after surgery can influence the Central BM in post-SMILE corneas, which was not further estimated in previous research.19 Our analysis suggests negative correlation between ΔCentral BM and postoperative time in SMILE cases may reflect both hydration and structure. As previously demonstrated, Brillouin measurements are sensitive to corneal hydration states.23 Individual variability in healing responses may account for the scatter in our data, even one month for LASIK patients was enough for recovery.24 Previous research has demonstrated that refractive surgery may influence both the morphology and function of corneal cells, as well as ocular surface status, three months post-operatively.31 Additionally, the stromal bed thickness has been shown to remain stable after one month following SMILE surgery.32 Based on these findings, we hypothesize that this correlation is attributable to both corneal hydration and structural effects.

The limitations of our study should be acknowledged: (1) The lack of baseline refractive error measurements and surgical correction amounts represents an important limitation, as these factors may influence post-operative corneal biomechanics. However, it's noteworthy that Brillouin microscopy measurements are less affected by residual corneal thickness variations. This characteristic permitted our focus on evaluating intrinsic post-operative biomechanical properties. (2) Recent research indicates that PRK has a smaller impact on corneal biomechanics than SMILE, while LASIK, a prominent refractive surgery procedure, was not considered in this study. This suggests the need for further development of corneal biomechanical studies using Brillouin microscopy. (3) The relatively small sample size of LASEK patients might limit the robustness of our conclusions. Nonetheless, the observed statistical significance suggest that the findings are robust and warrant validation in larger, prospective cohorts. A longer follow-up period is advisable to observe the long-term effects of Brillouin metric changes induced by different surgeries. (4) The absence of biomechanical parameters from other instruments necessitates further investigation into the consistency of Brillouin microscopy in measuring post-operative corneal biomechanical parameters. Therefore, future studies should address this issue. (5) Although the gender distribution between the two surgical groups did not differ significantly, the observed difference (49.37 % vs. 29.17 % female) raises the possibility of gender-related selection bias or physiological differences influencing surgical preference or outcomes. Future studies with larger sample sizes are warranted to explore this potential association.

Our study confirmed that SMILE and LASEK may induce a localized reduction in corneal biomechanical properties, as observed by Brillouin microscopy, while maintaining overall corneal biomechanics. When utilizing Brillouin microscopy for post-operative corneal biomechanics in a clinical setting, consideration of corneal tomographic and topographic parameters is vital.

Data are available on reasonable request.

Patient consent for publication

The study adhered to the tenets of the Declaration of Helsinki. The protocols were approved by the Ethics Committee of the Fudan University EENT Hospital Review Board (No.2020530).