Corneal Biomechanical Properties Demonstrate Anisotropy and Correlate With Axial Length in Myopic Eyes

Abstract

Purpose: The purpose of this study was to investigate the ex vivo and in vivo biomechanical characteristic of cornea in myopic eyes.

Methods: Fifty-one corneal stromal lenticules were obtained from myopic eyes during the SMILE procedure and were tested by a biaxial tensile system within 24 hours postoperatively. The material properties of the lenticules were described using stress-strain curves and were compared among axial length (AL) <26 mm and AL ≥ 26 mm group. Pre-operative stress-strain index (SSI) parameters were used to evaluate the biomechanical properties of the cornea in vivo.

Results: Compared with AL < 26 mm, the tangent modulus significantly decreased in horizontal and vertical directions when AL ≥ 26 mm (P < 0.05); SSI also significantly decreased when AL ≥ 26 mm (P < 0.05). Anisotropic parameter is positively correlated with AL (r = 0.307, P < 0.05). Compared with AL < 26 mm, anisotropic parameter significantly increased when AL ≥ 26 mm (P < 0.05). SSI was negatively correlated with AL (r = −0.380, P < 0.05) in the AL < 26 mm group but not in the AL ≥ 26 mm group (P > 0.05). Compared with 26 mm ≤ AL < 27 mm group, the tangent modulus significantly decreased in the horizontal direction (P < 0.05) but not in the vertical direction when 27 mm ≤ AL < 28 mm (P > 0.05).

Conclusions: The biomechanical properties of cornea decreased with the increase of AL. Tangent modulus significantly decreased in the horizontal direction compared with vertical direction. AL should be taken into account during calculation of corneal biomechanical parameters in order to improve validity.

Myopia is one of the common global public health problems with prevalence expected to reach half the world's population by 2050.¹ Myopia, especially high myopia, often have changes in numerous eyeball components, including sclera, choroid, cornea, and retina, and lead to an increased incidence of various sight-threatening complications, such as choroidal neovascularization, retinal detachment, and glaucoma.^2–4 The biomechanical properties of the cornea play an important role in maintaining the front surface of the eyeball. In addition, a better understanding of corneal biomechanical properties can help to explore the mechanism of myopia development, better diagnosis and staging of various corneal diseases, and optimization of corneal refractive surgery.^5,6 However, the knowledge of corneal biomechanics in myopic eyes is limited. 

There are many factors that affect the degree of myopia, such as the refractive power of the cornea, axial lengths (ALs), refractive index of the lenticules, and other components of the refractive system. The development of myopia is usually thought to be related to progressive axial elongation.⁷ The increase in AL attribute to the change of the biomechanical properties of the ocular wall, and myopic eyes have been suggested to show lower level of stiffness than emmetropic ones do.^7–9 It has been reported that axial elongation is associated with the flattening and thinning of the cornea and can lead to changes in corneal biomechanical properties.^10,11 Animal studies also have shown that AL changes as well as the shape of the anterior cornea during the process of myopia modeling.^12,13 The development and application of in vivo biomechanical devices made it possible to perform corneal biomechanical examinations in a noninvasive manner. In recent years, researchers have investigated the relationship between in vivo corneal biomechanics and AL.^14–19 

Corneal Visualization Scheimpflug Technology (Corvis ST) is one of the most common noninvasive measures for corneal biomechanics in vivo by recording the whole deformation process of the cornea under a specific air puff.¹⁵ The Corvis ST-based stress-strain index (SSI) is a new index of corneal stiffness obtained using a numerical simulation of model eyes and finite element analysis, and the superiority of which is that it is not affected by intraocular pressure (IOP) and central corneal thickness (CCT).^7,20 Several previous studies reported conflicting results between SSI and AL or refractive error.^7,21−23 Until now, there are no clear studies to confirm and explicitly state the rationale that how axial elongation of the eyes lead to changes in corneal biomechanics and the relationship between them. 

In this study, Corvis ST was measured for the in vivo corneal biomechanics in myopia participants, and stress-strain relationship of corneal stromal lenticules through biaxial mechanical tensile tests were conducted to investigate the anisotropic characteristic of the cornea. The purpose of this study was to investigate both ex vivo and in vivo biomechanical characteristic changes of cornea in myopic eyes and analysis factors that affects the results of in vivo biomechanical parameters. 

Materials and Methods

Specimen

This study adhered to the tenets of the Declaration of Helsinki and was approved by the Ethics Committee of the Eye and ENT Hospital of Fudan University (Project ID: 2021118-1). Written informed consent was obtained from all participants before enrolment. Cornea stroma lenticules extracted from the SMILE procedure were used as cornea samples to investigate the biomechanical behavior. In the SMILE procedure, a 500 kHz VisuMax femtosecond laser system (Carl Zeiss Meditec, Jena, Germany) was used with a pulse energy of 130 nJ. All SMILE surgeries were performed by the same experienced surgeon (author X.T. Zhou). The inclusion criteria of our study were as follows: (1) the surgical corrected astigmatism of the patient was less than or equal to 0.25 cylinder and (2) the lenticule of the patient was extracted and preserved intact. Only one eye was randomly selected if both eyes met the inclusion criteria. After extraction from the cornea, the lenticules were marked in the 12 o'clock direction and were immediately preserved in corneal storage media (Optisol GS; Bausch Lomb, Irvine, CA, USA) below 4°C. The lenticules were divided into two AL groups (A = AL < 26 mm and B = AL ≥ 26 mm). 

AL Measurement

The eye AL was measured by a commercial optical coherence biometry (IOLMaster; Carl Zeiss Meditec AG, Jena, Germany). Five measurements were continuously obtained from each eye at a sitting position, and the mean value of the five measurements was obtained for calculation. 

SSI Measurement

Corneal biomechanical assessment was performed using Corvis ST (OCULUS Optikgeräte GmbH). Corvis ST is a dynamic Scheimpflug analyzer, which can capture the corneal deformation process caused by an air puff. The cornea was recorded at 4330 images per second by using a built-in highspeed camera. SSI was recorded, which was established to eliminate the interference of IOP and corneal geometry and to estimate the stiffness of a material that differs from the stiffness parameter.²⁰ Corneal biomechanical measurements were performed by experienced examiners under the same lighting conditions. Only measurements with an “OK” quality index were included in the analysis. The parameters from Corvis ST were obtained for each eye. 

Biaxial Mechanical Testing

Biaxial tensile tests were performed on cornea stromal lenticule samples within 24 hours postoperatively and submerged into lactate buffered Ringer's solution bath at 37°C using a CellScale BioTester (BioTester 5000; CellScale, Waterloo, Canada) equipped with 5N loadcells. Test axes were aligned with horizontal and vertical directions of the lenticule, and specimens were attached using BioRakes. The central 4 mm × 4 mm square areas of the specimens were used for testing (Fig. 1). In order to make stability of force-displacement curves, specimens were applied equi-biaxial pre-cycle test to 120 micrometers (3% strain range) for 10 times. The stretching rates were close to 6 micrometers per second (strain rate = 9%/minute) to ensure that the specimens were subjected to quasi-static loading. The stress-strain curves in the 0.06 strain range were used to analyze the biomechanical properties of the specimens. The pre-stretching and formal experiments were carried out with displacement control protocol, and the stretching rates were close to 6 micrometers per second. The horizontal and vertical forces and displacements are recorded for calculating stress and strain. The central thickness of the lenticules were given by the SMILE procedure data. 

Bathing solution will affect the state of the corneal stroma, causing edema and increased thickness.²⁴ Considering the experimental cost, we used Ringer's lactate solution instead of Optisol preservation solution during the tensile test, it inevitably sacrificed some accuracy of the results. Most reported tensile tests were comparative experiments. We used the same processing method for all lenticule samples and set the same experimental conditions focused on the different mechanical behaviors between groups. 

Biomechanical Property Analysis

First Piola-Kirchhoff stress (engineering stress) was used, P_h = F_h/(t × w_h) and Pv = Fv/(t × w_v), where t and w_h,v were the initial thickness and width, respectively. F_h and F_v are the recorded horizontal and vertical forces, respectively. The strain was calculated by elongation (Δl/l_h or Δl/l_v), where l_h and l_v were the horizontal and vertical initial sizes of the boundary of the initial position of the rakes in formal tests. Where Δl was the change in displacement in that direction. The tangent modulus at 0.03 strain was also calculated to evaluate the mechanical properties of the cornea. In SSI measurements, the cornea was subjected to air puff loading. In this case, the deformation of the cornea obviously exceeds the physiological range. Therefore, we assumed that 3% strain was the physiological maximum strain²⁵ and used it to calculate tangent modulus and compared with the SSI value. When the strain is 0.03, the change of the corneal tangent modulus can be better to predict the mechanical properties of the cornea which with high IOP and axial growth occur. 

Assessment of Corneal Anisotropic Character

Horizontal and vertical stretches at 0.6 MPa biaxial stress level were used to assess the anisotropy²⁶ of the cornea as follows: A_0.6 = (λ₁ − λ₂)/0.5(λ₁ + λ₂). Where A_0.6  is the anisotropic parameter at 0.6 MPa. Where λ₁ and λ₂ represent the stretch in the horizontal and vertical direction, respectively. The selected value of 0.6 MPa is a stress value beyond the physiological range, and the anisotropic parameter represents the anisotropy that would exist when the cornea was stretched into this range. 

Statistical Analysis

Statistical analysis was performed with R 4.1.0 (the Foundation for Statistical Computing, Vienna, Austria). Descriptive statistical results are presented as the mean ± SD. Differences between the two AL groups were tested using 1-way ANOVA. Correlation was analyzed using Pearson correlation test and simple linear regression. Any P value less than 0.05 was considered a significant difference. 

Results

As shown in the Table, the eyes of 51 participants were studied, including 28 individuals in the AL < 26 mm group and 23 individuals in the AL ≥ 26 mm group. There were significant differences in spherical equivalent (SE) and thickness of the lenticule (T) between the two groups (P < 0.001). Age, sex, biomechanically corrected IOP (bIOP), CCT, and diameter of the lenticule (D) did not show statistical differences in the two groups (P > 0.05). 

Table.

 

Demographic and Clinical Characteristics of the Enrolled Corneal Stroma Lenticules

 

Corneal Biomechanical Properties Decreased With AL Elongation

Compared with AL < 26 mm, SSI significantly decreased when AL ≥ 26 mm (P = 0.048). Averaged stress-strain curves and their corresponding deviations of the AL < 26 mm and the AL ≥ 26 mm groups are shown in Figure 2A. The tensile mechanical properties in horizontal and vertical directions of AL < 26 mm group were stronger than those of the AL ≥ 26 mm group. Statistical analysis reveals that the horizontal and vertical tensile moduli at 0.03 strain in the AL < 26 mm group were significantly higher than that in the AL ≥ 26 mm group, respectively (Fig. 2B). Compared with the AL < 26 mm group, the tensile moduli at 0.03 strain of the AL ≥ 26 mm group decreased by 37.2% (13.23 ± 4.23 MPa vs. 8.31 ± 2.70 MPa) horizontally and 28.6% (13.49 ± 4.67 MPa vs. 9.63 ± 4.04 MPa) vertically. Compared with the vertical direction, the biomechanical property in the horizontal direction decreased more rapidly with the increase of the AL. 

Figure 1.

 

Schematic of biaxial tensile test.

Figure 2.

 

(A) Summary of the averaged experimental stress-strain results with deviations for the AL < 26 mm and the AL ≥ 26 mm groups tested using equibiaxial protocol. (B) The tangent modulus calculated at 0.03 strain in the horizontal and vertical directions for the AL < 26 mm and the AL ≥ 26 mm groups.

Anisotropic Parameter In Vitro is Positively Correlated With AL

There is a positive correlation between AL and anisotropic parameter in all data groups (r = 0.307, P = 0.028), Figure 3A. In the AL ≥ 26 mm group, the mean value of anisotropic parameters is positive, whereas in the AL < 26 mm group, it is negative. There is a significant difference between them (P = 0.029), Figure 3B. This reveals that the corneal biomechanical properties of AL ≥ 26 mm group were stronger in the vertical direction than in the horizontal direction, whereas the other group is opposite. Figure 4 shows that only the vertical tangent modulus at 0.03 strain of the AL ≥ 26 mm group has no correlation with AL. 

Figure 3.

 

(A) Relationship between AL and anisotropic parameter in the AL < 26 mm, AL ≥ 26 mm group, and all data. (B) Box plots with points plot overlay of anisotropic parameter of the AL < 26 mm and the AL ≥ 26 mm groups. Note, there are 3 data points of the AL < 26 mm group and 1 data point of the AL ≥ 26 mm group outside the vertical range of Figure 3B, they can be found in Figure 3A. *: 0.01 < P ≤ 0.05. AL is associated with anisotropic parameter in all data. AL is not associated with anisotropic parameter in the AL < 26 mm or the AL ≥ 26 mm group.

Figure 4.

 

Relationship between AL and tangent modulus calculated at 0.03 strain in the horizontal and vertical directions in the AL < 26 mm and the AL ≥ 26 mm group. There is a negative correlation ship between AL and tangent modulus in both directions in the AL < 26 mm group. In the AL ≥ 26 mm group, AL is negatively correlated with tangent modulus in the horizontal direction, whereas AL is not correlated with the tangent modulus in the vertical direction.

SSI In Vivo is Negatively Correlated With AL When AL < 26 mm but not for AL ≥ 26 mm

SSI is negatively correlated with AL in the AL < 26 mm, and AL in the AL ≥ 26 mm group, there is no correlation between SSI and AL (Fig. 5). The tangent modulus at 0.03 strain is negatively correlated with AL except in the vertical direction of the AL ≥ 26 mm group (Fig. 6B). 

Figure 5.

 

Relationship between AL and SSI in the AL < 26 mm and AL ≥ 26 mm groups. There is a negative correlation ship between the AL and the SSI in the AL < 26 mm group, but not in the AL ≥ 26 mm group.

Figure 6.

 

(A) Averaged experimental stress-strain results with deviations for the AL < 26 mm, 26 mm ≤ AL < 27 mm and 27 mm ≤ AL < 28 mm group tested using equibiaxial protocol. (B) The tangent modulus calculated at 0.03 strain in the horizontal and vertical directions for the 26 mm ≤ AL < 27 mm and 27 mm ≤ AL < 28 mm group.

Averaged stress-strain curves and their corresponding deviations of the AL < 26 mm, 26 mm ≤ AL < 27 mm, and 27 mm ≤ AL < 28 mm groups shown in Figure 6A. Compared with the 26 mm ≤ AL < 27 mm group, the horizontal and vertical corneal biomechanical property of the 27 mm ≤ AL < 28 mm group shows a decrease trend. Statistical analysis shows that the tangent modulus at 0.03 strain in the horizontal direction was statistically different between the two groups but not in the vertical direction (P = 0.011). 

SSI Was Significantly Correlated With the Tangent Modulus in the Vertical Direction in AL < 26 mm Group

The correlation analysis between SSI and tangent modulus have been analyzed (Fig. 7). SSI was significantly correlated with the tangent modulus in the vertical direction (P = 0.025, r = 0.422) in the AL < 26 mm group. When AL is greater than 26 mm, there is no correlation between SSI and the tangent modulus in the horizontal and vertical directions. 

Figure 7.

 

Relationship between the tangent modulus calculated at 0.03 strain in the horizontal and vertical directions and SSI. (A) AL < 26 mm group, and (B) AL ≥ 26 mm group.

Discussion

The study of corneal biomechanics in myopic eyes is of rising interest to investigators due to the scarcity of sources of normal corneal samples and the difficulty of measuring accurate biomechanical parameters in vivo. Meanwhile, corneal changes in patients with myopia have not been confirmed. Biaxial tensile tests and Corvis ST was conducted in a myopia participant in our study, and showed that biomechanical properties of the cornea decreased with the increase of AL. Compared with the vertical direction, the biomechanical property in the horizontal direction decreased more rapidly with the increase of the AL. In addition, the nonuniform deformation of the eyeball leads failure of SSI parameters to measure the mechanical properties of the cornea with the increase of AL. 

Considering that the anterior 40% of the central corneal stroma is stronger than the posterior 60% of the stroma,²⁷ a large amount of the posterior stroma in the lenticules may affect the results of the biaxial tensile test of the two groups. In this study, all the lenticules were almost taken from the anterior cornea (close to anterior 37.0% and 42.8% for the AL < 26 mm and the AL ≥ 26 mm groups, respectively). Furthermore, due to the anterior corneal lenticules that were used for biaxial tensile test, the tangent modulus in this study would be higher than that of the whole cornea reported in literature. 

The tangent modulus at 0.03 strain is about 13.4 MPa of the AL < 26 mm group when using the biaxial tensile test in this study. Spiru et al.²⁸ used a spherical indenter to apply the three-dimensional test force from the posterior surface, the results showed that the elastic modulus was about 8.22 MPa when the postoperative strain was 0.5% to 2%. Their lower modulus results may be caused by the fact that we chose the strain points, which is 1% larger than theirs. In addition, they used the effective elastic modulus and we used the tangent modulus, which is going to be a little bit larger. Kanellopoulos et al.²⁹ used biaxial tensile and measured the mechanical properties of the whole cornea, the tangent modulus at 0.1 strain is about 5.09 MPa for the SMILE control group. The reason why their results are at a higher strain and their tangent modulus is lower than ours may be due to the anterior cornea we used for testing, and the stiffer anterior cornea will lead to bigger tangent modulus. Another reason may be that their corneas had undergone SMILE surgery. 

The biaxial stress-strain curves show that the biomechanical properties of corneas in the AL ≥ 26 mm group decreased compared with those in the AL < 26 mm group (see Fig. 2A), and there is a significant decrease in the tangent modulus at 0.03 strain in both directions (see Fig. 2B). Compared with AL < 26 mm, SSI also significantly decreased when AL ≥ 26 mm. Our results demonstrate that the corneal biomechanical properties of the severely elongated group (AL ≥ 26 mm) were lower than those of the moderately elongated group (AL < 26 mm). This finding is in agreement with the results of previous studies which have been reported.^7,21 

Interestingly, the corneal biomechanical properties gradually change from horizontal enhancement (negative anisotropic parameter) to vertical enhancement (positive anisotropic parameter) with the increase of AL (see Fig. 3A). Although the values of the anisotropy parameters are close to 0, the differences in the anisotropy represented are large (Fig. 8). Moreover, with 26 mm as the dividing line, the AL < 26 mm group mainly show horizontal enhancement and the AL ≥ 26 mm group mainly show vertical enhancement (see Fig. 3B). This may be due to the inconsistent decline rates of corneal biomechanical properties in the horizontal and vertical directions, and that the average tangent modulus decline rate is 37.20% in the horizontal direction and only 28.62% in the vertical direction (see Fig. 2B). The results of correlation analysis show that the vertical tangent modulus at 0.03 strain of the AL ≥ 26 mm group has no correlation with AL, whereas the tangent modulus is negatively correlated with AL in the horizontal direction of the AL ≥ 26 mm group as well as in both directions of the AL < 26 mm group (see Fig. 4). 

Figure 8.

 

Schematic diagram of anisotropic parameter = 0.006.

Xue et al. tested the anisotropy of corneal stromal lenticules by uniaxial tensile test, and the results showed that the averaged mechanical properties in the vertical direction were slightly stronger than those in the horizontal direction.³⁰ Similar results have also been reported.³¹ Considering that uniaxial tensile test cannot measure the anisotropy of a specific sample, it can only measure the difference of the mean value of the horizontal and vertical mechanical properties of large samples. The difference between previous studies and our results may be caused by the moderate myopic corneal stromal lenticules they used. In the present study, a wider range of myopia was included. The biomechanical anisotropy of the cornea may be caused by nonuniform three-dimensional expansion of the eyeball during myopic development. Atchinson et al. reported that AL increased more significantly (0.35 mm/D) than in height (0.19 mm/D) or width (0.10 mm/D) in myopic eyes with the increasing severity of myopic degree.³² This means that the cornea stretches change more in the vertical direction as the degree of myopia increases. The slow decrease of biomechanical properties in the vertical direction with the increase of AL may be due to more activation of collagen fibers in the vertical direction of the cornea, as the higher stretch of the eyeball in the vertical direction than in the horizontal direction (Fig. 9). The schematic diagram of fiber activation is based on Jan's report in 2017.³³ 

Figure 9.

 

Schematic of conclusion. AL, axial length; CBP, corneal biomechanical properties; V, vertical; H, horizontal; BAP, biomechanical anisotropic parameter. Up and down arrows mean increase and decrease. The horizontal blue line indicates no significant change.

The increase in the AL is essentially the result of the expansion of the ocular wall.⁷ Sclera occupies more than 90% of the surface area of the eyeball,³⁴ and is the tissue with the highest tangent modulus of the eyeball.³⁵ The average tangent modulus of the middle three curves of the sclera under the biaxial tensile test reported in the literature was recalculated,³⁶ and the results showed that the tangent modulus of the sclera at 0.02 strain was 5 times that of the cornea (44.78 MPa vs. 9.04 MPa, Eilaghi's results versus our results, respectively). The tangent modulus of the sample under biaxial loading is larger than that under uniaxial loading. For example, a thin sample with incompressible linear elasticity has an effective modulus twice that of elasticity when stretched equibiaxially. Considering the nonlinear stress-strain relationship of soft tissue, the difference between the tangent modulus of the sclera and cornea will be larger with the increase of strain. These results suggest that the deformation of the sclera affects the mechanical environment of the cornea and changes the mechanical properties of the cornea. 

Previous studies suggested that owing to the fact that the sclera and cornea together form the outer layer of the eyeball, changes in cornea biomechanics could reflect structural and functional change in the sclera around the optic nerve caused by glaucomatous damage to some extent.^37–39 It has also been reported that corneal biomechanical properties are related to retinal deformation⁴⁰ and iris sectional parameters.^41,42 This may be due to the deformation of the sclera, which dominates the eyeball nonuniform deformation, leading to corneal mechanical properties connected with other tissue characteristics. 

The relationship between biomechanical characteristic and axial eye elongation showed conflicting result in vivo and ex vivo tests in our study. Tang et al. proposed that the longer AL associated with a more deformable cornea in both children and adults.⁴³ Additionally, previous clinic-based studies have shown that corneal biomechanics parameter corneal hysteresis (CH) measured by ocular response analyzer (ORA) has a negative correction with AL.^44,45 Lee et al. reported a positive correlation between deformation amplitude (DA) and AL, as well as a negative correlation between highest concavity radius (HCR) and AL, using Corvis ST.⁴⁶ In our study, biaxial tensile tests showed that with the increase of AL, the biomechanical properties decreased, and statistical difference can be observed in the different AL groups (see Fig. 6). The results of the correlation between SSI and AL show a negative correlation in the AL < 26 mm group, and no correlation in the AL ≥ 26 mm group (see Fig. 5). Although it was observed that SSI in the AL ≥ 26 mm group did not decrease with the increase of AL. Liu et al. speculated that this result may be explained by the nonuniform expansion of the eye during the development of myopia, that in the later stage of high myopia development the major nonuniform dilatation occurs mainly in the posterior pole and does not affect the mechanical properties of the cornea.²¹ Chu et al. also believed that the insignificant correlation may be related to the stabilization of the biomechanical properties of the cornea in the late stage of eyeball expansion.⁷ Considering the nonuniform expansion of the eyeball with myopic degree increase, the corneal mechanical properties show difference in the horizontal and vertical directions due to different degrees of extension, which means that the cornea begins to shown anisotropic characteristic. Because the cornea is assumed to be an isotropic material when calculates SSI,²⁰ this anisotropic change of corneal biomechanics may be the reason why SSI cannot measure the biomechanical properties of the cornea when AL ≥ 26 mm. 

The different corneal biomechanical characteristic obtained from biaxial tensile test and SSI could be due to differences in measurement as well as analysis methods. In the process of cornea response to air puff recorded by Corvis ST, the vertical fiber mainly carries capacity due to the stronger mechanical properties in the vertical direction. For SSI calculation, the cornea was assumed to be an isotropic material. This results in the inability of SSI to measure changes in the biomechanical properties of the cornea with AL greater than 26 mm. This means that to accurately evaluate the mechanical properties of the cornea in the AL ≥ 26 mm group, it is necessary to modify the computational model to take into account its anisotropic character. 

SSI was significantly correlated with the tangent modulus in the vertical direction in AL < 26 mm group (see Fig. 7). This provides evidence for the conclusion that SSI can accurately reflect the mechanical properties of the cornea when the AL is less than 26 mm. When the AL is greater than 26 mm, there is no correlation between SSI and the tangent modulus in the horizontal and vertical directions, which may be due to the vertical tangent modulus hardly changes, and the small range of variable changes leads to the inability to calculate the correlation with SSI. 

Limitations

First, although we provide evidence that corneal anisotropic character changes with AL growth, more direct evidence needs to be verified by magnetic resonance imaging and histological analysis. Second, we used the mechanical properties of the anterior stromal lenticules to represent the mechanical properties of the whole cornea. It was previously demonstrated that the existence of an elasticity gradient within the corneal stroma, and the elasticity of the posterior stroma was weaker than the anterior stroma.^27,47 With the lengthening of the AL and increase of myopia degree, the thickness of the removed corneal stromal lenticular also increases. Samples with longer AL will contain a more posterior portion of the stroma, which may lead to bias. Third, despite that we used the middle 4 × 4 mm² area of the lenticules for mechanical testing, the use of the central thickness for stress calculation also has some influence on the stress. Fourth, the sample size of this study is relatively small. 

Conclusion

The horizontal and vertical biomechanical properties of the cornea decreased with the increase of AL. With the increase of the AL, the nonuniform deformation of the eyeball leads to the increase of the vertical stretch of the cornea and increase of the vertical fiber activation, as well as the slow decline ratio of the vertical biomechanical properties with the increase of the axis of the eye. With the increase of AL, the mechanical properties in the horizontal direction decreased significantly. This results in the failure of SSI parameters to measure the mechanical properties of the cornea with the increase of AL. The graphical abstract is given in the Supplementary Material. 

Acknowledgments

The authors thank Yawei Zhao for her help in producing the pictures in this article. 

Funded by the National Natural Science Foundation of China (Grant Nos. 12072218, 12002231, 12172243, and 81770955), Project of Shanghai Science and Technology (Grant No. 20410710100). 

Disclosure: L. Chen, None; Y. Huang, None; X. Zhang, None; Y. Shi, None; Z. Gao, None; B. Sun, None; Y. Shen, None; L. Sun, None; Y. Cao, None; Q. Zhang, None; J. Guo, None; F. Li, None; W. Chen, None; X. Li, None; X. Zhou, None