Myopia is a common cause of vision loss. It is predicted that by 2050, 938 million people accounting for 9.8% of the world's population will have high myopia1. People with high myopia are prone to alters in retina, choroid and sclera due to axial length (AL) lengthening and scleral tissue remodeling. It may increase the risk for developing myopic retinopathy, retinal detachment or glaucoma, all of which can lead to incurable vision loss2,3.

Previous studies of histology have found that in axial myopia, the collagen fibers of the sclera elongated and deformed due to traction, which affected the biomechanical properties of the whole eyeball4,5. In addition, studies revealed that cornea, sclera and lamina cribrosa were all composed of similar extracellular matrix (ECM)6. Then the corneal biomechanical properties may to some extent represent that of these tissues.

Corneal Visualization Scheimpflug Technology (Corvis ST) is a relatively new non-contact tonometer with an ultra-high-speed camera system, which can be used to measure the dynamic corneal response parameters (DCRs). This measurement method has been demonstrated to have good reproducibility and repeatability7. Unlike the ocular response analyzer (ORA), it was reported to correlate directly with corneal stiffness (stiffer or softer)8,9,10.

The potential difference of corneal biomechanical parameters between different degrees of myopia has become a research hotspot. However, there is no unified conclusion. Lee et al.11 found significant differences in deformation amplitude, peak distance and outward applanation velocity in varying degrees of myopia. And both peak distance and outward applanation velocity were related with intraocular pressure (IOP), central corneal thickness (CCT) and refractive error. However, they did not consider AL lengthening, which is a major factor for myopia progression. Also, controlling the effects of CCT and IOP on corneal biomechanics is an important issue that is often overlooked in comparative studies.

The objective of this prospective study was to compare the DCRs using Corvis ST in patients with varying degrees of myopia after adjusting for biomechanically corrected intraocular pressure (bIOP) and CCT. And the potential influencing factors related to biomechanical properties were also investigated. It will help to understand the ocular characteristics of different degrees of myopia, which is conducive to the clinical prevention and control of high myopia.

Methods

Participants and ophthalmic examination

This prospective, cross-sectional study was performed at Beijing Tongren Eye Center and approved by institute ethics committee. This study was complied with tenets of Declaration of Helsinki and all participants signed a written informed consent.

Myopic spherical equivalent (SE) subjects with astigmatism less than 1.5 dioptre (D) were recruited. The exclusion criteria were as follows: (1) history of corneal or ocular disease (cataract, glaucoma or ocular hypertension), (2) history of refractive or other intraocular surgery, (3) systemic disease (hypertension, diabetes or connective tissue disease).

Participants were classified into 3 groups according to SE and AL: (1) Non-myopia (NM, SE > − 0.50 D and AL < 26 mm), (2) Mild-to-moderate myopia (MM, − 6.00 D < SE ≤ − 0.50 D and AL < 26 mm), (3) High myopia (HM, SE ≤ − 6.00 D or AL ≥ 26 mm).

Each subject underwent a general ophthalmic assessment, including visual acuity and subjective refraction, slit-lamp examination, CCT and AL measurement (Lenstar LS900, Haag-Streit Koeniz, Switzerland), fundus examination (Nonmyd WX 3D, Kowa Company Ltd., Japan). Corneal biomechanical parameters and bIOP measured by Corvis ST (Oculus, Wetzlar, Germany). All the above examinations were performed by the same experienced ophthalmologist between 9 AM and 5 PM.

Measurement of corneal biomechanical properties

The biomechanical parameters of cornea were measured by Corvis ST, a relatively new non-contact device with an automatic release mode. The corneal deformation process response to an air impulse was recorded by a high-speed Scheimpflug camera that could capture 4330 images/s. bIOP and ten DCRs were recorded for further analysis, where DCRs included the velocity at the first and second applanations (A1-velocity and A2-velocity), time from start until highest concavity (HC-time), distance between the two peaks at highest concavity of the cornea (PD), maximum deflection amplitude at highest concavity (DA-Max), whole eye movement in vertical direction at highest concavity (WEM-Max), the ratio of the deformation amplitude at the apex to that at 2 mm (DA Ratio-2 mm), the area under the inverse concave radius curve (integrated radius, IR), stiffness parameter at first applanation (SP-A1) and Corvis biomechanical index (CBI). Finally, measurements with a quality score of “OK” were included in subsequent analysis.

Statistical analysis

SPSS software version 22.0 (SPSS, Inc., Chicago, IL, USA) was used for statistical analysis. The normality of data was examined by Kolmogorov–Smirnov test. And the Levene test was used to check the homogeneity of the variance. The categorical and continuous variables were shown as counts (percentages) and means ± standard deviation respectively. Clinical characteristics for categorical variables among three groups were compared by chi-square test. The continuous variables across the groups were compared by one-way ANOVA test or the Kruskal–Wallis test. A linear mixed-effects model was conducted to account for the use of both eyes for the same subject. And the model was performed to evaluate how the corneal biomechanical parameter was influenced by varying degrees of myopia after adjusting for bIOP and CCT. Then the pairwise comparisons were conducted by Dunn-Bonferroni post-hoc test. Further, multiple linear regression with stepwise method of predictors (Entry P < 0.1; removal P > 0.2) was used to evaluate the relationship between corneal biomechanical parameters and SE, AL, CCT or bIOP. p < 0.05 was considered statistically significant.

Results

Clinical demographics of the study groups are shown in Table 1. A total of 304 eyes from 224 healthy myopic subjects were included. 95 eyes had NM, 122 eyes had MM, and 87 eyes had HM. The mean (SD) ages were 48.72 (15.85) years in the NM group, 47.02 (16.13) years in the MM group, and 44.16 (15.12) years in the HM group. Significant differences were found in SE and AL among the different groups (all p < 0.001). No significant differences were found in age (p = 0.147), gender distribution (p = 0.395), laterality (p = 0.109), best-corrected visual acuity (p = 0.426), CCT (p = 0.218) and bIOP (p = 0.152) among the three groups.

Table 1 Demographic and clinical characteristics of the study groups.
Full size table

The comparison of DCRs among three groups with CCT and bIOP as covariates is presented in Table 2 and Fig. 1. There were statistically significant differences for HC-time (p = 0.025), PD (p = 0.001), DA-Max (p = 0.002), WEM-Max (p < 0.001) and SP-A1 (p < 0.001) among the different myopic groups. The pairwise comparisons performed by Dunn-Bonferroni post-hoc test shown a statistically significant difference in mean WEM-Max between all pairs, and in mean HC-time between MM and HM group (p = 0.036), and in mean PD between NM and HM group (p = 0.001), and between MM and HM group (p = 0.006), and in mean DA-Max between NM and MM group (p = 0.012), and between NM and HM group (p = 0.004), and in mean SP-A1 between NM and HM group (p < 0.001), and between MM and HM group (p < 0.001).

Table 2 Comparison of the corneal biomechanical parameters in different myopia groups with CCT and bIOP as covariates.
Full size table
Fig. 1
figure 1

Histogram comparison of corneal biomechanical parameters in different myopia groups. NM, Non-myopia; MM, Mild-to-moderate myopia; HM, High myopia; A1-velocity, applanation1-velocity; A2-velocity, applanation2-velocity; HC-time, highest concavity-time; PD, peak distance; DA-Max, deflection amplitude-Max; WEM-Max, whole eye movement-Max; DA Ratio-2 mm, deformation amplitude ratio-2 mm; IR, integrated radius; SP-A1, stiffness parameter at first applanation; CBI, corvis biomechanical index. *means p < 0.05, **means p < 0.01 ***means p < 0.001.

Full size image

The results of multiple linear regression analysis to investigate the relationship between each corneal biomechanical parameter with SE, AL, CCT and bIOP are shown in Table 3. All DCRs were found to be correlated with CCT or bIOP. However, the relationship between parameters directly related to the degree of myopia (AL or SE) and corneal biomechanical parameters was different. A1-velocity, A2-velocity, DA Ratio-2 mm and IR were found to be independent of myopia parameters (AL or SE). Higher level of myopia (increased AL) was associated with decreased HC-time (B coefficient = − 0.065, p < 0.001), increased PD (B coefficient = 0.055, p < 0.001), increased DA-Max (B coefficient = 0.027, p = 0.001), decreased WEM-Max (B coefficient = − 0.024, p < 0.001), and decreased SP-A1 (B coefficient = − 2.509, p < 0.001). In addition, higher level of myopia (greater negative SE) was associated with increased CBI (B coefficient = − 0.007, p = 0.016). In another word, for each extra 1 mm increased in AL, there was a 0.065 ms decrease in HC-time, a 0.055 mm increase in PD, a 0.027 mm increase in DA-Max, a 0.024 mm decrease in WEM-Max, and a 2.509 mmHg/mm decrease in SP-A1. In addition, for every extra diopter in myopic SE, CBI increased by 0.007. The above statistically significant differences showed that corneal stiffness gradually decreased with increasing myopia and was independent of CCT and bIOP.

Table 3 Multiple regression analysis for variables predicting corneal biomechanical parameters.
Full size table

Scatter plots of changes in corneal biomechanical parameters as a function of AL and SE are shown in Fig. 2.

Fig. 2
figure 2

Correlation between corneal biomechanical parameters and AL and SE. HC-time, highest concavity-time; PD, peak distance; DA-Max, deflection amplitude-Max; WEM-Max, whole eye movement-Max; SP-A1, stiffness parameter at first applanation; CBI, corvis biomechanical index. AL, axial length; SE, spherical equivalent.

Full size image

Discussion

The current study dedicated to explore the corneal biomechanical response parameters in eyes with varying degrees of myopia, grouped according to SE and AL, using Corvis ST with a large sample size. In addition, the potential impact factors associated with corneal biomechanics were also investigated. In the present prospective study, after adjusting for bIOP and CCT, eyes with high myopia had a significantly shorter HC-time (p = 0.025), greater PD (p = 0.001), greater DA-Max (p = 0.002), smaller WEM-Max (p < 0.001) and reduced SP-A1 (p < 0.001) than the non-myopia and/or mild-to-moderate myopia. Moreover, multiple regression analysis revealed that five corneal biomechanics were significantly correlated with AL, and one parameter has significant relationship with SE. In other words, significant changes in corneal biomechanical parameters with increasing myopia (increased AL or greater negative SE), showed a progressive decrease in corneal stiffness, independent of bIOP and CCT.

Cornea is a tissue with viscoelastic properties. And the corneal biomechanical properties not only represent the viscoelastic properties of cornea but also contain the mechanical strength of stromal collagen fibers12. Recently, the study of corneal biomechanics has played a significant role in the diacrisis, management and treatment of various disorders such as glaucoma, corneal refractive surgery and different corneal diseases13,14.

The relationship between corneal biomechanics and myopia was also reported. A systematic review of 11 studies found that the corneal hysteresis (CH) and corneal resistance factor (CRF) measured by ORA were lower in high myopia than in low myopia15. In addition to ORA, the newly developed device Corvis ST could also be used to measure corneal biomechanics. It has been reported to be directly related to corneal stiffness (stiffer or softer)8,9,10. Although a number of studies have assessed the relationship between corneal biomechanical properties measured by Corvis ST and myopia, a few crucial questions remain unresolved. Some myopia is caused by high corneal refraction or the co-existence of corneal refraction and longer AL. AL elongation is the main cause of myopia progression and its effect on corneal biomechanics has been neglected in previous articles16,17. In addition, it has been shown that the biomechanical properties are connected with CCT. The thicker the CCT, the more resistant the cornea is to external forces, and the less likely it is to deform18. High IOP may mask abnormalities in corneal biomechanics, leading to apparently normal DA measurements19. Therefore, it is important to control the effects of CCT and IOP on corneal biomechanics in comparative studies. Furthermore, except for the Corvis ST’s classic biomechanical parameters, some new parameters should also be considered.

Our results show that the parameters associated with the highest concavity phase not only differ significantly with increasing myopia but also have the strongest correlation. Studies have suggested that the sclera affects corneal deformation through displaced fluid, thus, the sclera has the greatest influence on the cornea at the highest concavity20,21. Then the corneal biomechanical parameter may be used as an indicator for scleral mechanical strength in high myopia.

HC-time represents the time from initial state to highest concavity stage after the application of an air impulse at the corneal apex. We found significantly shorter HC-time in HM group (p = 0.036) than in MM group, and the decreased HC-time was associated with increased AL (B coefficient = − 0.065, p < 0.001). Then in the deformation-reformation cycle, the time for the cornea from the apex to the maximum deformation is shorter, implying that the cornea is softer and more prone to deformation.

Peak distance is defined as the distance between the two highest points of cornea at the maximum depression. Under the same IOP, the PD of hard cornea is smaller than that of soft cornea. In our study, after adjusting for bIOP and CCT, the PD was greater in high myopia eyes (p = 0.001). In addition, increased PD was found to correlate with increased AL (B coefficient = 0.055, p < 0.001), suggesting that the cornea of high myopia is softer and more deformable with the increase of AL.

The DA-Max represents the deformability of corneal resistance under air pressure, and this parameter compensates for the whole eye movement. The higher DA-Max the greater deformability of cornea, which is a stable parameter that can be used22,23. Eyes with high myopia or mild-to-moderate myopia demonstrated greater DA-Max compared with eyes in non-myopia group (NM, MM: p = 0.012; NM, HM: p = 0.004). It is suggested that the cornea becomes weaker and more deformed in myopic eyes. This observation is similar with the results of several previous studies16,17. In the same way, increased DA-Max was found to correlate with increased AL (B coefficient = 0.027, p = 0.001).

The whole eye movement is an index that reflects the eye movement during the measurement and represents the overall force profile of cornea, eyeball and the constant component. The eyeball itself moves slightly backward when the air is released, and forward again when the cornea returns to its initial contour. In our study, eyes in HM group had a significantly smaller WEM-Max than that in NM group (p < 0.001) or in MM group (p = 0.015). Similar findings have been reported in previous studies24. Decreased WEM-Max was also found to be related to increased AL (B coefficient = − 0.024, p < 0.001). Ocular compliance may explain these results. The longer the AL of high myopia, the greater the ocular compliance and the lower the ocular rigidity. During jetting, the eyeball was more likely to deform rather than move back, resulting in a lower WEM.

SP-A1 was a novel parameter reflecting the cornea stiffness and calculated as the the difference between air-puff pressure and bIOP, divided by deflection amplitude at the first applanation25. It described the resistance of cornea to deformation and was an effective index to assess the cornea biomechanics independent of tissue volume and IOP26. Lower SP-A1 was often accompanied by softer and more deformable corneas. Eyes with high myopia in our study had a significantly reduced SP-A1 than the non-myopia (p < 0.001) and mild-to-moderate myopia (p < 0.001). Further research found that decreased SP-A1 was also related to increased AL (B coefficient = − 2.509, p < 0.001). Which means that the cornea of patients with high myopia is softer and more deformable with the elongation of AL. As far as we know, myopia is one of the risk factors for glaucoma development. The decrease of SP-A1 in myopia may explain its pathogenesis. It is suggested that the lower SP-A1, to some extent, represent the more deformable lamina cribrosa tissue. It would be easier to bend posteriorly under IOP fluctuations then leading to the further glaucomatous injury of RGC axons. Previous study had confirmed our hypothesis and found that the greater posterior lamina cribrosa curvature was related with lower SP-A1 in glaucoma patients27.

In addition, increased CBI was associated with greater negative SE (B coefficient = − 0.007, p = 0.016). CBI is an effective auxiliary tool for keratoconus screening in clinical practice. And a cut-off of 0.5 suggested a high risk of corneal ectasia28. In our study, greater negative SE was associated with greater CBI, which may be related to the high corneal refraction.

We did not find statistically significant differences in the remaining five parameters among the three groups. Here are some reasons. Firstly, larger sample sizes are needed in detecting subtle differences in these parameters. Secondly, these parameters may not represent corneal biomechanical properties on their own and should be used with caution when accounting for the corneal biomechanics. Future studies with different ethnicities and larger sample sizes are needed to investigate the significance of these biomechanical parameters.

The cornea biomechanical parameters obtained by Corvis ST are some geometrical data that are mainly decided by three different factors: the corneal biomechanical properties, the air puff pressure and the IOP. Air puff pressure is constant in all cases. Previous studies confirmed that IOP and CCT were the two main factors that affect corneal biomechanical parameters in healthy eyes29. However, corneal biomechanical properties are much more difficult to be determined in vivo. Which may be changed during the development of certain ocular disorders such as keratoconus, glaucoma and myopia. Therefore, IOP and CCT may have different effects on corneal biomechanical parameters in specific diseases. In current study, all the biomechanical parameters were found to be correlated with CCT or bIOP. In briefly, the thinner the CCT and the lower the bIOP, the softer and more easily deformed the cornea. The hypothesis suggests that axial elongation due to elevated IOP may be one of the underlying mechanical factors leading to myopia and anisometropia in healthy white children30. However, the current study did not find significant difference in bIOP among groups with different degrees of myopia. This may be related to the fact that the study includes only the adult subjects, so the results cannot be extrapolated to children.

After statistical correction for these possible confounders (bIOP and CCT), this study also found a significant relationship between corneal biomechanical parameters and AL/SE. The significantly correlation between increasing myopia (increased AL or greater negative SE) with corneal biomechanics in highest concavity phase, means a “softer” cornea in patients with high myopia and longer AL. This possible conclusion is supported by a study reporting that lower tangential modulus of cornea and therefore lower corneal stiffness in patients with high myopia31.

The reasons why corneal biomechanical properties are related to the degree of myopia may be explained from the following aspects. The main feature of myopia is excessive elongation of the eyeball32. Studies have shown that the prolongation of AL is related to corneal thinning and flattening, which can result in corneal biomechanical properties changes33. In addition, myopia was thought to be closely related to scleral structural abnormalities and scleral remodeling34. Studies found that the elongation of AL is related to the changes of scleral collagen structure, fiber diameter and morphology35. With the increase of diopter in myopic patients, the rate of proteoglycan synthesis and the fiber diameter decreased significantly. Which may induce scleral thinning and scleral tissue loss, then the scleral mechanical property weakens and the deformability increases during scleral remodeling36,37,38,39,40. Moreover, histological studies have found that in axial myopia, the collagen fibers of the sclera elongated and deformed due to traction, which affected the biomechanical properties of the whole eyeball4,5. In addition, studies have proved that the cornea, sclera and lamina cribrosa were composed of similar ECM constituents41. Thus, similar changes may occur in cornea during myopic development. At the same time, the potential possibility of corneal biomechanical parameters as evaluation of the intact eyeball structure was discussed12,42. The stiffer sclera will limit corneal deformation. In contrast, the sclera is softer in highly and extremely myopic eyes and may allow greater corneal displacement at highest concavity stage20,21. Then the corneal biomechanics could be used as an index to evaluate the mechanical strength of sclera in high myopia, which has certain application value in clinical myopia management.

This study has several limitations as follows. First, longitudinal studies are needed to identify whether the differences in corneal biomechanical properties are responsible for the development of myopia. Second, the recruited subjects were all based on Chinese urban population. There may be racial differences in corneal biomechanics, and the conclusions need to be verified with other races. In addition, the results of this study cannot be extrapolated to children because only adult subjects were included. Lastly, more other available parameters updated by Corvis ST software, may afford more information about the biomechanical properties of the eye.

Conclusions

In summary, we compared corneal biomechanics in varying degrees of myopia using Corvis ST. After adjusting for confounding factors, eyes with high myopia demonstrated shorter HC-time, greater PD, greater DA-Max, smaller WEM-Max and reduced SP-A1. And significant changes in several corneal biomechanical parameters with increasing myopia (increased AL or greater negative SE) were also found. The above results suggest that with the increase of myopia, cornea becomes more deformable and susceptible to stress due to the decrease of corneal stiffness. Then the corresponding changes in corneal biomechanics may represent the scleral stiffness and be applied to the management of high myopia.