Age distribution and associated factors of cornea biomechanical parameter stress-strain index in Chinese healthy population

Corneal biomechanical determination is of great importance in clinical evaluation, such as diagnosis of keratoconus [1], assessment before corneal refractive surgery [2, 3], and measurement of corrected intraocular pressure [4, 5]. In recent years, various instruments that measured corneal biomechanics were developed, such as ocular response analyzer (ORA) and Corvis ST. Meanwhile, a variety of parameters to assess corneal biomechanics were provided. However, since cornea consists of a viscoelastic material, and its stress-strain behavior of biological tissue is nonlinear [6, 7], the cornea shows the biomechanical properties of changes under different intraocular pressure load. Hon reported a corneal indentation device (CID) that can measure the stiffness (S) and tangent modulus (E) of the cornea in vivo, but S and E due to IOP rise could not be ruled out [8, 9]. It was always a difficult problem to evaluate corneal biomechanics without the influence of IOP in vivo [10‐12].

A new method to obtain the stiffness of corneal material was proposed by Eliasy, which eliminated central corneal thickness (CCT) and biomechanical corrected intraocular pressure (bIOP) in vivo [10]. A biomechanical Corvis parameter, stress-strain index (SSI) has been proved to be almost independent of bIOP and CCT. It was shown as an ideal method to estimate the mechanical properties of corneal tissue material. The distribution of SSI in Italian and Brazilian populations has been reported by previous studies and SSI was found to be significantly affected by age [10]. Chua observed that there was a lower corneal hysteresis (CH) in Chinese than Indians, suggesting that corneal biomechanics may be varies among different populations [13]. At present, the SSI distribution of different age in Chinese population has not been reported.

Corneal stiffness is one of the important biomechanical parameters, which reflects the ability to resist stress deformation [14]. Previous studies have shown that corneal biomechanics could be affected by multiple factors. A larger anterior chamber volume (ACV) was associated with a lower maximum amplitude at the apex of highest concavity (DA) and a higher stiffness parameter (SP-A1) in primary angle closure (PAC) eyes [15]. In healthy people, corneal biomechanical parameters were significantly correlated with the corneal curvature, refractive error [16, 17] but not with ACV [18]. However, Lim found that corneal biomechanics was not associated with age, refraction error, but related to CCT, cornea-compensated intraocular pressure (IOPcc) and anterior curvature [19]. While other factors that may influence SSI besides age have not been reported. Therefore, the purpose of this study is to explore the distribution of SSI in different age in healthy Chinese population and investigate the factors associated with SSI via Corvis ST.

A total of 175 left eyes from 175 healthy patients who met the study inclusion criteria were analyzed. Among of them, 75 (42.9%) were male and 100 (57.1%) were female. SSI showed a normal distribution (P > 0.05). There were no significant gender differences existed in SSI (P > 0.05). The data are summarized in Table 1.

The average SSI was 1.14 ± 0.22 in all subjects and the values ranged from 0.66 to 1.78. To explore the differences of corneal stiffness between different age groups, patients were divided into seven groups according to age (Fig. 1). A significant rising trend was found in SSI with age increasing significantly after age of 35 (P < 0.05).

Due to the high correlation between the measurements, PLSLR and LME analysis was used to find the correlation between SSI and other parameters. After PLSLR calculation, five parameters were left, which were IOP, bIOP, AL, age and RsF, and then they were put into the LME model. The results showed that SSI was positively correlated with age, RsF and IOP, while it was negatively correlated with AL (Table 2).

This study described the distribution of corneal tissue material stiffness parameter SSI in different age groups and related factors in a healthy Chinese population. We found that SSI was relatively stable before age of 35, and then increased significantly with age. SSI was positively correlated with age, IOP, and anterior radius of curvature, meanwhile, it was negatively correlated with axial length. No significant effect was found in gender, ACV, CCT, or bIOP.

In our study, a nonlinear relationship was detected between age and SSI, showing that SSI increased with age significantly after age of 35. Age has been shown to be an important factor affecting corneal biomechanical in previous study [10]. Wang found a positive correlation between age and second applanation (A2L) in healthy Chinese adults [20]. In corneas of patients aged from 50 to 95 years, the tangent modulus increased with age [7]. Trend of corneal biomechanics with age could be accounted for changes in the molecular structure in cornea. Daxer and Malik observed that non-enzymatic crosslinking, collagen glycation, fibril diameter, and the number of collagen molecules increased with age over 40 years in corneal X-ray [21‐23]. These could explain the reason that SSI increased with age after 35. However, we found that SSI was basically stable before the age of 35. In previous studies, Kirwan used ORA to measure corneas of normal children aged 4 to 18 years, and found that CH was not correlated with age [24]. Another study found that there was no significant correlation between biomechanical parameters and age in healthy Chinese adolescents at 4–18 years of age [25]. Valbon found that deformation amplitude (DA) and other biomechanical parameters of the healthy eyes in population under 40 years old were not correlated with age [26]. Elsheikh measured the corneal tangent modulus (E) in vitro for 30–99 years old, and found that the growth rate of E was smaller in the younger, which also suggested that the changes in corneal biomechanics may be uneven with age [27]. Unfortunately, few studies have reported the relationship between corneal collagen and age in young people. Whether this relationship is the reason that no significant correlation is revealed between SSI and age in the young remains to be explored. Furthermore, the AL and refractive error increase with age in young people [28]. We speculate those may reduce the tendency of the cornea to harden with age. This inference needs to be further proved in future studies.

SSI was found to be negatively correlated with axial length, which indicated that the SSI includes a function of the whole eye biomechanics and not just the cornea. Previous studies indicated that the cornea and sclera were mainly composed of the same types of collagen [29]. In addition, when the collagen fibers of the sclera became longer and damaged in myopia progression, the overall arrangement of collagen fibers in corneal stroma also restructured [30‐32]. In infant monkeys and chicks, corneal astigmatism changes were also associated with induced eye growth [33, 34]. Chang’s study pointed out that the axial elongation led to corneal flattening and thickness reduction [35], suggesting that the increase of axial length may affect the biomechanics of the cornea and previous studies have proved this. Myopia in glaucoma and normal eyes would lead to biomechanical parameters changes such as corneal deformation amplitude (CDA), outward corneal applanation (OCA) and cornea stiffness (CS) [36, 37]. Especially the cornea in high myopia had faster outward corneal velocity (OCV) and higher CDA compared to emmetropia [16]. Long found in Chinese children, SP - A1 declined gradually between presbyopia, emmetropia and myopia groups [38]. However, in Lim’s study, CH and CRF were not significantly correlated with refractive errors by ORA measurement in corneal biomechanics in children aged 7–9 years. Lim believed the possible reason was the brief loading–unloading cycle of the ORA contrasts with the slower profile of scleral creep experiments and myopic deformation [19]. Actually, ORA exams CH or energy absorption but not corneal shape at maximum concavity. It may be more corneal specific and less surrounding affected and possibly buttressing sclera [14]. This may be the reason that biomechanical parameters measured by ORA are not correlated with refractive errors.

Consistent with the study of Eliasy, neither bIOP nor CCT was significantly correlated with SSI¹⁰, but SSI was found to be positively correlated with IOP. It was not surprising, since SSI reflected the corneal stiffness, and IOP measurement was affected by corneal stiffness [12, 39]. While it was emphasized that bIOP can exclude the influence of corneal thickness and age on intraocular pressure measurement [40], and can reflect more accurate intraocular pressure [41‐43]. Besides IOP, previous studies have shown that, parameters of corneal biomechanics in vivo were mostly affected by bIOP such as deflection area at the highest concavity and deflection amplitude (HC DefArea and HC DefA), SP - A1 [43], CH and CRF [44]. However, SSI based on finite element (FE) numerical modeling simulats the effects of Corvis ST air puff and bIOP to predict of corneal behavior and excludes the influence of bIOP, indicating that this is an ideal method for in vivo measurement of corneal tissue material stiffness [10].

There was no significant correlation between SSI and ACV in LME model. The results of this study were consistent with Gabor Nemeth and Hwang, that the ACV was not significantly related to biomechanical parameters [18, 45]. Cui’s study pointed out that decrease in ACV led to reduced corneal stiffness in the PAC suspects. This could be explained by the different population in the studies. Cui believed that the stiffness of cornea would reduce in PAC and compensated for the high IOP caused by the shallow anterior chamber [15]. But in healthy people, there was no significant effect of ACV on corneal stiffness.

This study found that the radius of anterior curvature was positively correlated with SSI. There has been controversy on the relationship between corneal curvature and biomechanics. It was found that SP-A1 was positively correlated with corneal asphericity (Q value) and radius of anterior surface [46]. In Nemeth’s study, the corneal curvature was also related to the amplitude of corneal deformation and the time taken to reach this applanation [18]. On the contrary, several studies found that there was no significant correlation between the curvature and CH or CRF [45, 47, 48]. The correlation between corneal curvature and biomechanics needs further study.

Our research had several limitations. Firstly, SSI was currently estimated with normal corneal topography, which could not be applied in corneas with biomechanical decline caused by pathological changes. The calculation method of corneas with keratoconus or ectasia needs to be further developed. Secondly, the relationship between refractive error and SSI was not analyzed. Previous studies have pointed out that refractive error was highly correlated with the axial length of the eye [49], and it has also been shown that the extension of the axial length had a significant correlation with the sclera and cornea tissue structure [32]. Therefore, we considered that the axial length was more significant as an evaluation factor.

In conclusion, this study for the first time conducted a statistical analysis of SSI in a healthy Chinese population and found that SSI was positively correlated with RsF, IOP and negatively correlated with AL. Moreover, after the age of 35, SSI increased with age.