Changes in Ocular Biomechanics During Adolescence and Its Relationship with Lifestyle and Myopic Progression: The Oporto Myopia Study

TL;DR

This study explores how eye biomechanics change during adolescence and how lifestyle factors like outdoor activity and screen time relate to myopia progression.

Contribution

The study identifies biomechanical stiffening and links lifestyle factors to myopic progression in adolescents.

Findings

Biomechanical stiffening was observed during adolescence.

Outdoor activity was associated with increased corneo-scleral stiffness and reduced myopization.

Eye rubbing and ocular surface symptoms correlated with higher myopization.

Abstract

The relationship between lifestyle, ocular biomechanical behavior, and myopia is not well established in the literature. The present study aims to describe changes in ocular biomechanics during adolescence and to explore their relationship with lifestyle factors and myopic progression. Prospective cohort study including 63 adolescents (126 eyes) with a mean age of 14.1 ± 2.6 years old examined twice over a 30 ± 0.9-month period. The data from biomechanics, biometry, corneal tomography, and lifestyle was addressed. The relationships between biomechanical changes, biometric and refractive variation, and lifestyle variables were analyzed using parametric and non-parametric statistics with a significance level of p < 0.05. A biomechanical stiffening trend was found. Axial elongation was 0.12 ± 0.17 mm, and refractive shift was −0.32 ± 0.87 D. The history of allergies was associated with greater axial growth (p = 0.032) and smaller increase in stress–strain-index (SSI) (p = 0.01). Myopization was higher in eyes with ocular surface symptoms (p = 0.049) and those with reported eye-rubbing habits (p = 0.04), with a lower gain in stiffness (p < 0.05). Outdoor activities were associated with higher gain in corneo-scleral stiffness (p < 0.05). Reduced myopization correlated directly with the increase in the SSI (p < 0.05) and inversely with the Integrated Radius (p < 0.05). Greater increases in axial length (AL), vitreous cavity length (VCL), and the ratio between VCL and AL (R_VCL/AL) correlated negatively with the increase in the SSI (p < 0.05). The increase in the R_VCL/AL correlated positively with the time spent on digital devices and negatively with the amount of outdoor activity (p < 0.05). Biomechanics may represent the physiological bridge between the environmental exposure and myopization, as lower gain in corneo-scleral stiffness was consistently associated with greater axial elongation and refractive myopization, with outdoor activity appearing to be protective.

Full text

1. Introduction

Myopia is a prevalent and complex ophthalmological condition, estimated to have affected approximately 2.5 billion people globally by 2020 [1,2]. With its prevalence on the rise, projections suggest that by 2050, myopia and high myopia will affect about 50% and 10% of the world’s population, respectively [2]. This increasing trend is expected to have significant economic implications, as robust modeling data from meta-analyses estimated that, in 2015, the global potential productivity loss due to vision impairment amounted to $244 billion from uncorrected myopia and$6 billion from myopic macular degeneration [3].

For decades, the ophthalmology community has been investigating the potential causes behind the rising prevalence of myopia in an effort to mitigate its associated burden. While genetic factors play a significant role, recent decades have shown that in economically developed societies, most cases of myopia emerge during the school years—commonly referred to as school myopia [4]—and only a small proportion of myopia cases are clearly familial, typically characterized by early onset, high severity, defined chromosomal loci, and, in some cases, causal genetic mutations [5]. The increase in myopia parallels the rise in levels of education [6] and decreased time spent outdoors [7,8,9,10].

Despite growing evidence linking children’s lifestyle factors to school-age myopia—particularly highlighted during the COVID-19 pandemic [11]—the biological mechanisms driving myopia progression remain incompletely understood. While the emmetropization theory provides the prevailing conceptual framework, the pathways translating environmental visual stimuli into structural ocular changes are not fully clarified [12,13].

Axial elongation, the structural hallmark of myopia progression, is primarily driven by the remodeling of the posterior sclera. Alterations in scleral biomechanics arise from changes in the extracellular matrix, including collagen fiber organization, cross-linking, and proteoglycan composition [14]. These processes are regulated by molecular signaling pathways involving transforming growth factor-β (TGF-β), matrix metalloproteinases (MMPs), tissue inhibitors of metalloproteinases (TIMPs), and mechanotransduction mechanisms that modulate scleral fibroblast activity [15]. A shift toward a more compliant scleral structure reduces resistance to intraocular pressure-related mechanical stress, thereby facilitating ocular elongation [16].

If environmental factors influence retinal signaling and downstream extracellular matrix remodeling, they may indirectly modulate scleral biomechanics and axial growth. Therefore, investigating ocular biomechanical behavior during childhood and adolescence may provide important insight into the relationship between lifestyle factors and myopia progression.

Nowadays, in vivo data from ocular biomechanical behavior can be obtained through the analysis of corneo-scleral tissue movement using ultra-high-speed Scheimpflug imaging during non-contact tonometry with the Corvis ST^®^ (OCULUS, Wetzlar, Germany) [17] and includes evolving several validated algorithms as descriptors of ocular biomechanical behavior and its stiffness [18,19,20,21,22]. However, the assessment of changes on ocular biomechanics throughout adolescence and its relationship with overall or segmental axial eye growth or with refractive changes have never been explored. Additionally, there is no evidence regarding the possible relationship between lifestyle and these changes during the growth phase in which school myopia develops.

The present study aims to describe the changes in ocular biomechanics during adolescence in a European cohort and to relate it with lifestyle parameters and myopic progression.

2. Materials and Methods

2.1. Design

This is a prospective cohort study. Data was collected from 2 timepoints, spaced by 30 ± 0.9 months (2.5 years).

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board (IRB) of Centro Hospitalar e Universitário de Santo António (protocol code 2020-201(158-DEFI-160-CE), 13 October 2021).

2.2. Setting

Centro Hospitalar e Universitário de Santo António, Oporto, Portugal.

2.3. Population Inclusion and Exclusion Criteria

The study population included individuals between 10 and 15 years old consecutively selected from the general ophthalmology clinic at our center. The exclusion criteria were: familial or personal ocular pathologic history besides myopia; amblyopia history: ocular surgery history; ocular inflammation at the time of examinations; corneal leucomas or other corneal or conjunctival pathologies; the inability to perform the exams; and the utilization of myopia control interventions, including atropine, contact lenses, or spectacle lenses with specialized optical designs.

2.4. Ocular Biomechanical Data

Biomechanical assessment was made at both timepoints by means of Scheimpflug camera, with Corvis ST^®^ (OCULUS, Wetzlar, Germany), through the dynamic corneal response (DCR) parameters. Only exams with ‘OK’ quality score were included. Both first-generation parameters—derived from the direct analysis of the corneal excursion image in different timepoints, including the time of first and second applanations (A1 and A2), the time of the highest corneal concavity (HC), and maximum whole eye movement (WEM)—and second-generation parameters, such as algorithms including biomechanically corrected-IOP 1 (bIOP), were considered. Corvis biomechanical index (CBI), tomographic and biomechanical index (TBI), stiffness parameter in A1 (SP-A1) and in HC (SP-HC), and stress–strain index (SSI) were analyzed. Deltas of variation were built for all parameters. Table 1 summarizes all Scheimpflug-based parameters used in the study and their explanation, including which tissue each of them is most associated with and the theoretically expected behavior in the presence of higher values.

2.5. Ocular Biometric Data

Biometric data from axial length (AL) and segmental ocular biometric lengths, including central corneal thickness (CCT), anterior chamber depth (ACD), and lens thickness (LT), were assessed at both timepoints by means of swept source optical coherence tomography (OCT) with the IOL MASTER 700^®^ (ZEISS, Oberkochen, Germany). The variable vitreous cavity length (VCL) was built [VCL = AL − (CCT + ACD + LT)]. The ratio between VCL and AL (R_VCL/AL) was calculated, and deltas of variation were built for AL (Δ_AL), VCL (Δ_VCL), and R_VCL/AL (Δ_R_VCL/AL).

2.6. Objective Refraction Data

Non-cycloplegic objective refractive status—(sphere (S), cylinder (C), and spherical equivalent (SE))—was assessed at both timepoints through the KR-800 Auto Kerato-Refractometer^®^ (TOPCON, Livermore, California (CA), United States of América). Deltas of variation were built for S, C, and SE (Δ_S, Δ_C, and Δ_SE).

2.7. Corneal Tomography Data

Keratometric values—minimum (K1) and maximum (K2)—were assessed through Pentacam HR^®^ (OCULUS, Wetzlar, Germany) at both timepoints, and deltas of variation were built (Δ_K1 and Δ_K2).

2.8. Lifestyle Data

General health data and lifestyle data from the follow-up interval were collected at the end of the follow-up through a questionnaire accepted and validated by the Centro Hospitalar e Universitário de Santo António Ethical Commission (nr 158-DEFI/160-CE), including age and questions regarding familiar history of myopia, allergies, sleeping position, dry eye symptoms, eye-rubbing habits, sun exposure, physical activity, digital device utilization, and reading habits. The anonymized data was collected via Google Form^®^ (Supplementary Materials—Questionnaire Oporto Myopia Study). Filling out the form was done by the individual with the help of parents when necessary.

2.9. Step 1—Inter-Individual Analysis

The variation (Δ) in the biomechanical parameters and ocular biometric, tomographic, and objective refraction data were compared to address the differences between individuals with family history of myopia, those with reported allergic diseases, and those with predominantly ventral sleeping position.

2.10. Step 2—Inter-Eye Analysis

The variation (Δ) in the biomechanical parameters and ocular biometric, tomographic, and objective refraction data were compared to address the differences between the groups of eyes with predominantly ipsilateral sleeping position, those with dry eye symptoms, and those with eye-rubbing habits but without these conditions.

2.11. Step 3—Study of Correlations

The variation (Δ) of ocular biomechanical parameters was correlated with the variation (Δ) in ocular biometric and objective refraction parameters and with the individual lifestyle variables of weekly hours of outdoor activity, weekly hours of physical activity, daily hours spent on digital devices, and daily hours spent reading/writing.

2.12. Statistical Analysis

Descriptive statistics of the dataset were calculated for ocular biomechanical, biometric, and refractive data. Normality of the data was tested with the Shapiro–Wilk and Kolmogorov–Smirnov tests. When the parametric analysis could be applied, the Student t-test was used to compare the variables. When non-parametric tests were needed, the Mann–Whitney test was applied. Pearson correlation coefficient and Spearman rank correlations were used in the correlation analysis. All analyses were performed using SPSS v26.0. All values are shown as mean ± standard deviation unless otherwise specified. All p-values (p) were two-sided, and p-values of < 0.05 were considered significant.

3. Results

The present study analyzed a prospective cohort including 63 adolescents (126 eyes) with a mean baseline age of 14.1 ± 2.6 years old.

Regarding baseline biometry, approximately 61% of eyes had a baseline axial length (AL) between 23 and 25 mm, while 13.5% had an AL greater than 25 mm. Concerning baseline refractive status, there was an equal distribution among eyes with myopia greater than 1 D, eyes with myopia less than 1 D, and hyperopic eyes. During follow-up, the mean ocular growth (ΔAL) was 0.123 ± 0.17 mm, and the mean myopic shift (ΔSE) was −0.321 ± 0.87 D. Biomechanical assessment revealed an overall increase in ocular stiffness over time, with mean increases in SSI, SP-A1, and SP-HC of 0.040 ± 0.12, 2.461 ± 13.2, and 0.422 ± 2.6, respectively. Additionally, a mean decrease of −0.394 ± 0.86 was observed in the integrated radius parameter.

Approximately 38% had a family history of myopia in at least one parent, 43% reported a history of allergies, and around 60% indicated the ventral position as their predominant sleeping posture. The data for each eye showed that about 41% of eyes were habitually subjected to an ipsilateral sleeping position, and symptoms of ocular surface discomfort and eye-rubbing habits were reported in approximately 44% and 67% of eyes, respectively. On average, outdoor activities occupied 3.4 ± 1.9 weekly hours and digital devices were used during 3.3 ± 1.5 daily hours.

In individuals with a family history of myopia, a significant decrease in A2 time (−0.014 ± 0.07 vs. 0.037 ± 0.06; p = 0.045) was observed during the follow-up, and the integrated radius function showed a significantly greater reduction (−0.644 ± 0.04 vs. −0.232 ± 0.068; p = 0.006) (Table 2). Individuals with allergies exhibited significantly greater axial growth (0.155 ± 0.2 vs. 0.087 ± 0.09; p = 0.032) and a significantly smaller increase in the SSI (0.012 ± 0.1 vs. 0.070 ± 0.13; p = 0.010). Individuals with a predominantly ventral sleeping position showed an increase in A1 velocity (0.004 ± 0.02 vs. −0.002 ± 0.01, p = 0.046) and a greater increase in HC DArc length (0.013 ± 0.2 vs. 0.003 ± 0.02, p = 0.049) (Table 2).

The eyes with ocular surface symptoms exhibited a higher degree of myopization (−0.49 ± 0.8 vs. −0.17 ± 0.9; p = 0.05), an increase in A1 deflection length (0.037 ± 0.31 vs. −0.019 ± 0.15; p = 0.032) and A1 deflection amp (0.002 ± 0.01 vs. −0.018 ± 0.02; p = 0.031) and an increase in the TBI (0.057 ± 0.17 vs. −0.040 ± 0.15; p < 0.001). The eyes with eye-rubbing habits showed a higher degree of myopization (−0.41 ± 0.79 vs. −0.125 ± 1.0; p = 0.04), a smaller increase in radius (0.193 ± 0.69 vs. 0.432 ± 0.46: p = 0.018), and an increase in A2 arc length (0.007 ± 0.03 vs. −0.001 ± 0.008; p = 0.032) (Table 3).

Weekly hours of outdoor activity, weekly hours of physical activity, daily hours spent on digital devices, and daily hours spent reading/writing were correlated with the variation in several biomechanical parameters during the follow-up period (p < 0.05) (Table 4).

We found correlations between the variation in several biomechanical parameters and Δ_SE, Δ_AL, Δ_VCL, and Δ_R_VCL/AL during the follow-up period (p < 0.05) (Table 5).

4. Discussion

The present study aimed to examine whether ocular biomechanics could play a pivotal role in the relationship between the myopization process and individuals’ lifestyles in a cohort of European Caucasian adolescents. Additionally, it describes the changes in ocular biomechanics during adolescence and investigates the relationship between these changes and variables related to family history and behavioral factors.

To the authors’ knowledge, this is the first study to describe the variation in ocular biomechanic behavior during adolescence in a European population. Thus, it may represent an important contribution toward establishing a normative range for these changes in biomechanics, which could eventually become an integral part of the myopization process assessment in the future.

This sample included predominantly eyes within the European normative AL range [23] and with low refractive error from individuals with a mean baseline age of 14 years old, which enhances the relevance of the results in the context of school myopia appearing in otherwise normal eyes. Additionally, regarding the inter-individual and inter-eye analysis, the n in each group was homogeneous, allowing for more consistent conclusions from the results.

Through the lifestyle-related analysis, the present study demonstrates a concerningly high ratio between the high amount of time that adolescents spend using digital devices and the low amount of time that they spend on outdoor activities, and this may explain the founded high prevalence of ocular surface symptoms and eye-rubbing habits.

Since the evidence suggest that the emmetropization process may involve the scleral tissue as a key substrate [24,25,26], potentially mediated through dopamine pathways [27] under direct genetic influence [28], we consider ocular biomechanics as pivotal in understanding the underlying mechanisms. Its assessment has made significant advancements since the ORA [29] to evolve with the inclusion of a Scheimpflug camera [17], which enabled the visualization of corneal excursion and the accessory lateral movement of the corneo-scleral transition. As a result, biomechanical analysis is no longer binary—classifying the cornea as simply more or less rigid—but rather the description of a unique biomechanical fingerprint for each individual. Although a detailed explanation of each parameter is beyond the scope of this study, it is important to highlight some that can be important in the myopization setting: (1) IR is an important descriptor of total cumulative corneal deflection, representing the area under the curve of the movement function until it returns to its initial position [17]; (2) the SSI estimates the in vivo biomechanical behavior of a given cornea with normal topography, adapted to each moment of the non-linear stress–strain curve [30]; (3) SP-A1 and SP-HC are considered to be the parameters that best describe the global stiffness of the eye for a given biomechanically corrected IOP [18]; and (4) whole eye movement (WEM) describes the accessory anteroposterior movement of the eyeball measured in the lateral part of the 8 mm corneal image [22]. Theoretically, a stiffer behavior is associated with lower IR values and higher SSI values [30], as well as elevated SP-A1 and SP-HC parameters [18]. Additionally, WEM is as high as the amount of energy that the cornea cannot absorb, dissipating towards the posterior pole.

In the present study, the eyes of individuals with a family history of myopia in at least one parent showed a corneal excursion that became significantly narrower over the follow-up period. While this decrease was theoretically expected with age [31], the authors hypothesize that the difference in these eyes may be related to a potential greater dissipation of energy to the posterior segment, which could be a factor in ocular elongation. However, no significant differences were found in the variation in WEM, AL, or SE during the follow-up. The authors advocate for the value of studying this hypothesis in future research. It is important to state that the average refractive error at baseline was within the range of low myopia, which is believed to be less dependent on genetic load.

The association between allergic disease and keratoconus is well-established [32]. Recently, a possible causal relationship between allergic conditions and myopia has also been investigated [33]. In the present study, individuals with a history of allergies showed significantly greater axial growth associated with a smaller increase in the SSI. Since it theoretically increases throughout life, the smaller gain in stiffness in these individuals highlights the potential pivotal role of ocular biomechanical study in the already documented [34,35] relationship between allergic disease and myopia.

There is evidence supporting the effect of orthokeratology on myopia progression; however, it remains controversial whether a mechanical mechanism alone accounts for this effect. Although no data currently exist in the literature addressing this specific hypothesis, the authors sought to explore whether a potential external mechanical pressure on the sclera during sleep—the so-called “pillow effect”—could be associated with the differences in changes in ocular biomechanical behavior, axial elongation, or myopization during the follow-up. The eyes of individuals with a predominantly ventral sleeping position also showed a smaller gain in corneo-scleral stiffness during the follow-up, with even an increase in the A1 velocity. Although less consistent as a general descriptor of corneo-scleral stiffness, this first-generation parameter is one of the most important in describing how the cornea dissipates the air puff energy, highlighting a potential pillow effect on corneal tissue remodeling in line with what has been reported in the setting of ectatic corneal disease [36]. There were also significant differences in the variation in keratometry in these eyes; however no relationship was found with axial elongation or refractive status changes. In the eyes corresponding to the preferred sleeping side, differences were observed in the variation in K2; however, no significant differences were found in any biomechanical, biometric, or refractive parameters.

Regarding eye-based analysis, a greater degree of myopization was found in the eyes with ocular surface symptoms and those with reported eye-rubbing habits, in line with recent literature [37]. Additionally, the increase in deflection amp in A1 and in the TBI in eyes with surface symptoms, as well as with the smaller increase in radius and the increase in A2 arc length in rubbed eyes, all support a lower gain in stiffness in these eyes during the follow-up. Since eye-rubbing is often associated with dry eye symptoms, the literature is scarce regarding its hypothetical relationship with myopia. The authors believe that this study, in addition to describing this association, provides evidence that it may occur through changes in ocular biomechanics.

Regarding the study of correlations, all can be found in Table 4 and Table 5, but the discussion will be focused on those considered most important to reduce the potential noise of a broader analysis.

The present study demonstrates that the more time adolescents spent engaging in outdoor activities during the follow-up, the greater the decrease in HC time and the integrated radius function and the greater the increase in the SSI, indicating altogether a greater gain in corneo-scleral stiffness. Among the lifestyle factors traditionally examined, this remains the one with the strongest evidence supporting a protective role against myopia progression [7,10]. The present study adds to the existing body of literature and is the first to provide evidence that this protective effect may be mediated by the influence of solar radiation, potentially acting as a natural corneal cross-linking (CXL) mechanism, modulating the biomechanical properties of ocular tissues [38].

The evidence is scarce regarding the association between myopia and physical activity [8,9]. In the present study, physical exercise was associated with a decrease in maximum WEM during the follow-up. This is the first description of this association in the literature. The authors conceptualize that this result may highlight the role of sports practice in stabilizing the posterior pole of the eye in response to microtraumas of the ocular surface and, consequently, in reducing axial elongation. However, in the present study, physical activity did not correlate with variations in biometric or refractive parameters.

There is growing evidence linking myopia and the use of digital devices [39,40]. In the present study, the increase in hours spent on near-vision activities (reading/writing) correlated with a greater decrease in SP-A1 overtime. Again, this is the first report of this association, and the authors believe that this may support the emmetropization theory suggesting that near-vision activates feedback mechanisms through the dopaminergic pathway [41], reducing scleral stiffness and leading to an increase in AL. However, no correlations were found between this variable and biometric or refractive variations during the follow-up. On the other hand, the amount of time spent on digital devices did not correlate with any of the main biomechanical variables.

Regarding the relationship between lifestyle parameters and the variation in biometric and refractive variables, the present study also shows that the Δ_R_VCL/AL correlates negatively with the amount of outdoor activity and positively with the time spent on digital devices. As a disproportion in the vitreous cavity growth could be related with higher degrees of myopia [42,43], the authors believe that these findings can be understood as evidence of the pivotal role of tissue biomechanics as the substrate for the relationship between lifestyle and myopization. Additionally, it highlights the role of sum-of-segments biometric approach in the study of myopia progression.

Finally, analyzing the direct relationship between variations in biomechanical parameters and biometric and refractive changes, we found the most consistent results: first, reduced myopization correlated inversely with the integrated radius function and directly with the increase in the SSI; second, greater increases in AL, VCL, and R_VCL/AL correlated negatively with the increase in the SSI. Thus, a lower overall gain in ocular stiffness during the follow-up was consistently associated with increased myopic progression in the present study, both in refractive and biometric terms.

The main strength of this study is the fact that it prospectively analyzes the modification of ocular biomechanics across 2.5 years during adolescence and its hypothetical crucial relationship with lifestyle habits and myopia progression, including a sum-of-segments biometric approach. Moreover, it encompassed the period of the COVID-19 pandemic, a period that markedly influenced lifestyle patterns known to contribute to myopia development.

The results obtained in this population with low refractive error and above childhood provide evidence for the concept of school myopia and gain greater relevance due to their potential for amplification in younger populations or those with higher degrees of myopia.

The main limitations are: (1) the inclusion of subjects of different ages, although no significant differences were found in comparative tests; (2) the execution of non-cycloplegic refraction due to restrictions imposed by the hospital’s ethics committee. However, this may add value to the results, as the greater accommodative ability at baseline could have masked some degree of myopization during follow-up; and (3) the data regarding lifestyle have some degree of subjectivity that is difficult to overcome in scientific reports.

5. Conclusions

This study provides novel evidence that ocular biomechanical changes during adolescence are closely linked to myopia progression and influenced by lifestyle factors. A lower gain in corneo-scleral stiffness was consistently associated with greater axial elongation and refractive myopization, while outdoor activity appeared protective, and near-work and ocular surface symptoms and eye-rubbing showed adverse effects. These findings suggest that ocular biomechanics may represent the physiological bridge between the environmental exposure and myopization, highlighting their potential role in future strategies for early detection and prevention of myopia.