Nowadays, myopia is the most prevalent refractive error and its increase is considered an important cause of concern in the population (1–4). In fact, over the last decades, has undergone a very significant increase in prevalence worldwide, especially in Asia. (5). Globally, it is estimated that almost 50% of the population will be myopic by 2050 (6) and according to World Health Organization (WHO) estimations, 2.8% of the general population had high myopia (HM) in 2010, and this percentage could reach 10% by 2050 (6–11).

Regarding children, the highest prevalence of myopia using cycloplegic refraction examinations was reported in East Asia and Singapore (60%), while in Europe the prevalence rates were lower (40%) (2, 12). Being the most sensitive age of myopia growth between 7 and 10 years old when it increased 5.1 times (from 4.5 to 23.0%), 15–18 years increased 2.1 times (from 21.5 to 45.0%), stabilizing from the age of 24, except in high myopic patients (13).

The relevance of this data is that early-onset myopia generally progresses through adulthood, leading to an increase in HM prevalence, which is a well known risk factor for the development of other ocular pathologies, such as glaucoma, cataract, retinal detachment or myopic maculopathy (14–16). Because of that reason, the prevention of myopia and HM has been considered by the WHO as one of its five priorities in the global prevention of blindness (17).

The complexity of myopia results from the fact that it is not caused by a single etiological factor. Some risk factors are well known, particularly genetic and environmental factors, although it has not been possible to determine the degree of influence of each factor in each patient (16, 18–20). Among the modifiable factors, time spent doing outdoor activities (OA), as a protective factor, and near vision activities (NVA), as a risk factor, are clearly the most important (16, 20–29).

The most frequently employed methodology in epidemiological studies for determining an individual’s time spent doing OA is the use of questionnaires (10, 11, 30, 31). However, this data may not be accurate due to recall bias, inaccuracy of questions or errors of interpretation (10, 21, 32), which may result in overestimation or underestimation of the actual or past time spent outdoors. Given the shifting population habits and the necessity for conducting objective research on myopia and its associated risk and protective factors, it is clear that an objective biomarker is needed. Recently, CUVAF (Conjunctival Ultraviolet Autofluorescence) area measurement, has been studied for this purpose (33). It was first described as a localized-area of autofluorescence under ultraviolet (UV) light in the bulbar conjunctiva (34–36), secondary to sunlight exposure (37–39). To detect CUVAF-area, a special custom-built device was designed initially, and subsequently, Lingham et al. validated the autofluorescence mode of the Heidelberg Spectralis HRA + OCT (optical coherence tomography), a device commonly used in ophthalmology clinics for other purposes (34, 40). CUVAF is based on the premise that conjunctival components may emit visible fluorescence as their structure is altered by UV exposure. Changes in intracellular protein content in endogenous cellular components, such as lysosomes, mitochondria, cytokines, growth factors and matrix metalloproteinases (MMPs), are also implicated in the pathogenesis of pterygium (39, 41).

Previous studies conducted by our research group with university students have demonstrated a negative correlation between the spherical equivalent (SE) of the subjects and the amount of time spent doing OA in myopic subjects (10, 42). This relationship has also been observed in children, with the CUVAF area increasing with greater sun exposure (taking into account the amount of exposed skin), and also increasing with age (43). Additionally, there is an inverse correlation between the individual’s refractive error (RE) and the CUVAF-area (10, 44). These results have been corroborated in a meta-analysis, which aimed to provide a comprehensive summary of the relevant evidence pertaining to the association between the CUVAF-area and myopia across different geographic regions and age groups (33).

To date, studies in this field have been conducted primarily in adults and young adults, however, the utility of CUVAF as an objective biomarker of myopia and its risk/protective factors in the pediatric population remains poorly understood. The objective of this study is to compare the CUVAF in children with and without myopia in order to validate its usefulness as a protective biomarker of childhood myopia.

The growing importance of myopia due to the rapid increase in its prevalence (7, 47, 48) and its possible severe visual impairment (14, 15, 49), makes it necessary to further study its etiopathogenesis and the location of biomarkers that help us to control and monitor it, especially in the pediatric age group (15, 50). In this sense, our results show that CUVAF is a useful biomarker for exposure to OA, as well as its inverse relationship with the degree of myopia, confirming its usefulness also in the pediatric population in both aspects. In addition, a direct relationship between CUVAF and age is observed, being more frequent its appearance after 8–10 years of age, presenting a greater discriminative capacity in the second decade of life with an increase in its specificity and sensitivity. In this regard, a dynamic cut-off point could be established according to the age of the subjects by means of the formula explained above in the results, which implies a progressive increase of the CUVAF area throughout life, obtaining additional information that could help us to establish the risk of patients with myopia.

The rise in myopia prevalence cannot be attributed exclusively to genetic factors, indicating that environmental influences also play a crucial role in its development (51). Although NVA contribute to this trend (21–23, 52), the most significant modifiable risk factor is the reduction of OA, as one hour of OA reduces the risk of myopia by 13% (1). Most studies show the dose–response protective effect of OA on myopia onset (1, 5), and its influence on progression (6). In this sense, our study showed no significant differences in NVA between groups, as other age-group studies (6), but did find that the control-group engaged significantly more hours of OA compared to the myopic-group, particularly in the HM group, underscoring the protective effect of OA. It is well known that various ocular components undergo growth and maturation in younger children and the ocular growth patterns may be more sensitive to environmental influences during this period (1). In terms of NVA, the SAVES study describe that they may have an impact on the early development of myopia, but only in young children (10, 53).

Regularly, the relationship between OA and the degree of myopia has been measured by lifestyle questionnaires which are prone to recall bias, making them an inaccurate method (41). This situation is exacerbated in the pediatric population, since these questionnaires must be filled out by a third person (legal-guardians) due to the age of the patients. Consequently, it is important to evaluate CUVAF as a biomarker for OA at the early age of myopia onset, as most studies have focused on older populations. In our study both questionnaire data on OA hours and CUVAF-area show statistically significant differences between controls and myopia groups. However, the measurement of CUVAF-area showed bigger statistical significance and more sensitivity. In this way, the results showed that having CUVAF-area protect to myopia 2.5 times, being almost 5 times the protection in case of HM. Additionally, when correlating these two measures, OA hours an CUVAF-area, although no statistically significant correlation was found, a clear trend was observed supporting the theory that the biomarker is more objective and more powerful predicting a lower risk of myopia development.

Since the direct relationship of CUVAF with OA is also observed in children, it is reasonable to consider its usefulness as a protective marker for myopia in this age group (34, 39, 48). Our results show significant differences between both groups, with the CUVAF-area being significantly higher in the control-group than in the group of myopic patients. A statistically significant difference is obtained when we separate the groups according to the degree of myopia, demonstrating how the CUVAF-area decreases as the degree of myopia increases. In the clinical practice these changes in the relative risk could play an interesting role, the simply describing the presence or absence of CUVAF could help us identify the children at higher risk that could benefit from a closer follow up and early treatment.

An interesting observation is that this relationship increases in the subgroup of children older than 12 years, which could imply that CUVAF may have a certain dynamic or cumulative component, or that this increase may be due to the difference in behavior of children of different ages, being more exposed to OA from the age of 12 years.

In our environment, we are constantly exposed to UV radiation, and its relationship to changes in the ocular surface is well known (54), as well as the alterations of proteins which provide these cells the ability to autofluorescence when excited by UV radiation (41, 54). Some researchers have observed that the CUVAF-area increased in intensity and surface-area, especially in subjects who do a lot of OA, confirming the theory of the cumulative effect of UV radiation (55). These data, which have been verified in different regions of the world (33), show the usefulness of CUVAF for the follow-up of ocular pathologies and adherence to lifestyle changes in patients (54–58).

Axial length is a primary factor linked to myopia development, and therefore, numerous studies examining CUVAF in both adults and children have sought to correlate CUVAF-area with axial length. However, to our knowledge, including the results of this study, none have found a significant correlation, despite a notable correlation with spherical equivalent. The absence of this correlation with axial length may be due to its dependence on several factors such as height, sex and age of the subjects. In our study, an adjustment for age and a ratio (AL/CCR) has been performed, with the intention of eliminating these confounding factors, thus obtaining a partial CUVAF-AL correlation (adjusted for age) and significant in the case of the ratio AL/CCR.

This has led to the hypothesis that light exposure influences not only axial length but also the balance of optical power during emmetropization. Consequently, CUVAF has correlated not only with axial length but also with corneal curvature radius and an index reflecting the relationship between these parameters (AL/CCR). Interestingly, this correlation between CUVAF and AL/CCR is statistically significant; suggesting that sunlight exposure indeed affects the balance between ocular optical power and axial length, encompassing both refractive and axial myopia origins. It could be interesting to evaluate this parameter in a larger sample, and even in adult population, since its possible impact has not been previously analyzed according to the current bibliography.

As the results of this study have shown, the implication of genetics in the development of myopia is clear, especially in those patients with an early onset of the disease. However, the results also manifest that in older pediatric patients (over 12 years of age) the environmental component plays a very important role that must be taken into account.

It could be hypothesized that parents with myopia have a lifestyle that favors this increase in myopia (more NVA and few OA), and that this is transmitted to their children, thus favoring an increase in myopia with a certain environmental, rather than genetic, character. Nevertheless, the data obtained have shown no significant differences in the size of the CUVAF-area according to family history (Supplementary Figure S1), which leads to the conclusion that the habits and lifestyle of the patients are not necessarily due to the existence of a family history of myopia.

The main limitation of this study is that most of the subjects were obtained from the Department of Ophthalmology, hence there are fewer controls and more myopic children. On the other hand, the strengths of this study are its large sample size, as well as the homogeneity of the sample in terms of geographical and socioeconomic status, so NVA and OA were the only differential variables regarding environmental risk factors. Even so, it would be necessary to do more studies with a larger sample size and in different environments.

The results of this study prove that CUVAF is a useful biomarker for exposure to OA, as well as its inverse relationship with the degree of myopia, confirming its usefulness also in the pediatric population in both aspects, especially in the second decade of life. In addition, a direct relationship between CUVAF and age is observed, being more frequent its appearance after 8–10 years of age, presenting a greater discriminative capacity in the second decade of life with an increase in its specificity and sensitivity. Future works are needed to consolidate the capacity of this biomarker to differentiate which patients may have a higher risk of developing myopia during the growing age, being able to provide individualized treatments based on that risk, as well as to assess adherence to the behavioral measures that may be proposed in the clinical practice as a first line of action for the treatment of childhood myopia.