Analysis of the Correlations between Relative Peripheral Refraction Errors and Myopia in Children and Adolescents

Myopia has become a public health problem worldwide, with a global prevalence of approximately 30%, which is projected to rise to 50% by 2050, and high myopia is expected to increase by 10% [1]. A recent multicenter study showed [2] that from 1998 to 2022, the prevalence of myopia in China increased significantly, with the highest rates in urban areas and among high school students; rates of high myopia continue to climb, with the rate among 16–18-year-old adolescents projected to rise from 7.3% in 2001 to 22.1% by 2050, leading to a continuous increase in the risk of related blindness.

However, the exact pathogenesis of myopia remains unclear. Relevant theories include the lag of accommodation [3], peripheral defocus [4], neurotransmitter-related theory [5], and scleral hypoxia theory [6]. Among these, peripheral retinal defocus is considered to play a significant regulatory role in the refractive development of the eye and may be an important mechanism affecting the onset and progression of myopia [7]. This study utilizes multispectral refraction topography (MRT) to measure the peripheral refractive status in children and adolescents. MRT primarily uses monochromatic light of different wavelengths to sequentially capture fundus images. Through extensively developed computer algorithms, it performs comparative analysis of the multispectral images after lens compensation, calculates and summarizes the actual refractive values of each pixel, and then draws the corresponding topography. This study employed MRT to measure the relative peripheral refractive error (RPRE) in children and adolescents of Jinhua City and explored the relationship between RPRE and myopia progression.

This study adhered to the tenets of the Declaration of Helsinki and was approved by the Clinical Research Ethics Committee of Jinhua Eye Hospital (Approval No. 2509261408747). Written informed consent was obtained from parents or guardians after fully explaining the purpose and content of the study, followed by ocular examinations.

Relative peripheral refractive error refers to a refractive state of the peripheral retina relative to the macular center retina. Smith et al. [7, 8] found that visual signals fed back by the macular fovea are not essential for the refractive development of the eye. In the case of foveal ablation, the peripheral retina of the eye can still receive visual signals and regulate the refractive development of the eye relatively independently. The peripheral refractive state is more critical than the foveal refractive state.

Multispectral refraction topography (MRT) is an optical system based on a multispectral fundus camera. By fixating on a stationary target, it can capture dozens of fundus images and measure millions of data points within seconds, meeting the need for rapid measurement [9]. MRT has been proven to be an objective method for measuring RPRE because it shows good repeatability of refractive measurements [10] and high measurement consistency compared to automated refractors [11]. This study used MRT technology to analyze the status of RPRE in different eccentricities and quadrants of the retina in children and adolescents, and for the first time included high myopia in children and adolescents in the study.

Atchison et al. [12] used magnetic resonance imaging (MRI) technology to study the retinal shape of emmetropic and myopic eyes in individuals aged 18–36 years and found that the retinal shapes of both emmetropic and myopic eyes were flat. As myopia increased, the flatness of the retina decreased, the size of all ellipsoids increased, and the increase in axial size was greater than that in the vertical direction, which in turn was greater than that in the horizontal direction (the increase ratio was approximately 3:2:1); this also revealed that the shape of the peripheral retina may be related to the progression of myopia. Hoogerheide et al. [13] found that 77% of pilots with emmetropic or hyperopic peripheral refraction developed myopia, while only 6% of those with myopic peripheral refraction developed myopia. Mutti et al. [14] evaluated the refractive status, AL, and RPRE of children with myopia and emmetropes before, at, and after the onset of myopia, and found that children with myopia had more hyperopic defocus in RPRE than emmetropes from 2 years before to 5 years after the onset of myopia, and the hyperopic defocus increased the most 1 year before the onset of myopia. Therefore, it was suggested that RPRE may be an important indicator for predicting the occurrence of myopia, and indicates that hyperopic defocus in RPRE is a risk factor for the development of myopia.

Our study found that children with different refractive states overall exhibited myopic defocus in TRDV. The TRDV values for the different refractive groups were as follows: H group (−0.41 ± 0.48) D, E group (−0.25 ± 0.42) D, LM group (−0.04 ± 0.38) D, MM group (−0.10 ± 0.34) D, and HM group (−0.07 ± 0.37) D. Compared with hyperopic and emmetropic eyes, myopic eyes showed a reduction in myopic defocus and an increase in hyperopic defocus. This precisely indicates that the eyeball morphology is flattened, and as myopia increases, the flatness of the retina decreases. In contrast, in a study on the internal balance of myopia and axial growth, Wallman et al. [4] pointed out that for myopic eyes, the elongation of the posterior part of the eyeball results in a prolate ellipsoidal shape, with the peripheral retina being in a hyperopic state, whereas hyperopic eyes are oblate in shape, with the peripheral retina being in a myopic state. Sng et al. [15], in a study on the relationship between peripheral refraction and central refraction in Chinese children living in Singapore, found that children with central myopic refraction had relative hyperopia in the periphery, while children with central hyperopic refraction had relative myopia in the periphery.

The discrepancy in research findings may be related to differences in measurement methods. Previous studies often measured RPRE at only one or several points within the peripheral retinal eccentricity range of 15°–45°, whereas our study measured data from 121 points across the full retinal eccentricity range of 0°–53° using MRT, providing more comprehensive data. Another possible reason is that most studies measuring RPRE were conducted under cycloplegia. The size of the pupil affects the amount of light entering the eye: a larger pupil allows more peripheral light to enter, which may lead to a more complex distribution of retinal defocus. In contrast, our RPRE measurements were performed under natural pupils using MRT. While the natural pupil state more closely aligns with daily living environments and can better explain the relationship between RPRE and myopia progression, differences in measurement results may arise due to smaller pupil size and the involvement of accommodation. A study comparing peripheral refraction before and after mydriasis using MRT showed that after mydriasis, peripheral refraction exhibited a significant hyperopic shift with increasing eccentricity, with peripheral refraction becoming more hyperopic than central refraction. The high myopia group demonstrated a greater hyperopic shift in the peripheral regions [16]. This further indicates that cycloplegia and pupil dilation affect RPRE, explaining why RPRE measured under natural pupils shows a reduced hyperopic shift, which is also the reason for the lower hyperopic defocus observed in our measurements.

Sng et al. [15] further conducted a comparative study of patients with different degrees of myopia and concluded that eyes with moderate and high myopia exhibited relative peripheral hyperopia at all eccentricities, while eyes with low myopia showed relative peripheral hyperopia only at eccentricities exceeding 30°. The results of a study measuring RPRE by MRT showed that in Chinese young adults aged 18–28 years, RPRE increased with increasing eccentricity. Patients with high and moderate myopia had relative hyperopic defocus in all directions, while patients with low myopia and emmetropia had relative hyperopia only at eccentricities exceeding 30° and 35°. RPRE between 20° and 35° may be closely related to refractive development and eyeball growth [17]. Another study found that in children aged 5–18 years, under cycloplegia, the refraction difference value (RDV) at retinal eccentricity of 30°–45° measured by MRT may be closely related to the development of myopia, and compared with emmetropia and low myopia, the RDV value of moderate myopia showed the highest hyperopic defocus [18]. However, our study found that the differences in RDV15–30, RDV30–45, and RDV45–53 among different refractive groups were statistically significant; both the hyperopia group (H) and the emmetropia group (E) showed myopic defocus in RDV15–30, RDV30–45, and RDV45–53, and as the eccentricity increased, the amount of myopic defocus increased; in the low myopia group (LM), as the eccentricity increased, the myopic defocus decreased, RDV15–30 (−0.11 ± 0.18) D, RDV30–45 (−0.04 ± 0.43) D, RDV45–53 (0.07 ± 0.73) D, and PDV45–53 showed hyperopic defocus.

Different studies have reported varying findings regarding RPRE in hyperopic and myopic eyes. However, most research indicates that relative hyperopic defocus at retinal eccentricities beyond 30° is associated with myopia progression, a result consistent with our study. Furthermore, we observed that for RPRE in the peripheral retinal regions RDV15–30, RDV30–45, and RDV45–53, as retinal eccentricity increases, myopic and non-myopic eyes exhibit opposite directional changes in RPRE: non-myopic eyes show an increase in myopic defocus, whereas myopic eyes show a decrease in myopic defocus and an increase in hyperopic defocus. This trend is particularly pronounced in the low myopia group, which may be related to the age range of low myopia subjects (10.85 ± 2.14 years), a period characterized by the most significant changes in refractive power and axial length. This is supported by a study investigating the effects of age and myopic changes on retinal development, which indicated that myopic refractive shifts and axial length growth are most significant in children under 12 years of age. Retinal layer thickness at the fovea increases in children aged 6–10 years, but decreases in adolescents over 13 years old. Moreover, the change in retinal thickness diminishes with age. In children who experience a myopic shift before the age of 9, the thickness of the most peripheral retina and peripheral retinal layers increases less or decreases more [19]. However, a study by Hu et al. [20] using MRT under natural pupils in children aged 10–13 years showed significant differences in RDV15–30 and RDV30–45 between myopic and emmetropic eyes, and these differences were negatively correlated with SE—that is, higher myopic refractive error was associated with reduced myopic defocus and increased hyperopic defocus. The discrepancy in results may be related to differences in the age distribution of the study populations. Several MRT-based RPRE studies are summarized in Table 4.

RPRE relative peripheral refractive error; MRT multispectral refractive topography; RDV refraction difference value; B: before cycloplegia; A: after cycloplegia; LH: low hypermetropia; E: emmetropia; LM: low myopia; MM: moderate myopia; HM, high myopia.

Ehsaei et al. [21] showed that in myopic eyes, the nasal-temporal retinal shape is asymmetric, and the temporal retinal shape is steeper than the nasal one, but the superior-inferior retinal shape difference is not significant. An MRT study of children aged 10–13 years under natural pupils showed that there were significant differences in RDV15–30, RDV30–45, RDV-S, and RDV-T between myopic and emmetropic eyes, and these differences were negatively correlated with SE, indicating that higher myopic refractive error was associated with decreased myopic defocus and increased hyperopic defocus [20]. However, our study found that there were significant differences in the nasal-temporal and superior-inferior directions among different refractive groups. RDV-I showed relative hyperopic defocus compared to RDV-S, and RDV-N showed relative hyperopic defocus compared to RDV-T. Moreover, the RDV-I and RDV-N of low myopia were 0.23 ± 0.57 D and 0.18 ± 0.68 D, respectively, showing more hyperopic defocus. This indicates that the nasal-temporal and superior-inferior retinal shapes are asymmetric across different refractive statuses, and myopic eyes exhibit hyperopic defocus in RDV-I and RDV-N. Such changes are more pronounced in the low myopia group. RDV-I and RDV-N may be associated with myopia progression; however, findings vary significantly across different studies. Further longitudinal investigations are warranted to clarify the relationship between RPRE in different retinal quadrants and the progression of myopia.

Our study found that SE was negatively correlated with TRDV, RDV15–30, RDV30–45, RDV45–53, RDV-S, RDV-I, and RDV-N (P < 0.001), although the linear correlations were weak (β < 0.2). The relationship between the refractive status of the peripheral retina and myopia is controversial. Current research cannot clearly determine whether peripheral retinal defocus is the cause or the result of myopia. Mutti et al. [22] posited that before the onset of myopia in children, their peripheral refractive status is usually relatively myopic, not hyperopic. Only after the onset and progression of myopia does the peripheral refraction gradually change to hyperopic defocus. This finding suggests that peripheral hyperopic defocus is not a risk factor for the occurrence of myopia, but a result after the occurrence of myopia. Qi et al. [23] studied the peripheral refraction and relative peripheral refraction characteristics of Chinese male children without myopia aged 14–16 years in the Air Force experimental class. The results showed that there was no significant difference in relative peripheral refraction between myopic and non-myopic eyes at baseline, but after 2 years, the relative peripheral refraction of myopic students turned to hyperopia, while that of non-myopic students remained relatively myopic or hyperopic. Atchison et al. [24] conducted a 2-year longitudinal follow-up study of emmetropic, hyperopic, and myopic children and found that in children who developed myopia, relative peripheral retinal hyperopic defocus was associated with the progression of myopia, but emmetropic children who showed hyperopic defocus remained emmetropic in the follow-up study. Compared with emmetropic children, the difference in baseline relative peripheral refraction (RPR) in children who developed myopia was not statistically significant. In addition, it was found that in children with baseline RPR showing myopic defocus, after the onset of myopia, RPR changed from myopic defocus to hyperopic defocus. Relative peripheral retinal hyperopic defocus could not predict the occurrence and development of myopia. Therefore, we believe that changes in RPRE in children and adolescents are not only associated with refractive error, but that other factors such as age, retinal development, the rate of refractive change, and accommodation must also play a role.

Our research has several limitations. First, our study used MRT technology under natural pupils to analyze the status of peripheral retinal defocus in children and adolescents. Due to the influence of accommodation and pupil size, the measured relative peripheral refraction was lower compared with other studies, and the effect sizes were relatively small. Second, the study subjects were children and adolescents aged 7–15 years who had established refractive development records at Jinhua Eye Hospital. The younger age group (7–12 years) consisted mainly of emmetropic and low myopic eyes, while the older age group (13–15 years) showed lower compliance in maintaining refractive development records due to academic demands. Consequently, certain challenges were encountered in recruiting sufficient samples for the hyperopia group (n = 191), moderate myopia group (n = 270), and particularly the high myopia group (n = 60), resulting in relatively small sample sizes in these categories. Therefore, it is important to note that the imbalanced sample sizes across groups may have diminished the statistical power of our analysis, potentially biasing the assessment of RPRE in the hyperopic, moderate myopic, and high myopic subgroups. Therefore, further research with larger and more balanced cohorts, particularly incorporating more hyperopic and moderate-to-high myopic eyes, is warranted to confirm the relationship between refractive error and RPRE.

In summary, RPRE differs significantly across various refractive statuses. Compared with hyperopic and emmetropic eyes, myopic eyes exhibit a reduction in myopic defocus and an increase in hyperopic defocus in RPRE, with these changes being more pronounced in low myopia. This suggests that RPRE is a relevant factor in the progression of myopia, rather than merely a consequence of eyeball growth. However, the variability in results and the presence of conflicting evidence indicate that other factors—such as age, retinal development, the rate of refractive change, and accommodation—also play essential roles. Thus, it cannot yet be concluded that RPRE is the decisive factor influencing refractive status. More longitudinal research data are needed to clarify the relationship between RPRE and myopia.