The refractive status of the eye typically changes throughout an individual's lifespan.1 During early childhood, most children exhibit hyperopia, which usually decreases during growth years, leading to emmetropia in adulthood. The elongation of ocular axial length usually maintains a delicate equilibrium with decreases in crystalline lens and cornea refractive power to preserve emmetropia.2 Although the cornea contributes two-thirds of the entire refractive power of the eye, the changes in corneal power are limited following a rapid reduction within the first two years of life.3 In contrast, during childhood growth, the lens undergoes compaction and remodeling of fiber cells in the lens interior, resulting in decreased axial thickness and a flattening and reduction in optical power.4,5

The crystalline lens comprises of five distinct layers, with a prominent cortex and nucleus surrounded by delicate cysts.6 Although the refractive power of the lens and its various lenticular components cannot be directly measured in vivo, this power is highly correlated with its morphology and can be calculated using a specific formula.7 Among the lenticular components, thickness and curvature play major roles in determining the refractive power during accommodation.3 Previous studies have reported significant alternations in lens curvature and radii during accommodation or cycloplegia, the latter of which represents a relaxed or low accommodative state.3,8 However, further analysis is needed to investigate the impact of accommodation or cycloplegia on lenticular components.

Recent studies have investigated the effects of accommodation lag on myopia progression.9,10 The Helmholtz accommodative theory proposes that the ciliary muscle plays a crucial role in accommodation.6,11 Relaxed ciliary muscle increases zonule tension to apply force on the lens, resulting in lens stretching and reduced lens curvature.12 Cycloplegic refraction is considered the gold standard for myopia evaluation due to the robust accommodation function of children.13 While previous studies have used cycloplegic eye drops to demonstrate the significant morphological alterations of the lens and the changes in lenticular refractive power, how the thicknesses of the main lens components change in accommodation or cycloplegia remains unknown.14

Optical coherence tomography (OCT) is a rapid, noncontact, and precise technology for in vivo and real-time biometric measurements.15 Depending on optical pixel density, OCT can be used to analyze the thicknesses of the three major components of the cataractous lens.16

In this study, we evaluated cycloplegia-induced alterations in the thickness of crystalline lens components in children aged 5–15 years and their potential correlations with ocular metrics. Additionally, we compared the baseline and altered values among three refractive statuses, including non-, low- and moderate-myopia.

In this study, we elucidated the altered thickness of crystalline lens components following tropicamide-induced cycloplegia in children aged 5–15 years, which was primarily attributed to changes in the lens nucleus rather than the cortex. Moreover, the changes in the posterior cortex thickness after complete cycloplegia may be negatively associated with PD (P = 0.042, Table 4). In addition, our findings demonstrated a significant correlation between SE and the thickness of specific lens components.

Based on the theory of OCT techniques and analysis of optical transparency, sweep-source OCT accurately stratifies the cataractous lens.16 Meanwhile, Brillouin microscopy effectively divides the lens into three layers based on its distinct biomechanical properties.17,18 Therefore, in this study, we initially categorized the lens into three major components: the anterior cortex, nucleus, and posterior cortex.

The mechanical properties of the lens cortex and nucleus are crucial factors influencing the lens's ability to change shape.19 Consistent with previous studies, our results revealed reduced LT and LNT following tropicamide-induced cycloplegia in both non- and low-myopia groups.20 Additionally, other studies have suggested that changes in nuclear thickness accounted for approximately 70 % of the total change in lens thickness in adults aged 29 years.21,22 Despite the variation in age range, we supposed that the cycloplegia-induced alteration in crystalline lens thickness can be attributed to changes in the nucleus. In contrast, the moderate-myopia group in the present study demonstrated no significant changes, possibly due to lower crystalline lens rise and thinner lens in higher myopic participants, as reported previously.23 Our findings further indicated differences between the non-myopia and both myopia groups in children aged 5–15 years, in which higher degrees of myopia were associated with thinner LT; however, low- and moderate-myopia eyes showed no significant differences. Moreover, we identified a significant correlation between lenticular thickness and other metrics including age and SE. With increasing age and myopia degree, the lens and lens nucleus thin, consistent with the observations reported by previous study in healthy adults.24

In this study, we did not observe alterations in ACT and PCT following cycloplegia across all three refractive groups, consistent with the findings reported by previous research.25 However, the magnitude of change in the cortex differs considerably among studies due to the use of different measurement methods. Using Scheimpflug imaging, Dubbelman et al. reported similar results, whereas Pour et al. found that the cortex accounted for 30 % of the thickness change in their analysis of finite element models.22,26 Furthermore, discrepancies in lens thickness changes may be attributed to an unclear definition of the cortex. The Hockwin system, which delineates 10 distinct turning points to indicate the extent of single lens layers, suggests a decrease in cortical thickness; however, the Oxford system, characterized by 8 different densitogram zones, indicates an increase in thickness during accommodation.26,27 These differences may be attributed to the varying methods of layer identification. In this study, we utilized OCT technique detailed images of the crystalline lens, providing new insights into the changes in component thickness. This approach differs from that of Scheimpflug imaging but aligns with previous classifications using similar OCT techniques.16,25 Our results also revealed a significantly negative relationship between ACT and PD before mydriasis. However, after cycloplegia, when all participants maintained the same accommodative status, ACT was not significantly correlated with PCT or PD.

In addition, ACT and PCT differed between the non- and low-myopia groups, in which the higher the degree of myopia, the thinner the cortex, and the older the children. Our results align with the general hypothesis regarding the balance between ocular axial length elongation and decreased lens refractive power.28 However, our findings contrast with previous studies, which reported a linear increase in cortex thickness with age in participants aged 4–80 years.4,29 According to Oxford system, the cortical zones could be divided into zone C1, C2 and C3.26 Previous studies have demonstrated that in the first decade there were only two distinct cortical zones, C1 and C2, and any decrease with age was due to compaction in zone C1.29 The participants in the present study had an average age of 9.5 years, therefore, the cortex thinning age can be attributed to this specific developmental stage.

Previous studies have confirmed cycloplegia-induced changes in ACD.30 Our results were consistent with those findings: after cycloplegia, ACD increased greatly, which could be attributed to the backward movements of the lens, and we also identified a negative relationship between ACD and LT.23,31 Furthermore, our results indicated that deeper ACD was associated with thinner lens and cortex, consistent with our previous Brillouin results.18 Similar to prior studies, we found that CCT significantly increased after cycloplegia across all three refractive groups.23,32 However, Zhang et al. reported no change in CCT after cycloplegia by tropicamide.20 These discrepancies may be due to the effect of corneal rehydration and the resolution limitation of OCT.33

Our study has several limitations. First, we focused on the components of the crystalline lens in eyes with different refractive error statuses while acknowledging that both lens morphology and refractive status are highly associated with age and may result in significant discrepancies in intragroup comparisons. Second, previous research has illustrated sex-based discrepancies in ocular metrics.34 However, in this study, the effect of sex on cycloplegia-induced alternation of the crystalline lens was statistically insignificant, which requires further investigation. Last, the use of five drops of 0.5 % tropicamide administered at five-minute intervals with a final waiting period of twenty minutes has been widely accepted for mydriasis in the Chinese pediatric population, this protocol is considered sufficient for achieving full mydriasis.13,23 Alternatively, commercially available 1 % cyclopentolate hydrochloride, following the standard protocol of two drops spaced fifteen minutes apart and an additional fifteen-minute waiting period, may also be a viable option.35

Our findings suggest that after cycloplegia, the lens nucleus plays a crucial role in the changes in the whole lens, while neither the anterior nor posterior cortexes change significantly, in children with non- and low-myopia. Additionally, SE was positively correlated with the thickness of the crystalline lens nucleus and the total lens thickness.

The study was approved by the Ethics Committee Office of the Eye and ENT Hospital of Fudan University. All participants provided informed consent to participate in the study (Ethics No.: 20200530).

Written informed consent was obtained from the patients for the publication of this paper. Patients’ names are not applicable.

Study conception and design: Jing Zhao and Jian Cao; Data acquisition: Yongle Bao, Zhirong Zhang, Zhanying Wang and Jing Zhao; Data analysis: Jian Cao and Zhirong Zhang; Data interpretation: Lingling Niu, Yong Ma and Yanze Yu; Manuscript drafting: Yongle Bao, Zhirong Zhang, Jian Cao and Jing Zhao; Manuscript revision: Xingtao Zhou, Xiaoying Wang and Jia Huang; Final approval: Jing Zhao. All authors read and approved the final manuscript.