The efficiency of 0.01% topical atropine on the myopia prevention in low refractive error population with different refractive errors: a retrospective cohort study

Abstract

Background

The aim of this study was to investigate the effectiveness of 0.01% topical atropine (ATR) on myopic prevention in a population with different low refractive errors.

Methods

A retrospective cohort study was conducted on children with low refractive error. They were classified according to the application of 0.01% ATR, and 55 and 67 eyes constituted the ATR and non-ATR groups, respectively. The primary outcomes were changes in spherical equivalent refraction (SER) and axial length (AXL) after a one-year interval. Statistical analysis was carried out using the independent T test and generalized linear model.

Results

The SER change and AXL elongation were significantly lower in the ATR group compared to the non-ATR group (P < 0.001 for both). For different initial SERs, SER and AXL progression in the individuals with plus initial SER were significantly lower than those with minus initial SER after ATR treatment (both P < 0.05). High outdoor activity was inversely correlated with myopic progression in the ATR group (P = 0.006). For the non-ATR population, young initial age, minus initial SER, and high near-work activity were related to myopic progression, while high solar exposure and high outdoor activity were inversely correlated with the development of myopic progression (all P < 0.05).

Conclusions

Application of 0.01% ATR might relates to myopic prevention effect in a population with low refractive error, especially for plus initial SER.

Introduction

Myopia is a growing medical issue in many countries, with a prevalence that could reach 75% in some Asian countries [1]. Myopia development features both long axial length (AXL) and steep corneal curvature, where long axial length is the main pathophysiology of acquired myopia [2, 3]. High myopia, which is mostly defined as a spherical equivalent refraction (SER) larger than − 6.00 diopter (D), would contribute to the development of retinal pigment epithelium disruption, retinal degeneration, and optic nerve damage [4]. Several co-morbidities associated with high myopia, involving retinal detachment as well as myopic choroidal neovascularization, could contribute to pronounced visual impairment [3].

Because of the negative effects of myopia on ophthalmic health, multiple approaches have been applied to prevent myopic deterioration [5, 6]. Instilling eyedrops of topical atropine (ATR) may be the most widely used myopia intervention in all countries [7]. In previous studies, ATR administration could significantly reduce SER and AXL progression [8, 9]. Except for topical ATR eyedrops, the orthokeratology contact lens has also been administered to reduce myopia progression with favorable efficiency and safety [1, 10]. The dual-focus contact lens is a daily disposable contact lens that was introduced for myopic management more than three years ago [11], with a favorable myopic control outcome regarding SER and AXL [12, 13]. In addition to the above three approaches, the defocus incorporated multiple segments’ spectacle lens had been introduced into clinical usage in the 2010s, with a satisfactory myopic control outcome [14].

Except for myopic control in the population with existing myopia, preventing myopia development in non-myopia children has been an important issue in recent years [15]. Low-concentration topical ATR has been administered in children with low refractive error to prevent myopia occurring [16]. Still, the various effects of low-concentration ATR on children with different low refractive errors have not been evaluated. In addition, lifestyle factors like near-work or outdoor activity may affect the preventive effects of low-concentration ATR on evaluating myopia needs.

As a consequence, the objective of this research was to evaluate the myopic prevention effect of 0.01% ATR on a population with different low refractive errors. In addition, certain demographic and lifestyle factors were analyzed regarding their influence on the myopia prevention effect of 0.01% ATR.

Materials and methods

Patient selection

This retrospective cohort study was conducted at the Nobel Eye Institute, a clinic network comprising more than 20 clinics across the Taiwan region. Participants were included in the study population if they met the following criteria: (1) aged between 6 and 12 years; (2) a low refractive error status with an initial SER ranging from 0.75D to -0.50D according to previous experience [17]; and (3) completed at least six follow-up visits within one year after the initial presentation. To standardize ocular conditions across all individuals, the following exclusion criteria were applied: (1) corrected distance visual acuity (CDVA) below 20/25; (2) presence of significant ocular diseases, including, but not limited to, corneal opacity, infantile glaucoma, corneal neovascularization, congenital cataract, optic nerve atrophy, retinal detachment, juvenile idiopathic arthritis-associated uveitis, retinopathy of prematurity, globe rupture, or microphthalmos; and (3) receipt of any additional myopia control treatment other than 0.01% ATR management (e.g., orthokeratology contact lenses) during the follow-up period. Regarding the protocol for managing a population with low refractive error in our institution, all children were advised to instill 0.01% ATR, and one type of topical 0.01% ATR (Atropine Eye Drops 0.01%; Sinphar Group, Luodong, Yilan County, Taiwan) would be prescribed if their parents agreed to the myopic prevention treatment after discussing with ophthalmologists. In addition, individuals were instructed to wear single-vision spectacles to maintain clear vision for the whole day if their uncorrected visual acuity (UDVA) was below 20/25. After collecting all the available medical documents after the exclusion process, individuals were categorized into the ATR (with 0.01% ATR instillation) and non-ATR (without 0.01% ATR instillation) groups. Moreover, only the right eye of each participant was included in the analysis. Ultimately, 55 and 67 eyes were included in the ATR and non-ATR groups, respectively.

Myopic evaluation

All individuals at the Nobel Eye Institute and included in this study were examined following a standardized myopic evaluation protocol, and relevant medical data for these individuals were extracted from our institute’s medical records. Baseline (i.e., pre-treatment) parameters included UDVA, CDVA, spherical and cylindrical powers, SER, flat and steep keratometry readings, and AXL. UDVA and CDVA were assessed using a standard Snellen chart. Objective refractive values, including spherical and cylindrical components, were obtained with an autorefractor (KR-8900; Topcon, Itabashi-ku, Tokyo, Japan) under cycloplegic conditions. Cycloplegia was induced using topical tropicamide (Better Eye Drops; Aseptic Innovative Medicine Co. Ltd., Taoyuan District, Taoyuan, Taiwan), which was administered at least three times. The degree of pupil dilation was evaluated by an optometrist, and when the pupil diameter exceeded 8 mm, three consecutive cycloplegic refractions were executed using the autorefractor, with the mean of these measurements recorded. The cycloplegic SER was defined as the spherical power plus half of the cylindrical power. Corneal curvature (i.e., flat and steep keratometry) was measured using a corneal topographer (TMS-5; Tomey Corporation, Nishi-ku, Nagoya, Japan), and AXL was determined with the optical biometer (IOL Master 500; Carl Zeiss, Göschwitzer Str., Jena, Germany). Once the individuals were followed up in our institutions, myopic assessments included CDVA, cycloplegic refraction, and AXL measurements. The primary outcome of this study was changes in SER and AXL. Baseline and one-year post-treatment measurements were collected and subjected to statistical analysis to evaluate the effectiveness of 0.01% ATR management.

Statistical analysis

Statistical analyses were performed using SPSS software (version 20.0; SPSS Inc., Chicago, IL, USA). The Shapiro–Wilk test was applied to assess the normality of medical data distributions in both groups, which demonstrated normal distributions (P > 0.05). The statistical power was 0.89 at an alpha level of 0.05 assuming a medium effect size, as calculated using G*Power software (version 3.1.9.2; Heinrich Heine University, Düsseldorf, Germany). Descriptive statistics were used to summarize baseline characteristics of the ATR and non-ATR groups. Either the chi-square test or independent T test was employed to compare baseline characteristics between the ATR and non-ATR groups depending on the variable type. Independent T tests were further executed to evaluate intergroup differences in SER progression and AXL elongation over different time intervals. Subsequently, the study population was divided into those with negative refractive status (0D to -0.50D) and those with plus refractive status (0.75D to 0D), and then the independent T test was applied again to compare the SER and AXL changes between the two groups with different refractive status. For the sensitive analysis, the generalized linear model was applied to identify potential associations between specific factors and myopic progression (an SER change more than − 0.25D or an AXL change more than 0.12 mm) within each group. The analyzed factors included young initial age (younger than 6 years), minus initial SER (SER lower than 0D), high solar exposure (exposure to sun for more than 1 h per day), high near-work activity (spent on reading or mobile device use after school for more than 3 h per day), and high outdoor activity (spent on outdoor/physical activity for more than 1 h per day). For each parameter, the odds ratio (OR) and corresponding 95% confidence interval (CI) of both groups were calculated to estimate the strength of association with myopic progression. Statistical significance was defined as P < 0.05 in the chi-square test or independent T test and as P < 0.01 in the generalized linear model using Bonferroni correction to decrease the incidence of false-positive results as possible.

Results

The basic characteristics of the two groups are demonstrated in Table 1. The mean age was 7.17 ± 0.84 and 6.94 ± 0.79 years old in the ATR and non-ATR groups, respectively, without a prominent difference (P = 0.123). Sex distribution, initial UDVA, and initial CDVA also demonstrated an insignificant difference between the two groups (all P > 0.05). For the other ocular parameters, sphere power, cylinder power, SER, and the keratometry reading all represented insignificant differences between the two groups (all P > 0.05). Similarly, the initial AXL values did not exhibit significant differences between the two groups (P = 0.606) (Table 1).

Table 1.

Baseline characteristics between the two populations

Characters	ATR group (N = 55)	Non-ATR group (N = 67)	P
Age (year)	7.17 ± 0.84	6.94 ± 0.79	0.123
Sex (male: female)	31:24	36:31	0.771
UDVA (LogMAR)	0.18 ± 0.14	0.20 ± 0.17	0.486
CDVA (LogMAR)	0.00 ± 0.01	0.00 ± 0.00	0.999
Sphere power (D)	0.37 ± 0.41	0.34 ± 0.38	0.676
Cylinder power (D)	-0.72 ± 0.44	-0.78 ± 0.46	0.466
SER (D)	0.01 ± 0.39	-0.05 ± 0.36	0.379
Flat keratometry (D)	42.59 ± 1.68	42.74 ± 1.83	0.641
Steep keratometry (D)	43.65 ± 1.93	43.82 ± 2.08	0.643
AXL (mm)	23.04 ± 0.33	23.01 ± 0.31	0.606

ATR: atropine, AXL: axial elongation, CDVA: corrected distance visual acuity. D: diopter, N: number, SER: spherical equivalent refraction, UDVA: un-corrected visual acuity

The initial SER was 0.01 ± 0.39D and − 0.05 ± 0.36D in the ATR and non-ATR groups, respectively (P = 0.379), and after the one-year follow-up period, the corresponding final SER was − 0.08 ± 0.34D and − 0.46 ± 0.39D (P < 0.001). The change in SER was significantly lower in the ATR group compared to the non-ATR group (-0.09 ± 0.07D versus − 0.41 ± 0.28D, P < 0.001) (Table 2). On the other hand, the initial AXL was 23.04 ± 0.33 and 23.01 ± 0.31 mm in the ATR and non-ATR groups, respectively (P = 0.606), and after the one-year follow-up interval, the corresponding final AXL was 23.08 ± 0.33 and 23.18 ± 0.34 mm (P = 0.104) (Table 2). The elongation in AXL was significantly lower in the ATR group than the non-ATR group (0.04 ± 0.03 versus 0.17 ± 0.10 mm, P < 0.001). For those with an initial SER of 0D to -0.50D, SER and AXL progression in the ATR group was significantly lower than that in the non-ATR group (both P < 0.001). Similarly, that progression in the ATR group was significantly lower than that in the non-ATR group in children with an initial SER of 0.75D to 0D (both P < 0.001) (Table 3). The children with plus initial SER, regardless of whether they were in the ATR or non-ATR groups, demonstrated a lower amount of SER and AXL progression compared to those with minus initial SER (all P < 0.05) (Table 3).

Table 2.

Progression of spherical equivalent refraction and axial length after the follow-up period

Index	ATR group	Non-ATR group	P
SER (D)
Pre-treatment	0.01 ± 0.39	-0.05 ± 0.36	0.379
Post-treatment	-0.08 ± 0.34	-0.46 ± 0.39	< 0.001*
Progression	-0.09 ± 0.07	-0.41 ± 0.28	< 0.001*
AXL (mm)
Pre-treatment	23.04 ± 0.33	23.01 ± 0.31	0.606
Post-treatment	23.08 ± 0.33	23.18 ± 0.34	0.104
Progression	0.04 ± 0.03	0.17 ± 0.10	< 0.001*

AXL: axial elongation, D: diopter, N: number, SER: spherical equivalent refraction

* denotes a significant difference between the two groups

Table 3.

Myopic progression between the two groups with different initial myopia degrees

Myopia progression	ATR group	Non-ATR group	P
0D to -0.50D
SER (D)	-0.17 ± 0.11	-0.56 ± 0.37	< 0.001*
AXL (mm)	0.08 ± 0.04	0.25 ± 0.14	< 0.001*
0.75D to 0D
SER (D)	-0.04 ± 0.02	-0.32 ± 0.15	< 0.001*
AXL (mm)	0.01 ± 0.01	0.12 ± 0.07	< 0.001*

AXL: axial elongation, D: diopter, N: number, SER: spherical equivalent refraction

* denotes a significant difference between the two groups

Regarding specific parameters, high outdoor activity was inversely correlated with the development of myopic progression in the ATR group (P = 0.006) (Table 4). For the non-ATR population, young initial age, minus initial SER, and high near-work activity were correlated with the development of myopic progression (all P < 0.05), while high solar exposure and high outdoor activity exhibited an inverse correlation (both P < 0.05) (Table 5). All children adhered to the ATR treatment throughout the study period. In addition, no severe adverse events were observed in any of them, with only three eyes/children (2.46%) presenting with ocular itchy sensation which subsided after instilling anti-histamine medication. Since all the children completed the whole treatment course, the discontinuation rate was 0.

Table 4.

Indicators for myopic progression in the atropine group

Parameter	OR	95% CI	P
Young initial age	1.05	0.96–1.13	0.148
Minus initial SER	1.14	1.02–1.27	0.012
High solar exposure	0.92	0.86–0.98	0.022
High near-work activity	1.07	1.00-1.14	0.051
High outdoor activity	0.90	0.83–0.97	0.006*

CI: confidence interval, OR: odds ratio, SER: spherical equivalent refraction

* denotes a significant correlation between a parameter and myopic progression

Table 5.

Indicators for myopic progression in the non-atropine group

Parameter	OR	95% CI	P
Young initial age	1.10	1.04–1.17	0.001*
Minus initial SER	1.18	1.07–1.30	< 0.001*
High solar exposure	0.90	0.82–0.98	0.009*
High near-work activity	1.15	1.05–1.25	< 0.001*
High outdoor activity	0.88	0.84–0.94	< 0.001*

CI: confidence interval, OR: odds ratio, SER: spherical equivalent refraction

* denotes a significant correlation between a parameter and myopic progression

Discussion

Myopia could contribute to several ocular disorders according to the literature [2]. Myopic maculopathy is associated with disruption of Bruch’s membrane and the outer neurosensory retinal layer [3], and in advanced forms of this disorder, choroidal neovascularization can interfere with visual acuity [2, 3]. Despite the arrangement of treatment, the prognosis of myopic maculopathy is unknown and persistent visual impairment may still occur [18]. Retinal detachment is another major co-morbidity of both general myopia and high myopia [3], where a 200-fold risk of rhegmatogenous retinal detachment was observed in a high-myopia population compared to the general populace [2], but the postoperative visual condition of rhegmatogenous retinal detachment cannot be promised [19]. In addition, glaucoma is another common ophthalmological disease in individuals with high myopia, which may contribute to irreversible visual insult or blindness [3, 20]. Correspondingly, preventing myopia development and progression is a major issue for public health. Nowadays, myopic control interventions, which include ATR eyedrops, orthokeratology contact lenses, dual-focus contact lens, and defocus incorporated multiple segments’ spectacle lenses have demonstrated favorable myopic control ability in [21]. Although the exact myopic control mechanism of ATR is not established, non-accommodative mechanisms including muscarinic receptor suppression, epidermal growth factor receptor reduction, dopamine release, and glycosaminoglycan synthesis decrement may account for parts of the myopic control effect of ATR [22]. However, the myopic prevention method has not been explored as much as the above myopic control tools. We think that 0.01% ATR can be an effective myopic prevention tool, while the myopic prevention effect would differ in patients with different refractive errors. The above hypotheses are proven by the findings of this study.

The application of 0.01% low-concentration ATR contributed to a significant myopic prevention effect in children with low refractive error compared to those with low refractive error without ATR treatment, and myopic progression is significantly lower in those ATR and non-ATR children with plus initial SER. In a previous study, 0.01% ATR could effectively control myopia progression, which is not inferior to the ATR therapy with higher concentration [8]. In addition, 0.01% ATR has been proposed for myopic prevention management for the general population due to AXL control [17]. Nevertheless, the myopic prevention effect of 0.01% ATR in specific population and refractive statuses have not been thoroughly investigated. In this study, the amplitude of myopic shift was numerically lower in the children with plus initial SER in both groups, which correspond to the concept of hyperopia reserve for myopic prevention [23]. To our knowledge, our findings may serve as preliminary evidence that illustrates the prominent myopic prevention effect of 0.01% ATR in the Taiwanese population with a low refractive error with plus initial SER than those with minus initial SER. In addition, the baseline features between the ATR and non-ATR groups in this study demonstrated an insignificant difference; thus, the influence of other confounders for myopia development may be minimal. In addition, only one type of 0.01% ATR was administered in our institution, and the biased result from different medications may not be prominent. Correspondingly, the effect of 0.01% ATR on myopic prevention could indeed be higher in the population with low refractive error with plus initial SER. On the other hand, the changes in SER and AXL were significantly lower in the ATR group than the non-ATR group, and even the ATR group with minus initial SER showed a numerically lower myopic progression than did the non-ATR group with plus initial SER. These findings further strengthen the application of 0.01% ATR for myopia prevention.

Concerning the sensitive analysis, all the demographic and lifestyle factors were significantly associated with myopia development in the non-ATR population. Young age is a known predisposing factor for myopia development and progression, in which patients younger than 9 years demonstrated a higher amplitude of AXL progression than their older counterparts did [24]. In addition, a higher initial myopia degree is associated with a higher chance of myopia progression [25]. For the lifestyle factor, near-work activity is a universal risk factor for myopia development and progression, which may account for the rapid growth of myopia prevalence [8]. In addition, outdoor activity has been proposed as an efficiency management solution for myopic control in the Taiwanese population [26], and solar exposure may also be a protective factor against myopia development [27]. As a result, the results in the non-ATR group correspond to previous concepts. Still, only high outdoor activity significantly correlated with myopia development in the ATR group. There are few papers that report this phenomenon. We propose that the myopic prevention effect of 0.01% ATR offsets the influences of certain risk and protective factors. Still, the correlation between near-work activity and myopia development in the ATR group nearly reaches significance, and the statistical results may change if we include more individuals. Further study with more individual numbers may be needed to clarify this relationship.

Regarding the effectiveness of myopia control in this study, annual SER progression in the ATR group was − 0.09 ± 0.07D. In a previous study, SER progression in those patients that received 0.01% ATR treatment was − 0.58D in one year [28], and another study with 0.01% ATR and spectacle lens application demonstrated − 0.30D [29]. AXL elongation one year after 0.01% ATR treatment was 0.04 ± 0.03 mm. In the previous publications that measured the changes in AXL after 0.01% ATR administration, one-year AXL elongation ranged from 0.10 to 0.27 mm [30]. In previous major 0.01% ATR randomized clinical trial and review articles, annual SER progression from − 0.25D to -0.59D and AXL elongation from 0.12 to 0.36 mm were observed [31–33]. Accordingly, the SER progression (-0.09 ± 0.07D) and AXL elongation (0.04 ± 0.03 mm) amounts between the ATR groups in this study and the population reported in previous major 0.01% ATR studies may be comparable. On the other hand, SER progression and AXL elongation in the non-ATR groups of this study were − 0.41 ± 0.28D and 0.17 ± 0.10 mm, respectively, which is similar to the trend of natural myopic progression [34]. These results may indicate that the 0.01% ATR treatment in our institution is effective on myopic prevention and highlight the importance of myopic prevention even in children with low refractive error.

There are certain limitations in this study. First, the total number of eyes is relatively inadequate; only 122 eyes were enrolled. This scale of study population may contribute to statistical bias regardless of the normal distribution and sufficient statistical power of our population. Second, all the individuals in this study are Han Taiwanese, and the external validity of our results may be low. The follow-up duration of this study was also short. In addition, the retrospective design of this study may reduce the homogeneity of the two groups, despite the initial characteristics that related to myopia not showing a significant difference in statistical analysis. Although baseline characteristics appear similar between groups, and the educational intensity (all children spent 3–4 h of studying after school according to parents) as well as adherence to treatment (all children adhere to the ATR management) could also be similar, we cannot ensure the homogeneity of socioeconomic status and other parameters between groups. Cyclopentolate is not available in Taiwan, so we can only use tropicamide as the cycloplegia agent, which may not lead to a full cycloplegic effect, although we confirmed that all pupils reached at least 8 mm before cycloplegic refraction. Finally, the decision to receive 0.01% ATR treatment depends on the discussion with the parents, not totally by the ophthalmologist based on the objective conditions and parameters, due to the retrospective design of this study (all the individuals were routine cases in our clinics). This pronounced selection bias could reduce the integrity of our analyses and results. Overall, the inherent limitations of this retrospective study and limited control of significant confounding factors including socioeconomic status and environmental influences could prominently reduce the integrity of the result interpretations of this study.

In conclusion, the application of 0.01% topical ATR might own some relationship to lower SER and AXL progressions compared to those without ATR therapy in children with low refractive error. Furthermore, the myopic prevention effect of 0.01% ATR might be slightly better in those with plus initial SER. Consequently, 0.01% ATR management might be recommended once a child reaches low-refractive-error status and present with plus initial SER to reduce as much as possible the chances of myopia development. Further large-scale prospective studies that evaluate the myopic prevention effect among 0.01% ATR, other ATR concentrations, and other myopic control interventions are mandatory.

Abbreviations

ATR

Atropine

AXL

Axial length

CDVA

Corrected distance visual acuity

CI

Confidence interval

D

Diopter

N

Number

OR

Odds ratio

SER

Spherical equivalent refraction

UDVA

Uncorrected visual acuity

Author contributions

Conceptualization, C.-K.C.; methodology, S.-F.Y.; software, C.-K.C.; formal analysis, J.-Y.H. and Y.-L.C.; data curation, P.-H.H.; writing—original draft preparation, C.-Y.L.; writing—review and editing, C.-K.C.; validation, P.-H.H.; supervision, C.-K.C. All authors have read and agreed to the submitted version of the manuscript.

Funding

None.

Data availability

The data applied in this study is available from corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

All procedures described in this study complied with the ethical standards of the Declaration of Helsinki in 1964 and its subsequent amendments. The study protocol was approved by the Institutional Review Board of the National Changhua University of Education (project code: NCUEREC-112-071; approval date: 27 September 2023). Owing to the retrospective nature of this study, the requirement for written informed consent was waived by the National Changhua University of Education.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.