Timing of height growth peaks and axial elongation in atropine-treated myopic children Authors: Dongheon Surl, Jinu Han, Yuri Seo Published: 2026 ABSTRACT PurposeTo investigate the association between systemic height growth rate and ocular axial elongation rate in atropine-treated children with myopia, and to explore whether the timing of height growth provides clinically relevant information regarding periods of axial elongation during treatment.MethodsChildren with myopia (< 15 years) who underwent serial ocular biometry and anthropometric measurements over ≥ 3 visits spanning > 1 year were included. Participants were classified into a control group (no atropine) and an atropine group (0.05% nightly, initiated for axial elongation ≥ 0.4 mm/year). Axial elongation and height growth rates were calculated for each inter-visit interval. Peak growth ages were defined as those with the greatest inter-visit changes. Subgroup analyses in the atropine group were based on the median peak height growth age (10.45 years) and median annual axial elongation rate (0.21 mm/year).ResultsA total of 139 eyes from 70 children were analyzed (56 control, 83 atropine-treated). Height growth rate and axial elongation rate were not significantly correlated in either group. In the control group, peak height growth age and peak axial elongation age showed similar unimodal distributions. In the atropine group, however, the distribution of peak height growth age showed a broadened pattern with marginal deviation from unimodality (P = 0.051) and associated strongly with peak axial elongation age (β = 0.91, P < 0.01). Children with earlier height growth peaks had greater annual axial elongation (0.27 ± 0.17 vs. 0.15 ± 0.12 mm/year; P < 0.01) despite identical dosing.ConclusionsAmong atropine-treated children, earlier systemic growth peaks were associated with faster axial elongation. Monitoring height growth timing may help identify rapid ocular growth periods and optimize treatment strategies.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12886-026-04954-y. Introduction Myopia is a growing global public health concern, with increasing prevalence and earlier onset among children [1–4]. Among various interventions to slow myopia progression, low-concentration atropine eye drops have shown efficacy in reducing ocular axial elongation—a key factor associated with disease progression and long-term ocular complications [5–7]. In clinical practice, decisions regarding the initiation or adjustment of atropine therapy are often guided by the rate of axial elongation, which can be measured quickly and noninvasively, without requiring cycloplegia [8]. However, axial elongation is not always predictable at the time of initial diagnosis or during early follow-ups. We hypothesized that the rate of systemic physical growth might parallel the rate of ocular axial elongation in children. Prior studies have explored associations between ocular axial length and anthropometric parameters, such as height, body mass index (BMI), and foot size [9–13]. While some reported positive correlations between stature or growth velocity and axial length, others failed to establish consistent or predictive relationships. These discrepancies may stem from differences in study design, age ranges, ethnicity, or lack of treatment stratification. Despite widespread use of atropine in clinical settings, no study has examined whether systemic growth markers can predict axial elongation in children undergoing pharmacological treatment. Given that 25%–42% of atropine-treated children are classified as non-responders in previous trials, identifying predictors of treatment response remains an important unmet need. In prior studies, non-responders are typically defined as those demonstrating continued myopia progression, such as refractive progression of ≥ 0.5 D/year or axial elongation of approximately ≥ 0.3 mm/year despite treatment [14–17]. If systemic growth trajectories—such as the timing or velocity of height spurts—can indicate future axial elongation, clinicians may better anticipate periods of rapid myopia progression. This could facilitate prognostic discussions with families regarding and optimize the timing of treatment initiation or adjustment. Additionally, understanding whether certain growth characteristics are linked to reduced responsiveness to atropine could support personalized management strategies for pediatric myopia. Rather than examining a direct causal relationship between systemic growth and ocular axial elongation, this study aimed to characterize the overall relationship between axial elongation rate and systemic height growth patterns, and to further explore whether the timing of somatic growth is associated with inter-individual variability in axial elongation among children undergoing atropine treatment. Methods This retrospective cohort study included children with myopia under 15 years of age who underwent serial ocular biometry and anthropometric assessments at the Department of Ophthalmology, Yongin Severance Hospital, between January 2020 and September 2022. Eligible children had attended at least three clinic visits over a period exceeding one year. Participants were classified into two groups: a control group with no atropine treatment, and an atropine group prescribed 0.05% atropine once nightly, initiated due to axial elongation of ≥ 0.4 mm/year and the absence of contraindications such as cardiac or respiratory conditions [18]. This threshold was selected to identify individuals with clinically meaningful progression who were more likely to benefit from intervention, based on prior studies and common clinical practice patterns [19]. We acknowledge that lower rates of progression, such as 0.3 mm/year, may also warrant treatment in certain cases; however, a higher threshold was applied in this study to capture a population with more pronounced progression. The baseline visit for the atropine group was defined as the time of treatment initiation. Inclusion criteria required a myopic refraction of less than − 0.50 D and astigmatism less than − 2.5D at the initial visit, attendance at a minimum of three clinic visits over a follow-up period exceeding one year, and complete ocular biometric and anthropometric data for all visits, including axial length, anterior chamber depth, lens thickness, white-to-white distance, keratometry, central corneal thickness, height, weight, and BMI. Exclusion criteria included hyperopia at baseline, a history of retinopathy of prematurity, congenital cataract, amblyopia or strabismus requiring surgical correction; congenital, neurological or syndromic disorders; growth hormone therapy or gonadotropin-releasing hormone analogs for precocious puberty, and prior use of orthokeratology lenses; myopia control spectacles including Defocus Incorporated Multiple Segments (DIMS) lenses, or soft contact lenses. Children in the control group wore single-vision spectacles only. In total, 906 children were screened. After applying the above inclusion and exclusion criteria, 70 children (139 eyes) were included in the final analysis, while 836 children were excluded. The final study population consisted of 28 children (56 eyes) in the control group and 42 children (83 eyes) in the atropine treatment group. A flow diagram of participant selection is provided in Supplementary Fig. 1. The primary outcome was the axial elongation rate. Axial elongation and height growth rates were calculated between visits by dividing the change in axial length or height by the time interval, expressed in millimeters per month (mm/month) and centimeters per month (cm/month), respectively. Axial length was measured using optical biometry (IOLMaster 700, Carl Zeiss Meditec, Germany), and height was measured at each visit using a wall-mounted stadiometer, with children barefoot and standing upright. All anthropometric and ocular measurements were obtained on the same visit day. For subgroup comparisons, an average annual axial elongation rate was calculated for each eye by dividing the total axial length change by the total follow-up duration, expressed in millimeters per year (mm/year) to facilitate interpretation. Peak ages for height growth and axial elongation were defined as the visit age corresponding to the greatest change rate between two consecutive visits. Subgroup analyses were conducted within the atropine group. Participants were divided into early or late peak growth subgroups based on whether their peak occurred before or after the median age. Comparisons of annual axial elongation rates were made between these groups. Participants were also stratified into high or low axial elongation subgroups using the median annual axial elongation rate as the threshold. Peak ages for height and axial elongation were then compared across subgroups. Statistical analyses were performed using R software (version 4.3.0). Baseline characteristics were compared using independent t-tests. Linear mixed-effects models were used to evaluate the associations between height velocity and axial elongation while accounting for repeated measures. Height was treated as a subject-level variable, while axial elongation was analyzed at the eye level using mixed-effects models with subject-level random effects. To control for age-related variations, participants were stratified into quartiles based on visit age (Q1: ≤ 8.92 years; Q2: 8.93–10.45 years; Q3: 10.46–11.77 years; Q4: ≥ 11.78 years), and these strata were incorporated into graphical and modeling analyses. Distributional characteristics of peak ages were assessed using the Shapiro-Wilk test (normality), Levene’s test (homogeneity of variance), Kolmogorov–Smirnov test (distributional differences), and Hartigan’s dip test (modality). Skewness and kurtosis were also calculated to assess distributional shape. Linear mixed-effects model was used to evaluate the association between height peak age and axial elongation peak age. A p-value < 0.05 was considered statistically significant. This study was approved by the Institutional Review Board of Yongin Severance Hospital, Yongin, South Korea (approval number: 2024-0339-001), and conducted in accordance with the Declaration of Helsinki. The author declare no conflicts of interest. Results A total of 139 eyes from 70 children were included in the analysis (56 eyes in the control group and 83 eyes in the atropine group). The median follow-up interval was significantly shorter in the atropine group compared to the control group (3.88 months [interquartile range, 3.68–4.16] vs. 7.13 months [interquartile range, 5.75–11.96], P < 0.001). Baseline biometric and anthropometric characteristics are summarized in Table 1. At the initial visit, there were no significant differences in height, weight, BMI, or ocular biometry measurements between the two groups, except for axial length and spherical equivalent. Initial axial length was significantly longer in atropine group compared to the control group (24.76 ± 1.03 mm vs. 24.18 ± 0.98 mm, P < 0.01). Initial spherical equivalent was also significantly more myopic in the atropine group (–3.19 ± 1.89D vs. − 2.11 ± 1.85D, P < 0.01). Comparison of biometric and anthropometric change rates revealed significantly lower axial elongation in the atropine group (0.018 ± 0.014 mm/month vs. 0.024 ± 0.011 mm/month, P < 0.01). Corneal diameter (white-to-white) increased more in the atropine group than in the control group (0.003 ± 0.007 mm/month vs. − 0.001 ± 0.008 mm/month, P < 0.01); however, this difference was minimal and likely within the range of measurement variability. No significant differences were observed for other parameters between the groups (Supplementary Table 1). Table 1Baseline characteristics and initial ocular biometry in the control and atropine groupsVariableControl(n = 56, Mean ± SD)Atropine(n = 83, Mean ± SD)p-valueAge9.47 ± 2.089.14 ± 2.030.36Sex (% of male)27.3%37.4%0.65Height(cm)138.05 ± 14.03134.31 ± 14.630.13Weight(kg)38.29 ± 12.0134.65 ± 12.090.08BMI(kg/m2)19.63 ± 3.2718.65 ± 3.250.09Axial length(mm)24.18 ± 0.9824.76 ± 1.03< 0.01*ACD(mm)3.72 ± 0.233.78 ± 0.230.12LT(mm)3.38 ± 0.183.38 ± 0.150.84WTW(mm)12.12 ± 0.3612.2 ± 0.460.26Mean K43.56 ± 1.1143.18 ± 1.630.11CCT(㎛)559.91 ± 36.25551.07 ± 32.360.14SE–2.11 ± 1.86–3.19 ± 1.89< 0.01*ACD: anterior chamber depth; LT: lens thickness; WTW: white to white; CCT: central corneal thickness; SE: spherical equivalent; BMI: body mass index. *p < 0.05 is considered as statistically significance. Table 1Baseline characteristics and initial ocular biometry in the control and atropine groupsVariableControl(n = 56, Mean ± SD)Atropine(n = 83, Mean ± SD)p-valueAge9.47 ± 2.089.14 ± 2.030.36Sex (% of male)27.3%37.4%0.65Height(cm)138.05 ± 14.03134.31 ± 14.630.13Weight(kg)38.29 ± 12.0134.65 ± 12.090.08BMI(kg/m2)19.63 ± 3.2718.65 ± 3.250.09Axial length(mm)24.18 ± 0.9824.76 ± 1.03< 0.01*ACD(mm)3.72 ± 0.233.78 ± 0.230.12LT(mm)3.38 ± 0.183.38 ± 0.150.84WTW(mm)12.12 ± 0.3612.2 ± 0.460.26Mean K43.56 ± 1.1143.18 ± 1.630.11CCT(㎛)559.91 ± 36.25551.07 ± 32.360.14SE–2.11 ± 1.86–3.19 ± 1.89< 0.01*ACD: anterior chamber depth; LT: lens thickness; WTW: white to white; CCT: central corneal thickness; SE: spherical equivalent; BMI: body mass index. *p < 0.05 is considered as statistically significance. Baseline characteristics and initial ocular biometry in the control and atropine groups ACD: anterior chamber depth; LT: lens thickness; WTW: white to white; CCT: central corneal thickness; SE: spherical equivalent; BMI: body mass index. *p < 0.05 is considered as statistically significance. Because age is known to affect both height growth and axial elongation [19–21], the influence of age on height growth rate was evaluated before analyzing its relationship with axial elongation. In the control group, height growth rate decreased significantly with age (β = − 0.028 cm/month per year, P = 0.012); Supplementary Fig. 2. In contrast, no significant association was observed in the atropine group (β = 0.004 cm/month per year, P = 0.716), suggesting more heterogeneous somatic growth trajectories among atropine-treated children. The association between height growth rate and axial elongation rate was evaluated using linear mixed-effects models while adjusting for age (Fig. 1). In the atropine group, height growth rate was not significantly associated with axial elongation (β = − 0.021 mm/month for each 1 cm/month increase in height growth rate, P = 0.726). Axial elongation rate showed a modest but significant decrease with age (β = − 0.011 mm/month per year, P = 0.011), although this association was no longer significant after adjusting for both age and height growth rate (P = 0.942). In the control group, axial elongation rate also decreased significantly with age (β = − 0.048 mm/month per year, P = 0.003), while the effect of height growth rate remained non-significant (β = − 0.537 mm/month for each 1 cm/month increase in height growth rate, P = 0.123). After adjusting for age, the interaction term between height growth rate and age was also not statistically significant (P = 0.182). To account for age-related variation in growth patterns, data points in Fig. 1 were stratified by visit age quartiles (Q1: ≤ 8.92 years, Q2: 8.93–10.45 years, Q3: 10.46–11.77 years, Q4: ≥ 11.78 years), showing that the lack of association between height and axial elongation persisted across age groups. In conclusion, height growth rate showed no significant association with axial elongation rate in either group even after adjusting for age. Fig. 1Association between height change and axial elongation by treatment group. Scatter plots and linear regression lines illustrate the relationship between height change and axial length change per visit in the control and atropine groups. Data points are grouped by visit age quartiles to account for age-related effects (Q1: ≤ 8.92 years, Q2: 8.93–10.45 years, Q3: 10.46–11.77 years, Q4: ≥ 11.78 years), and lines represent linear regression fits. Linear mixed-effects modeling, incorporating visit age as an interaction term, confirmed no significant association between height velocity and axial elongation in either group (P > 0.05). Stratification by age quartiles and inclusion of age in the model indicate that height change alone does not predict ocular growth, regardless of age Association between height change and axial elongation by treatment group. Scatter plots and linear regression lines illustrate the relationship between height change and axial length change per visit in the control and atropine groups. Data points are grouped by visit age quartiles to account for age-related effects (Q1: ≤ 8.92 years, Q2: 8.93–10.45 years, Q3: 10.46–11.77 years, Q4: ≥ 11.78 years), and lines represent linear regression fits. Linear mixed-effects modeling, incorporating visit age as an interaction term, confirmed no significant association between height velocity and axial elongation in either group (P > 0.05). Stratification by age quartiles and inclusion of age in the model indicate that height change alone does not predict ocular growth, regardless of age Our initial hypothesis was that axial elongation rate would be associated with height growth rate. However, as shown in Fig. 1, this association was not observed. To further investigate this discrepancy, we analyzed the distribution of peak ages for height growth rate and axial elongation rate in each group. The peak age was defined as the visit age at which the greatest change rate occurred for each eye. The distributions are shown in Fig. 2. Fig. 2Distribution of individual peak ages of height growth and axial elongation by group. The peak age was defined as the visit age at which the highest rate of change occurred, adjusting for visit interval. Each panel displays the distribution of peak ages for height growth (left) and axial elongation (right). In each histogram, gray bars represent the atropine group and white bars (unfilled) represent the control group. Overlaid density plots show smoothed distribution, with solid lines for the control group and dotted lines for the atropine group. In the control group, both height growth and axial elongation peak ages followed a unimodal, approximately normal distribution (peaks at 11.09 and 11.03 years, Shapiro–Wilk test, P = 0.172 and 0.406, respectively; solid line). In the atropine group, height growth peak age demonstrated a broadened distribution with increased heterogeneity in peak timing (Hartigan’s dip test, P = 0.051; dotted line), with two local maxima observed on visual inspection at 8.81 and 10.75 years. Axial elongation peak age in the atropine group also showed a visually broader distribution (peaks at 9.77 and 11.5 years), although statistical tests did not confirm the deviation from unimodality (Hartigan’s dip test, P =0.091; Shapiro–Wilk P = 0.478) Distribution of individual peak ages of height growth and axial elongation by group. The peak age was defined as the visit age at which the highest rate of change occurred, adjusting for visit interval. Each panel displays the distribution of peak ages for height growth (left) and axial elongation (right). In each histogram, gray bars represent the atropine group and white bars (unfilled) represent the control group. Overlaid density plots show smoothed distribution, with solid lines for the control group and dotted lines for the atropine group. In the control group, both height growth and axial elongation peak ages followed a unimodal, approximately normal distribution (peaks at 11.09 and 11.03 years, Shapiro–Wilk test, P = 0.172 and 0.406, respectively; solid line). In the atropine group, height growth peak age demonstrated a broadened distribution with increased heterogeneity in peak timing (Hartigan’s dip test, P = 0.051; dotted line), with two local maxima observed on visual inspection at 8.81 and 10.75 years. Axial elongation peak age in the atropine group also showed a visually broader distribution (peaks at 9.77 and 11.5 years), although statistical tests did not confirm the deviation from unimodality (Hartigan’s dip test, P =0.091; Shapiro–Wilk P = 0.478) In the control group, the peak ages for both height growth and axial elongation rates were normally distributed (Shapiro–Wilk test, P = 0.172 and 0.406, respectively), and nearly identical (11.09 ± 1.92 years vs. 11.03 ± 2.16 years), with no significant difference between them (paired t-test, 0.37P = 0.62). These results suggest a synchronized pattern of systemic height and ocular axial growth in the absence of atropine treatment. Conversely, the atropine group showed a broader distribution of peak ages for height growth (Hartigan’s dip test, P = 0.051). Although the distribution did not show two sharply separated peaks, visual inspection of Fig. 2 reveals a flattened central portion and a wide spread of peak ages, suggesting heterogeneity in growth timing among treated children. The distribution of axial elongation peak age in the atropine group appeared similarly variable, though statistical tests supported approximate unimodality (Shapiro-Wilk test: P = 0.478; Hartigan’s dip test: P = 0.091). The two most prominent peak ages for height growth in the atropine group were 8.81 and 10.75 years, while those for axial elongation were 9.77 and 11.5 years. Despite the appearance of multiple peaks, the mean peak age of axial elongation was 10.59 ± 2.1 years, representing the central tendency. Summary statistics, including skewness, kurtosis, and results of normality and variance tests, are provided in Supplementary Table 2. To further explore variability in peak growth timing within the atropine group, participants were categorized as “early growth peak (n = 42)” and “late growth peak (n = 41)” based on the median peak age of height growth (10.45 years). A significant positive association was observed between peak height growth age and peak axial elongation age in the atropine group, using a linear mixed-effects model with subject-level random effects (β = 0.91, P < 0.01; Fig. 3). Earlier peak height growth was associated with earlier axial length elongation, suggesting close alignment in the timing of systemic and ocular growth spurt. However, after adjusting for baseline axial length, spherical equivalent, and age, the association was attenuated and did not remain statistically significant. Fig. 3Association between peak age of height growth and axial elongation in the atropine group. Scatter plots show a strong positive association between the peak age of height growth and the peak age of axial elongation in atropine-treated children The regression line represents estimates from a linear mixed-effects model with subject-level random effects. (β = 0.91, P < 0.01). Data points are stratified into early and late height growth peak groups based on the median value (10.45 years) Association between peak age of height growth and axial elongation in the atropine group. Scatter plots show a strong positive association between the peak age of height growth and the peak age of axial elongation in atropine-treated children The regression line represents estimates from a linear mixed-effects model with subject-level random effects. (β = 0.91, P < 0.01). Data points are stratified into early and late height growth peak groups based on the median value (10.45 years) Axial elongation rate per year was also compared between the two subgroups. The early height growth peak group exhibited a significantly faster axial elongation rate than the late growth peak group (0.27 ± 0.17 mm/year vs. 0.15 ± 0.12 mm/year, P < 0.01; Fig. 4). A similar pattern was observed when the groups were defined by axial length peak age (median = 10.28 years), with the early peak group showing faster elongation (0.27 ± 0.17 mm/year vs. 0.15 ± 0.12 mm/year, P < 0.01). This subgroup analysis is presented in Supplementary Fig. 3. Fig. 4Comparison of total axial length elongation rate between early and late growth peak subgroups in the atropine group. Box plots compare axial elongation rate (mm/year) between children with early and late height growth peaks, classified based on the median height growth peak age (10.45 years). Eyes from early growth peak group (n = 42) exhibited significantly faster axial elongation than those from the late group (0.27 ± 0.17 mm/year vs. 0.15 ± 0.12 mm/year, P < 0.01) Comparison of total axial length elongation rate between early and late growth peak subgroups in the atropine group. Box plots compare axial elongation rate (mm/year) between children with early and late height growth peaks, classified based on the median height growth peak age (10.45 years). Eyes from early growth peak group (n = 42) exhibited significantly faster axial elongation than those from the late group (0.27 ± 0.17 mm/year vs. 0.15 ± 0.12 mm/year, P < 0.01) To further assess the relationship between growth timing and myopia progression in the atropine group, participants were stratified by the median axial elongation rate (0.21 mm/year). The high axial elongation rate group (n = 41) showed significantly earlier peak ages than the low-rate group (n = 42) for both height growth (9.66 ± 1.91 vs. 11.06 ± 1.82 years, P < 0.001) and axial elongation (9.6 ± 2.11 vs. 11.05 ± 1.72 years, P < 0.001). However, within each subgroup, the timing of height growth and axial elongation did not differ significantly (P = 0.85 for high-rate group; P = 0.56 for low-rate group), suggesting that systemic height and ocular axial growth remained synchronized regardless of the overall rate of elongation (Fig. 5). Fig. 5Comparison of peak age for height growth rate and axial elongation rate between high and low axial elongation rate subgroups in the atropine group. Box plot illustrates the timing of height growth and axial elongation peaks in eyes stratified by axial elongation rate (cutoff = 0.21 mm/year, n = 41 for high group; n = 42 for low group). Both height and axial elongation peaks occurred significantly earlier in the high axial elongation rate group (height peak age: 9.66 ± 1.91 years vs. 11.06 ± 1.82 years; axial elongation peak age: 9.6 ± 2.11 years vs. 11.05 ± 1.72 years, P < 0.001 for both comparisons). However, within each group, the temporal difference between systemic height and ocular axial length growth peak timing was not statistically significant (P = 0.85 for high group; P = 0.56 for low group), indicating synchronized growth patterns regardless of elongation rate Comparison of peak age for height growth rate and axial elongation rate between high and low axial elongation rate subgroups in the atropine group. Box plot illustrates the timing of height growth and axial elongation peaks in eyes stratified by axial elongation rate (cutoff = 0.21 mm/year, n = 41 for high group; n = 42 for low group). Both height and axial elongation peaks occurred significantly earlier in the high axial elongation rate group (height peak age: 9.66 ± 1.91 years vs. 11.06 ± 1.82 years; axial elongation peak age: 9.6 ± 2.11 years vs. 11.05 ± 1.72 years, P < 0.001 for both comparisons). However, within each group, the temporal difference between systemic height and ocular axial length growth peak timing was not statistically significant (P = 0.85 for high group; P = 0.56 for low group), indicating synchronized growth patterns regardless of elongation rate Discussion We hypothesized that the rate of axial elongation would increase in parallel with systemic height growth rate and that this association could influence the efficacy of atropine in suppressing axial elongation. However, no significant correlation between axial elongation rate and systemic height growth rate was found in either the atropine or control groups (Fig. 1). To investigate this discrepancy, we analyzed the distributions of peak age for height growth and axial elongation. In the control group, the peak ages for height growth and axial elongation were closely synchronized (11.09 ± 1.92 vs. 11.03 ± 2.16 years), both following a unimodal, normal distribution. In contrast, the atropine group showed a broader distribution for the peak age of height growth, with a similar pattern of dispersion observed in the distribution of axial elongation peak age, although statistical tests did not confirm significant deviation from unimodality (Fig. 2). Subgroup analyses were conducted within the atropine-treated group (Figs. 3, 4, and 5). Earlier onset of height growth spurts was significantly associated with earlier axial elongation peaks (β = 0.91, P < 0.01; Fig. 3). Moreover, patients with an earlier height growth peak exhibited significantly faster annual axial elongation compared to those with later height growth peaks (P < 0.01; Fig. 4). When participants were stratified by axial elongation rate, those with higher elongation rates demonstrated significantly earlier peak ages for both height growth and axial elongation compared to those with lower elongation rates (Fig. 5). Notably, within each group, the peak ages of height growth and axial elongation remained closely synchronized (Fig. 5). These findings suggest that the temporal pattern of systemic growth—specifically, the timing of the height growth spurt—is more closely aligned with axial elongation than with overall growth velocity, regardless of atropine treatment. Importantly, the absence of a significant linear association between age and axial elongation rate in the atropine-treated cohort should not be interpreted as age-independent ocular growth. Rather, this finding likely reflects the heterogeneity of growth trajectories and the modifying effects of atropine treatment, which may obscure a simple age-related pattern within this clinically selected population. These considerations also highlight the importance of interpreting between-group comparisons with caution. Because the atropine group was defined based on clinically significant axial elongation (≥ 0.4 mm/year), it inherently represents a cohort of faster progressors rather than a population comparable to the control group. Accordingly, the control group does not reflect the untreated natural history of the atropine group, and direct comparisons between groups should be interpreted with caution. In this context, the primary value of the present study lies in the intra-group analysis within the atropine-treated cohort, which allows for the characterization of heterogeneity in growth patterns despite uniform treatment exposure. Interestingly, while both height growth rate and axial elongation rate followed unimodal distributions in the control group, the atropine group exhibited a trend toward bimodal patterns for both parameters—even though the statistical tests support unimodality in both parameters. This pattern is likely not a direct consequence of atropine treatment but may reflect heterogeneity in underlying growth trajectories that could influence treatment response. Given the retrospective nature of this study, group classification was based on real-world clinical decisions: children with myopia of less than − 0.50 D were followed longitudinally, and those demonstrating axial elongation ≥ 0.4 mm/year were initiated on atropine treatment, forming the atropine group, while those with slower progression comprised the control group. Within the atropine group, patients with later height growth peaks exhibited peak ages (10.75 years for height growth and 11.5 years for axial elongation) similar to those observed in the control group (11.09 and 11.03 years, respectively). In contrast, patients with earlier height growth peaks demonstrated markedly earlier timing for both height growth (9.77 years) and axial elongation (8.81 years). This heterogeneity likely contributed to the lack of a consistent linear relationship between height growth rate and axial elongation rate. Nevertheless, a strong positive association was observed between the timing of height growth peak and axial elongation peak within the atropine group (β = 0.91, P < 0.01), indicating that systemic and ocular growth spurts tended to occur in close temporal proximity, regardless of whether the peak occurred early or late (Fig. 3). The attenuation of the association after adjusting for baseline ocular parameters suggests that the relationship between height growth timing and axial elongation may be partially explained by underlying baseline characteristics. This finding supports the possibility that the observed association may be confounded by the selection of rapid progressors in the atropine group, rather than representing an independent pharmacological effect of atropine. Although the control group also demonstrated close alignment between systemic and ocular growth peaks—evidenced by nearly identical mean peak ages and unimodal distributions—there was no significant correlation between the magnitude of height change and axial elongation rate (Fig. 1). These findings suggest that while the timing of systemic and ocular growth is often synchronized, the extent of growth in height and axial length is not necessarily correlated. Another clinically important finding is that, within the atropine group, children with earlier height growth peaks showed significantly greater annual axial elongation than those with later growth peaks, despite receiving the same atropine concentration and dosing frequency (Fig. 4). While systemic and ocular growth appear to be temporally synchronized under natural conditions, as observed in the control group, the clinical relevance of height growth timing becomes more evident in the atropine-treated cohort. In the control group, this synchronization reflects a general physiological pattern of development. In contrast, the atropine group demonstrated greater heterogeneity in growth timing, and this variability was associated with differences in axial elongation despite uniform treatment exposure. In this context, the timing of peak height growth may serve as a practical clinical marker for identifying children at higher risk of continued axial elongation during atropine treatment. Monitoring growth trajectories may therefore help clinicians anticipate periods of increased progression and support more individualized treatment strategies. This also suggests that early-onset systemic growth may be associated with reduced responsiveness to atropine in suppressing axial elongation. To further assess the predictive value of height growth peak timing, we performed a subgroup analysis based on axial elongation rate. When participants were stratified into high (≥ 0.21 mm/year) and low (< 0.21 mm/year) axial elongation rate groups based on the median value, those in the high-rate group demonstrated significantly earlier peak ages for both height growth and axial elongation compared to the low-rate group (Fig. 5). Notably, within each subgroup, the timing of height and axial elongation peaks remained closely synchronized, with no significant difference between the two. These findings indicate that children who experience an earlier systemic growth spurt may be more likely to exhibit faster axial elongation and diminished response to atropine treatment, underscoring the potential of height growth peak age as a clinically meaningful marker for predicting treatment outcomes in pediatric myopia management. Previous studies have demonstrated that ocular axial elongation often coincides with periods of accelerated physical growth during childhood, suggesting a temporally synchronized relationship between systemic and ocular development [9, 11, 22]. Specifically, earlier onset of peak height velocity has been associated with earlier axial elongation and earlier myopia onset, highlighting the importance of growth timing as a risk factor for myopia progression [9]. However, these studies were limited to observational designs and did not assess whether growth patterns may influence responsiveness to pharmacologic interventions such as atropine [9, 11, 12]. In the present study, we analyzed a clinically defined cohort in which children with documented rapid axial elongation were treated with uniform atropine therapy. Among these treated children, those with earlier height growth peaks showed significantly earlier axial elongation peaks and greater annual axial elongation, despite receiving identical atropine concentrations and regimens. This finding suggests that early biological maturation may be associated with attenuated responsiveness to atropine in suppressing ocular growth. Additionally, the broadened distribution of growth peak ages in the atropine group likely reflects underlying heterogeneity in systemic growth trajectories that may have influenced treatment initiation timing. While previous reports have examined anthropometric correlates of axial length in adult populations [10] or identified behavioral and biometric predictors of myopia progression without accounting for treatment exposure [13], our study provides a unique perspective by linking physiologic growth patterns with both axial elongation timing and treatment efficacy in a pediatric population with myopia. Axial elongation and linear height growth are governed by multifactorial biological processes, yet they involve distinct mechanisms. Axial elongation is driven by factors such as scleral remodeling regulated by endoplasmic reticulum stress pathways, biomechanical forces including ciliary muscle activity, and genetic predisposition [23–26]. These elements collectively contribute to progressive ocular elongation and the development of myopia. In contrast, height growth is regulated by endocrine signals, genetic factors, and environmental influences. Given these mechanistic differences, parallel progression of axial and height growth is not inherently expected. Nevertheless, several longitudinal studies have consistently reported a positive association between increases in height and axial length during childhood, particularly during early school age and preadolescence [11, 22, 27, 28]. The observed association between height and axial length appears to be largely explained by shared genetic influences, as supported by twin studies attributing up to 89% of the covariance to genetic factors [29]. Additionally, hormonal mediators such as growth hormone and insulin-like growth factor I—both of which regulate skeletal growth—have been proposed to influence axial elongation through their effects on ocular tissues [30]. While these associations are most prominent during periods of rapid physical development, the relationship tends to weaken after the age of 10–12 years, as growth in both stature and eye size decelerates [11, 31]. These findings suggest that although axial and height growth share overlapping developmental windows and regulatory signals, the extent of their interdependence is limited and modulated by age and individual variability. These factors may have constrained the present study’s ability to more clearly delineate the relationship between height growth and axial elongation. In addition, this study was conducted using a single atropine concentration (0.05%), which may limit the generalizability of the findings to lower-dose regimens such as 0.01% or 0.025%. While the observed association between growth timing and axial elongation likely reflects underlying biological processes rather than a dose-specific effect, the magnitude of treatment response and its interaction with growth patterns may vary depending on atropine concentration. Further studies incorporating multiple dosing regimens are needed to validate these findings. This study had some additional limitations. First, its retrospective design and real-world clinical setting may introduce selection bias—particularly in assigning patients to the atropine group based on documented rapid axial elongation. Notably, the absence of a control group comprising children with axial elongation ≥ 0.4 mm/year who did not receive atropine limited our ability to isolate treatment effects from inherent growth tendencies. This comparison was not feasible due to the retrospective design of the study. Second, in addition to peak height growth, previous studies have shown that pubertal stage as well as various genetic and environmental factors are closely associated with myopia progression. However, these factors were not directly assessed in the present study. Instead, height was used as a surrogate marker to reflect the combined influence of systemic growth-related processes. Therefore, the findings should be interpreted with caution. Third, while we adopted axial elongation ≥ 0.4 mm/year as the clinical threshold for initiating atropine therapy, this objective cut-off may not fully reflect the broader range of clinical considerations that guide treatment decisions in real-world practice. Fourth, we acknowledge that identification of peak axial elongation age may be influenced by visit schedules and the timing of atropine initiation. In the atropine group, however, because follow-up intervals rarely exceeded 6 months, the potential imprecision in peak age estimation is unlikely to exceed approximately 3 months. Finally, the relatively small sample size and broad age range may limit the ability to fully characterize age-specific growth patterns. Although age-related variation was partially addressed through stratification into age quartiles, more granular analyses in larger cohorts are warranted. Moreover, while peak height growth and axial elongation showed close temporal alignment, the present study does not establish a temporal or causal relationship between these variables. Despite these limitations, the study benefits from a well-characterized longitudinal dataset with serial biometric and anthropometric measurements, allowing for precise identification of peak growth timing. The inclusion of a pharmacologically treated cohort further enabled evaluation of not only the temporal relationship between systemic and ocular growth but also its potential influence on treatment responsiveness. By quantitatively analyzing individualized growth trajectories, this study provides novel insights into the dynamic interplay between physical development and myopia control and supports the clinical utility of height growth peak timing as a potential biomarker for guiding atropine treatment in pediatric populations. In conclusion, these findings underscore the importance of monitoring the timing of systemic growth spurts in the management of pediatric myopia. Height growth peak age may help identify periods of increased risk for rapid axial elongation despite atropine treatment. Given the close temporal alignment observed between systemic and ocular growth peaks, even children with delayed height growth spurts may experience later-onset axial elongation, reinforcing the need for continued biometric monitoring beyond the early years. Clinicians may consider integrating growth trajectory assessments into routine care and adjusting the timing or intensity of atropine therapy based on each child’s individualized growth pattern. Conclusions Systemic height growth rate was not linearly associated with axial elongation rate in children with myopia, regardless of atropine treatment. However, in atropine-treated children, the age at peak height growth was closely aligned with the timing of axial elongation and associated with greater annual axial elongation. Children with earlier height growth peaks exhibited faster axial elongation despite receiving identical atropine regimens. Monitoring height growth timing may help identify periods of increased risk for rapid axial elongation and support individualized myopia management. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1