Clinical outcomes and baseline factors associated with one-year axial elongation during orthokeratology wear in children: a retrospective study

Abstract

Background

Myopia has emerged as a major global public health concern, especially in Asia, with high myopia substantially increasing the risk of complications such as retinal degeneration and glaucoma. This study aimed to evaluate the efficacy of orthokeratology (Ortho-K) in controlling axial length (AL) elongation in myopic children over a one-year period and to identify baseline predictors of treatment response.

Methods

This retrospective, single-group, pre-post control study included 122 children (244 eyes) with myopia who were fitted with orthokeratology lenses at Jiangyin People’s Hospital between January 2024 and December 2025 and completed one year of follow-up. Baseline data, including age, body mass index (BMI), spherical equivalent (SE), pupil diameter (PD), and axial length (AL) were collected. Changes in AL were analyzed using paired t-tests, Linear mixed-effects model analysis, and receiver operating characteristic (ROC) curve analysis were performed.

Results

The mean AL elongation after one year was 0.23 ± 0.15 mm. Linear mixed-effects model analysis of factors associated with one-year axial length growth revealed that baseline age (estimate = − 0.04, p < 0.001), baseline SE (estimate = 0.02, 0.004), baseline BMI (estimate = 0.01, p < 0.001), and baseline pupil diameter (estimate = − 0.08, p < 0.001) were independently associated with AL change. ROC curve analysis identified exploratory cutoff values associated with slower axial elongation during orthokeratology wear, including age 11.50 years (overall), axial length of 24.33 mm, and pupil diameter of 5.10 mm. Sex-stratified analysis suggested age-related differences, with corresponding cutoff values of 10.5 years for males and 11.5 years for females.

Conclusions

Children undergoing orthokeratology showed a mean one-year axial elongation of 0.23 mm. Older age, higher baseline myopia, larger pupil diameter and BMI were associated with axial elongation during treatment, with notable sex differences. These findings may help inform individualized orthokeratology management in pediatric myopia.

Introduction

Myopia has emerged as a significant global public health challenge, with a particularly high and escalating prevalence in East Asia [1]. In China, the prevalence of myopia among adolescents aged 16–18 years reaches 84.8%, with high myopia accounting for 19.3% [2]. The early onset and rapid progression of myopia substantially elevate the lifetime risk of pathological myopia and associated vision-threatening complications, including retinal detachment and maculopathy [3]. Consequently, effective intervention to control myopia progression in children is a critical priority in contemporary eye care.

Orthokeratology (Ortho-K), which involves the overnight wear of rigid gas-permeable contact lenses, reshapes the corneal profile. This not only corrects central refractive error but also induces relative peripheral myopic defocus, a key mechanism believed to inhibit axial elongation. Its efficacy in slowing myopia progression has been well-documented in numerous studies [4, 5]. However, significant inter-individual variability in treatment response is observed in clinical practice. Identifying and validating baseline predictive factors is crucial for personalizing treatment strategies, optimizing clinical decision-making, and managing patient expectations. This retrospective study aims to assess the one-year efficacy of Ortho-K in a cohort of myopic children and to systematically investigate key baseline characteristics that influence treatment outcomes.

Methods

Study design and participants

This retrospective, single-arm, pre-post study included 122 myopic children (244 eyes) who were first fitted with Ortho-K lenses at Jiangyin People’s Hospital between January 2024 and December 2025 and completed a one-year follow-up. The cohort comprised 62 boys (124 eyes) and 60 girls (120 eyes). Written informed consent was obtained from all participants and their legal guardians. The study protocol was approved by the Institutional Review Board of Jiangyin People’s Hospital (Approval No. 2026-KY013-01) and adhered to the tenets of the Declaration of Helsinki.

Inclusion and exclusion criteria

Inclusion criteria

(1) Age 8–14 years at initial examination; (2) Spherical equivalent (SE) refractive error between − 1.00 D and − 6.00 D; (3) Best-corrected visual acuity (BCVA) of 1.0 or better (decimal); (4) Completion of the one-year follow-up.

Exclusion criteria

(1) Previous history of contact lens wear; (2) Current or prior use of medications known to affect myopia progression, such as low-dose atropine; (3) Corneal pathology, active ocular inflammation, or history of ocular surgery/trauma; (4) Other ocular diseases (e.g., strabismus, amblyopia, cataract, glaucoma); (5) Systemic diseases potentially affecting ocular health (e.g., diabetes, immune disorders); (6) Poor lens wear adherence (average wear time < 6 h/night) or incomplete follow-up data.

Examinations and follow-up

All participants underwent comprehensive ophthalmic examinations at baseline and at the one-year follow-up visit. These assessments included uncorrected and best-corrected visual acuity, non-contact tonometry, slit-lamp biomicroscopy, fundus examination, cycloplegic autorefraction (after instilling 1% cyclopentolate three times), corneal topography (Pentacam, Oculus), corneal endothelial cell count (CEM-530, NIDEK), and optical biometry (IOL Master 500, Zeiss). Axial length (AL) was measured using the IOL Master.

Pupil diameter and baseline corneal topography were measured under standardized indoor ambient lighting conditions (photopic conditions) using a corneal topographer (Oculus, Germany). Pupil diameter was recorded separately for each eye. To minimize measurement variability, all examinations were performed under consistent illumination and fixation conditions. Three images were selected based on the manufacturer’s recommended optimal quality indices and retained for subsequent corneal topography analysis.

Ortho-K lenses (Euclid Systems Corporation, USA) were fitted according to the manufacturer’s guidelines and based on corneal parameters. Initial lens selection was guided by corneal topography data (flat and steep keratometry values, corneal astigmatism, symmetry) and refractive error. Lens fit was assessed after a settling period using fluorescein staining to evaluate lens centration, movement, and tear exchange pattern. The final lens was chosen based on good centration, adequate coverage of the pupillary zone, and an acceptable movement of 1–2 mm upon blinking. Follow-up visits were scheduled for 1 day, 1 week, 1 month after lens delivery, and then every two months thereafter. In our clinical follow-up protocol, adverse events were routinely monitored and documented at each visit through slit-lamp examination and patient/parent interviews, including corneal staining, infiltrative events, lens binding, allergic conjunctivitis, and treatment discontinuation. No serious adverse events, such as microbial keratitis, were observed during the study period.

Statistical analysis

Statistical analyses were performed using SPSS software (version 27.0, IBM Corp.). Continuous variables were expressed as mean ± standard deviation (SD). Paired t-tests were used to compare changes in parameters from baseline to the 1-year follow-up. Baseline age, SE, AL, body mass index (BMI), and pupil diameter were included as fixed effects to identify factors associated with one-year axial length growth, with estimated coefficients (β) and corresponding p-values reported. An eye was classified as a “responder” if the annual AL elongation was < 0.25 mm, and as a “non-responder” if ΔAL was ≥ 0.25 mm [6–8]. Receiver operating characteristic (ROC) curve analysis was used to evaluate the predictive ability of baseline factors for responder status. The optimal cut-off value was determined using the Youden index, and the area under the curve (AUC) was calculated. An AUC > 0.7 was considered to indicate good discriminatory ability. A p-value < 0.05 was considered statistically significant. A post hoc power analysis was performed using G*Power software (version 3.1.9.7, Heinrich Heine University Düsseldorf, Germany). The calculated statistical power exceeded 0.80, indicating that the sample size was sufficient for the present study.

Results

Baseline characteristics and treatment response

The mean age of the 122 participants was 10.76 ± 1.86 years. Baseline characteristics are summarized in Table 1. Girls had a significantly shorter baseline AL compared to boys (p < 0.001). After one year of Ortho-K wear, the overall mean AL elongation was 0.23 ± 0.15 mm, representing an approximate 40–50% reduction compared to expected annual progression rates (~ 0.40 mm/year) for untreated children in this age range [9]. Based on the responder definition (ΔAL < 0.25 mm), 150 eyes (61.5%) were classified as responders, and 94 eyes (38.5%) as non-responders. A comparison of baseline characteristics between responder and non-responder eyes is presented in Table 2. Significant differences were observed in baseline age, ΔAL, baseline AL, SE, PD, height, weight, BMI, and corneal curvature (Es). No significant changes were noted in corneal endothelial cell density, coefficient of variation in cell area, percentage of hexagonal cells, or corneal thickness after one year (data not shown), indicating good short-term safety.

Table 1.

Baseline demographic and clinical characteristics in children with myopia

Parameters	Baseline	1 Year	t	P
Eyes, n	244	244	-	-
Sex
Male, n (%)	62(50.82)	-	-	-
Female, n (%)	61(49.18)	-	-	-
Age, years	10.76 ± 1.86	-	-	-
Ks, D	43.86 ± 1.45	-	-	-
Kf, D	42.69 ± 1.31	-	-	-
Es	0.60 ± 0.12	-	-	-
Ef	0.56 ± 0.08	-	-	-
SE, D	-2.91 ± 1.45	-	-	-
IOP, mmHg	14.54 ± 2.50	-	-	-
Pupil diameter, mm	5.26 ± 0.60
AL, mm	24.58 ± 0.90	24.81 ± 0.56	-23.83	< 0.001
Height, cm	150.45 ± 13.00	155.82 ± 12.38	-23.67	< 0.001
Weight, kg	43.06 ± 10.21	48.31 ± 10.50	-15.69	< 0.001
BMI, kg/m2	18.75 ± 2.24	19.83 ± 2.95	-6.75	< 0.001
Corneal endothelial cell density, cells/mm2	3007 ± 339	2997 ± 224	+ 1.62	0.11
Corneal endothelial cell coefficient of variation	36.25 ± 19.80	35.49 ± 7.13	+ 0.56	0.55
Corneal endothelial cell hexagonality, %	54.38 ± 8.86	53.64 ± 9.13	+ 1.55	0.12
Corneal thickness, mm	531.26 ± 33.64	532.39 ± 32.53	-1.00	0.14

Significant P values < 0.05 are shown in bold. Ks, the direction and value of the meridian with the maximum refractive power; Kf, the direction and value of the meridian perpendicular to Ks; Es, the curvature value in the meridian with the maximum corneal refractive power; Ef, the curvature value in the meridian perpendicular to Es; SE, spherical equivalent; IOP, Intraocular pressure; AL, axial length; BMI, body mass index

Table 2.

Comparison of clinical characteristics between responder eyes and non-responder eyes

Parameters	Responder group	Non-responder group	t	P
Baseline age, years	11.37 ± 1.91	9.80 ± 1.30	7.62	< 0.001
AL growth, mm	0.14 ± 0.07	0.38 ± 0.13	-16.39	< 0.001
Baseline AL, mm	24.73 ± 0.87	24.34 ± 0.90	3.40	< 0.001
Baseline SE, D	-3.16 ± 1.52	-2.50 ± 1.22	-3.55	< 0.001
Baseline pupil diameter	5.47 ± 0.49	4.93 ± 0.62	7.70	< 0.001
Baseline IOP, mmHg	14.63 ± 2.48	14.41 ± 2.56	0.66	0.51
Baseline height, cm	154.87 ± 12.58	143.40 ± 10.32	7.75	< 0.001
Baseline weight, kg	45.69 ± 10.28	38.85 ± 8.60	5.61	< 0.001
Baseline BMI, kg/m²	18.82 ± 2.34	18.65 ± 2.07	0.56	0.58
Baseline corneal endothelial cell density, cells/mm²	2984 ± 376	3044 ± 265	-1.35	0.18
Baseline corneal endothelial cell coefficient of variation	36.90 ± 25.02	35.24 ± 5.09	0.62	0.54
Baseline corneal endothelial cell hexagonality, %	55.08 ± 8.56	53.30 ± 9.25	1.50	0.13
Baseline Ks, D	43.76 ± 1.49	44.00 ± 1.38	-1.23	0.22
Baseline Kf, D	42.62 ± 1.36	42.80 ± 1.24	-1.08	0.28
Baseline Es	0.59 ± 0.12	0.63 ± 0.11	-2.50	0.01
Baseline Ef	0.55 ± 0.08	0.57 ± 0.09	1.84	0.07
Baseline corneal thickness, mm	532.87 ± 31.90	528.66 ± 36.31	0.92	0.36

Significant P values < 0.05 are shown in bold. Data are presented as mean ± standard deviation. Ks, the direction and value of the meridian with the maximum refractive power; Kf, the direction and value of the meridian perpendicular to Ks; Es, the curvature value in the meridian with the maximum corneal refractive power; Ef, the curvature value in the meridian perpendicular to Es; SE, spherical equivalent; IOP, Intraocular pressure; AL, axial length; BMI, body mass index

Factors influencing axial length change

Linear mixed-effects model analysis of factors associated with one-year axial length growth revealed that baseline age (estimate = − 0.04, p < 0.001), baseline SE (estimate = 0.02, 0.004), baseline BMI (estimate = 0.01, p < 0.001), and baseline pupil diameter (estimate = − 0.08, p < 0.001) were independently associated with AL change. (Table 3).

Table 3.

Linear mixed-effects model analysis of factors associated with one-year axial length growth

Parameters	β (Estimate)	SE	t	P	95%CI
Baseline age, years	-0.04	-0.394	-6.35	< 0.001	-0.05∼-0.03
Baseline SE, D	0.02	0.008	2.90	0.004	0.01∼0.04
Baseline BMI, kg/m²	0.01	0.004	3.49	< 0.001	0.01∼0.02
Baseline pupil diameter	-0.08	-0.015	-5.27	< 0.001	-0.11∼-0.05

Significant P values < 0.05 are shown in bold. SE, spherical equivalent; BMI, body mass index

Threshold analysis for predicting treatment response

ROC curve analysis identified optimal baseline cut-off values for predicting responder status (ΔAL < 0.25 mm) in the overall cohort: age of 11.50 years (AUC = 0.74, 95% CI: 0.68–0.80), AL of 24.33 mm (AUC = 0.63, 95% CI: 0.56–0.70), and pupil diameter of 5.10 mm (AUC = 0.76, 95% CI: 0.70–0.82) (Fig. 1). Sex-stratified analysis revealed different optimal age cut-offs, with 10.5 years for males (AUC = 0.70, 95% CI: 0.61–0.79) and 11.5 years for females (AUC = 0.78, 95% CI: 0.70–0.86) (Fig. 2A, B). Children older than the identified age cut-offs, with longer baseline AL, or with larger pupils showed significantly slower AL progression (all p < 0.001 for trend).

Fig. 1.

Receiver operating characteristic (ROC) curve analysis of treatment outcomes based on baseline age, baseline pupil diameter, and baseline axial length. The cut-off points for baseline age, baseline axial length, and baseline pupil diameter were 11.50 years, 24.33 mm, and 5.10 mm, respectively, for achieving optimal treatment response

Fig. 2.

Receiver operating characteristic (ROC) curve analysis of treatment outcomes based on baseline age in males and females. A: The optimal cut-off point for predicting treatment response in males was 10.5 years; B: The optimal cut-off point for predicting treatment response in females was 11.5 years

Discussion

This retrospective study confirms the observed axial elongation was lower than values commonly reported in untreated children, although the absence of a concurrent control group limits causal interpretation. Our findings align with recent meta-analyses and large-scale studies [4, 5]. More importantly, this study highlights substantial inter-individual variability in axial elongation during orthokeratology wear and identifies several baseline factors—including age, refractive error, pupil diameter, and BMI—that were independently associated with axial length growth during treatment.

Baseline factors associated with axial elongation: implications for patient selection

Consistent with previous research, [9] baseline age emerged as the strongest independent predictor of AL change (β = −0.394, p < 0.001). Children older than the identified threshold of 11.5 years demonstrated significantly slower annual AL elongation. This is partly attributable to the natural deceleration of myopia progression with age [10]. However, our sex-specific findings (male threshold 10.5 years, female 11.5 years) suggest a more nuanced interaction, potentially linked to the differential timing of puberty and associated hormonal changes (e.g., growth hormone, estrogens) which may influence scleral remodeling and ocular growth dynamics [11]. These possible mechanisms remain speculative and were not directly examined in the present study, but may warrant further investigation into potential neuro-endocrine-ocular pathways involved in myopia control.

The association between higher baseline myopic SE (greater magnitude) and slower axial elongation during orthokeratology wear (β = 0.226, p = 0.011) is consistent with the proposed optical mechanism of Ortho-K. A higher degree of myopia requires a greater magnitude of central corneal flattening, which in turn induces a more pronounced steepening of the mid-peripheral cornea. This creates a stronger and more extensive relative peripheral myopic defocus signal [12, 13]. The effect might exhibit a ceiling effect at very high myopia where corneal biomechanics limit further shape change, a point warranting investigation in future studies with broader SE ranges.

Our results strongly corroborate the role of pupil diameter as a key effect modifier. Baseline pupil diameter > 5.10 mm was associated with slower axial elongation during orthokeratology wear (AUC = 0.76). The optical rationale is compelling: under mesopic conditions (e.g., overnight wear), a larger pupil allows more peripheral light rays, which have undergone greater refraction through the treated corneal zone, to enter the eye. This enhances the signal strength of relative peripheral myopic defocus on the retina, thereby amplifying the inhibitory effect on axial growth [14, 15]. This finding reinforces the dose-response relationship between the defocus signal and treatment efficacy.

BMI as a potential marker of lifestyle and systemic factors: implications for myopia management

A novel and clinically significant finding of our study was the independent association between higher baseline BMI and greater axial elongation during orthokeratology wear. This suggests that systemic metabolic state may be a modifiable factor influencing ocular growth, an area garnering increased attention [16, 17]. The mechanisms are likely multifactorial, involving both behavioral and metabolic pathways.

The association between higher BMI and greater axial elongation may reflect broader lifestyle or systemic influences rather than a direct mechanistic effect. Behaviorally, Higher BMI is often correlated with factors such as reduced outdoor activity, lower physical activity, dietary patterns, and increased near-work, all of which may contribute to myopia progression [18, 19]. Metabolically, pathways involving extracellular matrix remodeling (e.g., matrix metalloproteinases [MMPs]), insulin resistance, and IGF-1–related signaling may also contribute, although these mechanisms remain speculative and were not directly assessed in the present study [20, 21]. Further prospective studies incorporating behavioral and metabolic measures are needed to clarify whether BMI has an independent role in axial elongation during orthokeratology wear.

Clinical implications and personalized cut-offs

The baseline thresholds derived from ROC analysis (age 11.5 years, AL 24.33 mm, PD 5.10 mm) offer a practical, evidence-based framework for pre-treatment counseling. Children meeting these “optimal profile” criteria can be counseled to expect a high likelihood of excellent control. Conversely, for children with less favorable baseline characteristics (e.g., younger age, shorter AL, smaller pupils, higher BMI), clinicians should set realistic expectations, emphasize the importance of lifestyle modifications, and consider more frequent monitoring (e.g., at 3–6 months). For those showing rapid progression despite Ortho-K alone, early adoption of combination therapy, such as Ortho-K plus low-dose atropine (particularly higher concentrations), may be warranted, as emerging evidence suggests a synergistic effect [22, 23].

Ocular safety

In line with established literature [4], we observed no significant changes in corneal endothelial cell density or morphology after one year of Ortho-K wear, supporting its short-term safety profile. However, long-term studies using more sensitive imaging techniques like confocal microscopy are needed to fully elucidate potential subtle effects on corneal microstructure (e.g., stromal keratocyte density, sub-basal nerve morphology) over extended wear periods [24]. The low risk of serious adverse events like microbial keratitis, when stringent hygiene and follow-up protocols are followed, remains a key tenet of Ortho-K practice.

Study limitations

The findings should be interpreted in the context of several limitations. First, the single-center, retrospective design is susceptible to selection bias, and the lack of a concurrent control group limits causal inference, although the comparison with expected progression rates is informative. Second, Treatment-related parameters, including treatment zone size, lens decentration, and corneal topographic response, were not systematically recorded and may partly explain variability in axial elongation outcomes. Third, no specific exclusion criterion was applied for anisometropia, which may have contributed to inter-eye variability despite the use of a linear mixed-effects model to account for within-subject correlation. Fourth, the one-year follow-up period, while sufficient to assess short-term efficacy, cannot capture the durability of these predictive factors over the longer term (e.g., 3–5 years). Fifth, pupil diameter was measured only at baseline under standardized conditions; diurnal variation and changes during treatment could modulate the effect. Finally, the model’s R² of 0.48 indicates that other unmeasured factors (e.g., genetic predisposition, precise outdoor time, near-work parameters, parental myopia) contribute to the variability in response and external validation in independent prospective cohorts will be necessary before these cut-offs can be considered for broader clinical application. Prospective, multicenter studies with larger, diverse cohorts and more comprehensive data collection are necessary to develop robust, multi-factorial prediction models for personalized myopia management.

Conclusion

In conclusion, this study confirms the one-year efficacy of orthokeratology in slowing axial elongation, with axial elongation rates lower than those commonly reported in untreated children, although the absence of a concurrent control group limits causal interpretation. One-year axial elongation during orthokeratology wear was significantly associated with baseline age, refractive error, pupil diameter, and BMI. Older age (> 11.5 years), larger pupil diameter (> 5.10 mm), and longer baseline axial length (> 24.33 mm) were associated with slower axial elongation, with exploratory age-related differences observed between males and females. In addition, the independent association between higher baseline BMI and greater axial elongation suggests that broader lifestyle or systemic factors may contribute to variability in treatment outcomes.These findings may provide useful reference points for individualized clinical assessment, helping to inform treatment planning, set realistic expectations, and identify children who may benefit from closer monitoring or additional interventions. Future prospective, long-term studies incorporating detailed metabolic and behavioral phenotyping are essential to further refine predictive models and advance precision medicine in pediatric myopia control.

Author contributions

S.C.: Conceptualization, Formal Analysis, Investigation, Resources, Writing - original draft, Writing - review &editing. X.Z.: Conceptualization, Data curation, Methodology, Funding acquisition, Writing - review & editing. W.X.: Formal Analysis, Investigation. S.C., W.X., M.Z.: Resources. S.C. and M.Z.: Investigation. X.X.: Funding acqui-sition. X.X. and G.L.: Conceptualization, Funding acquisition, Project administration, Supervision, Writing - review &editing. The final version of the manuscript was approved by all the authors.

Funding

National Natural Science Foundation of China (82070936); Jiangyin Science and Technology Innovation Special Fund Research Grant Project (JY0603A011014250004PB).

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Written informed consent was obtained from all participants and their legal guardians. The study protocol was approved by the Institutional Review Board of Jiangyin People’s Hospital (Approval No. 2026-KY013-01) and adhered to the tenets of the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.