Orthokeratology for stable mild-to-moderate keratoconus: a pilot study on safety and corneal remodeling via quantitative OCT analysis

Abstract

Objective

To assess the efficacy and safety of overnight orthokeratology (Ortho-K) in patients with stable, mild-to-moderate keratoconus and myopia, with a focus on epithelial and Bowman’s layer structural changes.

Methods

A prospective single-center cohort study enrolled 13 patients (24 eyes) with stable, mild-to-moderate keratoconus (Amsler-Krumeich grades I-II, age 14–21 years) who were fitted with Euclid Emerald Ortho-K lenses and followed for 18 months to evaluate clinical outcomes. Primary outcome measures included uncorrected visual acuity (UCVA), best-corrected visual acuity (BCVA), refractive error, axial length, corneal topographic parameters, corneal and corneal epithelial thickness (central, nasal, temporal), and quantitative analysis of the cross-sectional area of Bowman’s layer using optical coherence tomography (OCT). Treatment-related adverse events and patient dropout were also recorded.

Results

After 18 months of intervention, UCVA improved significantly from logMAR 0.59 ± 0.29 to 0.19 ± 0.16 (p < 0.05). The mean myopic refraction significantly decreased from − 2.46 ± 1.25 D to − 0.62 ± 0.24 d (p < 0.05). Characteristic corneal epithelial remodeling was observed: significant thinning in the central zone (from 54 ± 9 to 41 ± 8 μm, p < 0.05) and significant thickening in the nasal and temporal zones (from 51 ± 11 to 64 ± 17 μm and from 62 ± 13 to 71 ± 9 μm, respectively; p < 0.05 for both). The key safety finding was the absence of statistically significant changes in the cross-sectional area of Bowman’s layer at the nasal and temporal locations before and after treatment (p > 0.05). Axial length remained stable (mean elongation: 0.14 ± 0.13 mm, p = 0.583). Total higher-order aberrations (HOA) showed a decreasing trend that did not reach statistical significance. Five of the 18 initially enrolled patients discontinued treatment, primarily due to visual symptoms (nocturnal glare, halos) or tolerance issues (dry eye, handling difficulties). A total of 23 episodes of mild corneal punctate staining were recorded, all of which resolved completely following temporary lens discontinuation and pharmacological treatment, with no serious complications.

Conclusion

In rigorously selected patients with stable, mild-to-moderate keratoconus, overnight orthokeratology induces predictable epithelial remodeling—characterized by central thinning and paracentral thickening—leading to significant improvements in uncorrected visual acuity and refractive stability. Critically, high-resolution OCT revealed no significant change in the cross-sectional area of Bowman’s layer over 18 months, suggesting that this intervention does not compromise deep stromal structural integrity. These findings support orthokeratology as a viable, non-surgical adjunct for daytime spectacle independence in a highly selected ectatic population, provided that stringent fitting protocols and longitudinal structural monitoring are implemented.

Introduction

Keratoconus is a progressive corneal ectatic disorder characterized by stromal thinning and conical protrusion, typically emerging during adolescence and stabilizing after the age of 30–40 [1, 2]. With a reported male-to-female ratio of approximately 3:1, this condition leads to significant visual impairment, adversely affecting patients’ daily activities and quality of life [3]. Consequently, effective vision correction strategies remain a central focus in keratoconus management.

Rigid gas permeable contact lenses (RGPCLs) have been widely adopted as a primary nonsurgical intervention for improving visual quality in early to moderate stages [4]. Longitudinal studies suggest that long-term RGPCL wear does not exacerbate disease progression and may even help maintain corneal stability [5]. However, patient tolerance is often limited due to discomfort and mechanical complications, highlighting the need for alternative approaches [6].

Orthokeratology (ortho-K), which employs specially designed rigid lenses to temporarily reshape the cornea during sleep, has proven effective for myopia control. This corneal reshaping mechanism theoretically addresses the irregular astigmatism central to keratoconus visual impairment. Despite its distinct mode of action from conventional RGPCLs, the safety and efficacy of ortho-K in patients with pathological corneal ectasia remain largely unexplored and potentially controversial, given the altered biomechanical properties of the keratoconic cornea.

This study explores an off-label yet clinically relevant application of ortho-K lenses in a challenging ectatic population by systematically evaluating the clinical outcomes and safety profile of ortho-K lenses in patients with stable, mild-to-moderate keratoconus. We will assess visual acuity, corneal topographic changes, and adverse event rates over a 18-month follow-up period, thereby providing preliminary evidence for its potential role as a novel therapeutic option.

Methods

This was a single-center, prospective cohort study that evaluated the outcomes of Ortho-K lens wear over an 18-month follow-up period.Inclusion criteria were: (1) stable, mild-to-moderate keratoconus (grade I or II based on the Amsler-Krumeich classification) [7] with stability defined as the absence of disease progression for at least 12 months prior to enrollment. Progression criteria were: an increase in Kmax of ≥ 1.0 D within 1 year, a change in astigmatism of ≥ 3.0 D within 6 months, or the need for new contact lens fitting more than once in 2 years [8] (2) age ≥ 14 years; (3) minimum corneal thickness of 420 μm; (4) myopia ranging from plano to − 4.00 diopters (D); (5) refractive astigmatism ≤ 8.00 Dand (6) steep keratometry (K) value < 52.00 D. Exclusion criteria comprised severe dry eye disease, history of herpetic eye disease, active anterior or posterior segment pathologies, autoimmune diseases, and myopia exceeding − 4.00 D.

At the Ophthalmology Department of Tongde Hospital of Zhejiang Province, all included patients received ortho-K lenses (Euclid Systems Corporation, Sterling, VA, USA) and were followed up for 18 months.

Procedures for ortho-K lens fitting

Ortho-K lenses were fitted for a cohort of 13 patients, comprising a total of 24 treated eyes. The participants had a mean age of 17.68 ± 2.36 years (range: 14–21). Treatment was bilateral in 11 patients (22 eyes), while the remaining 2 patients received unilateral treatment (one right eye and one left eye).

Over an 18-month follow-up, all patients were managed with Euclid Emerald ortho-K lenses. The lens prescription was guided by the need to accommodate the paracentral corneal protrusion, necessitating a toric design for every case. Peripheral corneal curvature was quantified via axial mapping with a Medmont E300USB topographer, and these measurements directly informed the selection of the trial lens base curve (K value).

A structured follow-up protocol was implemented: visits were scheduled at 1 day, 1 week, and 1 month post-fitting. Thereafter, the interval was extended to every three months. To monitor tolerability, patients were contacted monthly to report any discomfort. They were also provided with the researcher’s direct contact details to report any adverse events occurring between scheduled visits. All participants were instructed to wear the lenses nightly for 7 to 9 h during sleep.

The study involved comprehensive ophthalmic evaluations. Primary assessments focused on visual function and ocular biometrics, encompassing best-corrected visual acuity (BCVA), uncorrected visual acuity (UCVA), subjective refraction, and axial length. Corneal status was evaluated using both corneal topography and slit-lamp examination. Additionally, each visit included a detailed assessment of the ortho-K lens fit and documentation of any lens-related complications.

All visual acuity assessments were conducted at a standard testing distance of 5 m. Subjective refraction was carried out monocularly under non-cycloplegic conditions using a Tomey RC-500 autorefractor (Japan), adhering to the “maximum plus for best visual acuity” endpoint to determine optimal spherical correction. In cases where the autorefractor failed to provide a reliable myopic refraction, the refractive error was measured in the less-affected eye. The value for the fellow eye was then derived by applying the interocular axial length difference to this measurement.

Axial length measurements were obtained with a non-contact partial coherence interferometer (IOLMaster-500; Carl Zeiss, Germany). To acquire corneal parameters, topography was performed by a trained technician using the Medmont E300USB topographer (Medmont Corporation, Australia).Furthermore, a Mediworks S39H slit lamp (Shanghai Meibo Precision Instrument Co., Ltd., China) was utilized for all key clinical procedures, including ortho-K lens fitting, fluorescein staining evaluation, and follow-up examinations.

Higher-order aberrations (HOAs) were quantified across the central 6-mm corneal zone with the Medmont corneal topographer. A single, trained investigator (ZC) was tasked with performing all measurements under standardized scotopic conditions. Relevant Zernike coefficients were computed automatically by the instrument’s integrated software. These included vertical coma, horizontal coma, vertical trefoil, oblique trefoil, spherical aberration, and the root mean square of higher-order aberrations (RMS-HOA).

Analysis of Topographic and OCT Images Follow-up topographic analysis consistently demonstrated a characteristic pattern: central corneal thinning, peripheral thickening, and a defined ortho-K treatment zone (comprising base and reverse curves) with a radius of approximately 3.1 mm in each eye. Quantitative assessment was performed by measuring corneal and epithelial thickness at two defined locations using optical coherence tomography (OCT, Zeiss, Germany):the corneal apex (defined as the geometric center of the cornea, corresponding to the corneal light reflex on baseline topography); and a point 3.1 mm distant from the apex along the horizontal meridian (nasal and temporal directions).The 3.1 mm distance was chosen because it matches the outer boundary of the ortho-K treatment zone identified on topography (mean radius: 3.1 mm), where the reverse curve exerts maximal epithelial redistribution. The horizontal meridian was selected due to the lens’s symmetric design and previous evidence of pronounced remodeling along this axis. To ensure reproducibility, the corneal apex served as a fixed reference for all follow-up measurements. This redistribution involves centrifugal migration of epithelial cells from the central zone toward the mid-peripheral region. As a result, central corneal epithelial thickness decreases, while epithelial thickness increases in the mid-peripheral reverse curve area. The limbal region itself maintains its structural integrity and stem cell function, continuously supplying new cells to sustain this redistributed epithelial profile during long-term lens wear. Subsequently, the obtained OCT images underwent processing and analysis utilizing Image-Pro Plus software (Version 6.0). The resulting data are presented in Fig. 1 and Table 1.

Fig. 1.

The blue arrow denotes Bowman’s layer. The red line delineates the corneal epithelial thickness (μm)

Table 1.

Assessment of corneal and epithelial thickness at multiple corneal locations (μm)

	Baseline	1 mo	3 mo	6 mo	9 mo	12 mo	15 mo	18 mo
Central corneal thickness	511 ± 23	501 ± 12	491 ± 13	497 ± 9	488 ± 24	492 ± 12	496 ± 16	498 ± 12
Nasal corneal thickness	486 ± 47	493 ± 31	489 ± 22	499 ± 36	502 ± 16	496 ± 36	506 ± 8	501 ± 23
Temporal corneal thickness	503 ± 31	513 ± 22	509 ± 18	518 ± 21	522 ± 28	513 ± 31	511 ± 23	512 ± 19
Central corneal epithelial thickness	54 ± 9	48 ± 11	47 ± 6	42 ± 13	39 ± 17	44 ± 9	42 ± 12	41 ± 8
Nasal corneal epithelial thickness	51 ± 11	59 ± 13	66 ± 12	67 ± 7	59 ± 12	60 ± 13	62 ± 9	64 ± 17
Temporal corneal epithelial thickness	62 ± 13	66 ± 22	69 ± 14	65 ± 12	77 ± 12	67 ± 7	72 ± 13	71 ± 9a

mo, month

The cross-sectional area of Bowman’s layer was quantified through analysis of OCT images [9, 10], with results detailed in Fig. 2 and Table 2. To delineate the layer’s contour, the OCT image analysis employed graph theory combined with Dijkstra’s algorithm [11]. The segmented anterior boundary of Bowman’s layer is indicated by a blue line in the figure. Subsequently, the posterior boundary of Bowman’s layer was delineated using the same method and is indicated by a yellow line in Fig. 2b. The distance between the blue line and the yellow line represents the thickness of Bowman’s layer. Figure 2c illustrates the region enclosed between these two lines, with the area between them calculated and represented by the red-shaded portion in Fig. 2c [10].

Fig. 2.

The blue line and blue arrow in a shows an OCT image of the cornea. b indicates the segmented anterior boundary of Bowman’s layer. The yellow line and yellow arrow in B indicates the segmented posterior boundary of Bowman’s layer. The distance between the blue line and the yellow line represents the thickness of Bowman’s layer. The red-shaded area between these two lines in c represents the cross-sectional area of Bowman’s layer

Table 2.

The cross-sectional area of the Bowman’s layer (mm²)

	Baseli	1 mo	3 mo	6 mo	9 mo	12 mo	15 mo	18mo
Nasal	1.83 ± 0.43 × 10–2	1.72 ± 0.23 × 10–2	1.78 ± 0.38 × 10–2	1.74 ± 0.51 × 10–2	1.67 ± 0.41 × 10–2	1.61 ± 0.33 × 10–2	1.71 ± 0.23 × 10–2	1.65 ± 0.22 × 10–2
Tempo ral	2.02 ± 0.34 × 10–2	2.08 ± 0.24 × 10–2	1.97 ± 0.33 × 10–2	1.92 ± 0.19 × 10–2	1.95 ± 0.23 × 10–2	1.93 ± 0.11 × 10–2	1.86 ± 0.29 × 10–2	1.89 ± 0.18 × 10–2

mo, month

The Quality of Vision (QoV) questionnaire was used to assess ten distinct visual symptoms: glare, halos, hazy vision, starbursts, blurred vision, double or multiple images, distortion, fluctuating vision, focusing difficulties, and impaired depth or distance perception. For each symptom, three aspects were rated: frequency (0 = never, 1 = occasionally, 2 = quite often, 3 = very often), severity (0 = not at all, 1 = mild, 2 = moderate, 3 = severe), and bothersomeness (0 = not at all, 1 = a little, 2 = quite, 3 = very) [10]. In addition, participants rated their overall satisfaction with the procedure on a six-point scale ranging from 0 (very dissatisfied) to 5 (very satisfied), with intermediate categories including dissatisfied, general, less satisfied, and satisfied.

For statistical analysis in this study, SPSS version 25.0 (SPSS Inc., Chicago, IL, USA) was used, with a significance level set at p < 0.05. Paired t-tests were applied to assess the statistical significance of differences in corneal parameters before and after ortho-K lens fitting. To account for inter-eye correlation within subjects, a linear mixed-effects model was employed. P values for multiple comparisons across follow-up time points were adjusted using the Bonferroni method. The inherent intra-subject correlation between eyes was handled by conducting a paired-samples analysis on the bilateral data.

The research protocol has received approval from the Ethics Committee at Tongde Hospital, with approval number: Zhetongderunshen 2024 Research No. 184-JY, and it adheres to the guidelines outlined in the Helsinki Declaration. Written informed consent was obtained from all participants (and from their legal guardians for participants under the age of 18) prior to enrollment. This study was initiated in 2024 as a small-sample exploratory pilot investigation. At the time of study design, prospective registration was not mandatory for such pilot studies at our institution, and thus the trial was not registered in a public registry. All data are traceable and the authors take full responsibility for the integrity of the findings.

Result

Over 18 months, Ortho-K significantly improved UCVA and reduced myopic refraction, mediated by significant central epithelial thinning and peripheral epithelial thickening. Crucially, this remodeling occurred without altering the cross-sectional area of Bowman’s layer (all p > 0.05). Axial length remained stable.

Prior to ortho-K lens wear, the mean uncorrected visual acuity (UCVA) was logMAR 0.59 ± 0.29. This improved significantly to logMAR 0.19 ± 0.16 following lens wear. UCVA improved by 0.40 logMAR units (95% CI: 0.28–0.52; Cohen’s d = 1.38).The mean refractive error before fitting was − 2.46 ± 1.25 diopters (D) for myopia and − 3.72 ± 2.42 D for astigmatism. At the final examination, a significant reduction in myopia was observed, with a mean value of − 0.62 ± 0.24 D (p < 0.05). Compared to baseline, the mean refractive astigmatism demonstrated a significant decrease to − 2.02 ± 0.89 D at the final visit (p < 0.05). Notably, these changes occurred within the first week of ortho-K lens wear, stabilized within one month, and were maintained throughout the entire follow-up period (Fig. 3). The mean axial elongation was 0.14 ± 0.13 mm in all treated eyes. No statistically significant change in axial length was observed before and after the treatment (p = 0.583).

Fig. 3.

Corneal topographic changes following ortho-K for keratoconus

Before lens fitting, the average central corneal thickness was 511 ± 23 µm, while the nasal and temporal corneal thicknesses were 486 ± 47 µm and 503 ± 31 µm, respectively. The average central corneal epithelial thickness was 54 ± 9 µm; the nasal and temporal corneal epithelial thicknesses were 51 ± 11 µm and 62 ± 13 µm, respectively. At the final follow-up, the central corneal thickness had decreased to 498 ± 12 µm, a change that was not statistically significant compared to pre-fitting values. Changes in overall corneal thickness were non-significant, with nasal and temporal measurements reaching 501 ± 23 µm and 512 ± 19 µm, respectively. Conversely, corneal epithelial thickness exhibited a distinct remodeling pattern: the central region demonstrated significant thinning to 41 ± 8 µm (p < 0.05). In comparison, significant epithelial thickening was observed in the nasal (64 ± 17 µm, p < 0.05) and temporal (71 ± 9 µm, p < 0.05) regions.

The cross-sectional area of the Bowman layer is used to quantitatively assess microstructural distortions of the cornea in conditions such as keratoconus [9, 10]. Prior to lens fitting, the nasal and temporal areas measured (1.83 ± 0.43) × 10⁻² mm² and (2.02 ± 0.34) × 10⁻² mm², respectively. At the final follow-up, the values were recorded as (1.65 ± 0.22) × 10⁻² mm² and (1.89 ± 0.18) × 10⁻² mm², indicating a numerical decrease from baseline that did not reach statistical significance (p = 0.567).

At baseline, the measured values were: spherical aberration, 0.017 ± 0.004 μm; coma, − 0.006 ± 0.003 μm; and the root mean square (RMS) of higher-order aberrations (HOA), 7.32 ± 2.38 μm. After 18 months, the corresponding values were: spherical aberration, 0.011 ± 0.003 μm; coma, 0.005 ± 0.002 μm; and RMS HOA, 5.62 ± 1.54 μm. Comparative analysis revealed no statistically significant differences between baseline and 18-month measurements for spherical aberration (p > 0.05), coma (p > 0.05), or total RMS HOA (p > 0.05). Although the RMS HOA showed a declining trend, this change did not achieve statistical significance, potentially due to considerable inter-subject variability reflected in the large standard deviations.

The primary visual symptoms before wearing orthokeratology (OK) lenses were blurred vision (95.56%), glare (82.29%), and starbursts (67.32%). At 18 months after wearing OK lenses, the main symptoms reported were fluctuation in vision (75.64%), double or multiple images (72.26%), and halos (59.16%). However, a statistically significant difference was observed in blurred vision before and after wearing OK lenses (P < 0.05), while no significant differences were found in other individual symptom scores or total scores. No significant correlations were detected between overall Quality of Vision (QoV) scores, overall satisfaction, this observation remained consistent across different diopter subgroups (all P > 0.05).

To further characterize the extent of corneal morphological changes induced by orthokeratology, we calculated the following indices before and after 18 months of lens wear: central corneal curvature (K-value), inferior–superior asymmetry (I-S), corneal astigmatism (AST), skew percentage of the steepest radial axis (SRAX), and the KISA% index [12]. The results are presented in Table 3.

Table 3.

Topographic index values

	Before	After	P value
K value	52.78 ± 9.78	48.52 ± 10.32	< 0.01
I-S value	3.62 ± 3.04	2.82 ± 2.95	> 0.05
AST	3.32 ± 1.79	3.86 ± 3.45	> 0.05
SRAX	15.78 ± 16.37	13.52 ± 16.51	> 0.05
KISA%	9327.57 ± 10,783.37	8737.67 ± 9283.37	> 0.05

mo, month

Of the 18 patients initially enrolled and fitted with orthokeratology lenses, 5 discontinued treatment during the 18-month follow-up period, leaving 13 patients (24 eyes) who completed the full follow-up for final analysis. Two of these patients reported a constellation of visual symptoms, such as impaired night vision, increased glare, halos around light sources, and monocular diplopia [13]. One patient withdrew due to lens removal difficulty caused by dry eye, and one withdrew because of dry eye syndrome. The remaining volunteer withdrew due to an inability to adapt to the discomfort caused by rigid ortho-K lenses. None of the patients required lens replacement during the follow-up period.

During the course of treatment, a total of 5 cases involving 23 episodes of mild corneal staining were observed. Recombinant bovine basic fibroblast growth factor eye drops were administered three times daily, and lens wear was discontinued for one week. Upon verification of full corneal healing at follow-up, Ortho-K wear was successfully reinstated without sequelae. Notably, no other corneal adverse events or lens-related issues necessitating discontinuation occurred among the remaining participants. Moreover, all lenses retained their therapeutic effectiveness for the duration of the study, and no replacements were required.

Discussion

This study demonstrates that overnight ortho-K lens wear in patients with mild to moderate keratoconus induces characteristic corneal epithelial remodeling—central thinning and peripheral thickening—without causing a significant change in the cross-sectional area of Bowman’s layer. Given that the core pathology of keratoconus involves structural disorganization and biomechanical weakening of the stromal collagen [14], these findings suggest that the mechanical effect of ortho-K may be primarily confined to the epithelial layer, without imposing a measurable additional impact on the already compromised deep stromal architecture.

Bowman’s layer is an acellular collagen layer beneath the epithelial basement membrane that maintains anterior corneal stiffness. Structural disruption of this layer occurs early in keratoconus [15]. Thus, it is critical to assess whether orthokeratology affects Bowman’s layer. This study is the first to systematically evaluate the long-term impact of orthokeratology on the cross-sectional area of Bowman’s layer in keratoconus patients using high-resolution OCT. After 18 months, Bowman’s layer parameters remained stable, indicating that the mechanical forces are primarily confined to the epithelium and superficial stroma, without disrupting Bowman’s layer—a marker of deeper structural integrity. These findings support the layer-specific mechanical effects of orthokeratology and alleviate safety concerns that the therapy might exacerbate underlying stromal weakness in keratoconus.

The corneal epithelial remodeling pattern observed in this study—characterized by significant central thinning and compensatory mid-peripheral thickening—is consistent with the classic changes induced by Ortho-K in healthy myopic eyes [16] This finding confirms that even in keratoconic corneas with underlying stromal abnormalities, the corneal epithelium retains its plasticity in response to sustained mechanical stimuli. Notably, this remodeling exhibits a spatially complementary relationship to the typical pathological changes in early keratoconus, namely epithelial thinning at the cone apex accompanied by reactive thickening in the surrounding region [17]. The remodeled epithelial thickness profile remained stable over the 18-month follow-up period, suggesting that this active epithelial redistribution may enhance peripheral biomechanical support, partially counteracting the abnormal thickness distribution driven by anterior stromal protrusion. This mechanism could contribute to the stabilization of overall corneal architecture and may partly explain the clinical safety profile of Ortho-K observed in this cohort..

The primary goal of contact lens therapy is to effectively enhance visual function while ensuring safety. The results of this study demonstrate that after wearing ortho-K lenses, patients’ uncorrected visual acuity and refractive error showed rapid and sustained significant improvement. This functional benefit closely corresponds to the formation of a regular “treatment zone” observed in corneal topography, confirming the efficacy of this therapy in reshaping the anterior corneal surface to correct irregular astigmatism and improve optical quality [18]. Although total higher-order aberrations showed a clinically meaningful decreasing trend, the change did not reach statistical significance. This result is likely due to the substantial inter-subject variability in higher-order aberrations among keratoconus patients, combined with the limited sample size of this study, leading to insufficient statistical power. Nevertheless, this does not diminish the substantial clinical value of Ortho-K in refractive control and visual improvement.

Among the five patients who discontinued treatment, two withdrew due to visual symptoms (nocturnal glare, halos, monocular diplopia, and fluctuating vision). Fluctuating vision was the most frequently reported symptom at 18 months (75.64%), reflecting its clinical significance. These visual disturbances likely result from the inherent corneal irregularity in keratoconus, which can compromise lens centration and create an irregular treatment zone. Such irregularities are known to induce higher-order aberrations and degrade visual quality, particularly causing fluctuations that vary with tear film dynamics and pupil size [13]. Two additional patients withdrew due to dry eye-related issues. Keratoconus is frequently associated with dry eye disease [19], and orthokeratology lens wear may further compromise tear film stability [20], contributing to ocular discomfort and lens handling difficulties. Collectively, these findings suggest that visual quality and ocular surface tolerability are key determinants of long-term adherence, underscoring the importance of careful patient selection, individualized lens fitting, and proactive dry eye management.

Overnight ortho-K may protect against keratoconus progression by mitigating nocturnal mechanical pressure from sleep posture, with the rigid lens potentially redistributing pressure away from the cone apex. This aligns with our observation of no progression in this adolescent cohort despite high-risk age, hinting at a disease-modifying effect beyond optical correction. The lack of significant HOA reduction reflects ortho-K’s inherent induction of coma and spherical aberration, explaining why total HOAs did not decrease despite improved vision—a trade-off that underscores the need for patient selection and expectation management. Looking forward, freeform ortho-K designs offer promise for better managing irregular astigmatism in keratoconus, and future studies should assess whether they can reduce HOAs while maintaining epithelial remodeling benefits, potentially expanding therapeutic options.

Contact lenses represent a cornerstone of nonsurgical management for keratoconus. Long-term wear of RGPCLs offers established efficacy and safety in controlling disease progression and improving visual quality, functioning as an external optical surface without permanently altering corneal structure [21]. In contrast, Ortho-K aims to achieve daytime spectacle independence by inducing reversible corneal epithelial remodeling through overnight lens wear. These two modalities embody different safety paradigms, yet the evidence presented here indicates that in carefully selected patients with stable, mild-to-moderate keratoconus, Ortho-K also demonstrates a considerable safety profile—its mechanical effect appears confined primarily to the epithelial layer without compromising the structural integrity of Bowman’s layer.

Therefore, Ortho-K should not replace conventional RGPCLs but serve as a valuable complementary strategy. It may be particularly suitable for patients intolerant to daytime RGPCL wear or those strongly desiring complete daytime optical freedom [22]. The prerequisite for successful Ortho-K application is ensuring that remodeling remains strictly limited to the epithelial layer, supported by standardized structural follow-up including Bowman’s layer monitoring. This approach seeks to improve visual acuity while safeguarding the already compromised corneal architecture.

Several limitations of this study should be acknowledged. First, as a small-sample pilot study, it was not prospectively registered in a public clinical trials registry, which may affect transparency and reproducibility. Future studies should undergo prospective registration to enhance the level of evidence. Second, corneal biomechanical properties—including corneal hysteresis (CH), corneal resistance factor (CRF), and intraocular pressure (IOP)—were not assessed. These parameters, measured by devices such as the Ocular Response Analyzer (ORA) or Corvis ST [23], are critical for understanding the biomechanical behavior of the keratoconic cornea in response to overnight lens wear. Future studies should incorporate corneal biomechanical evaluation to better elucidate this interaction and refine patient selection criteria.Third, epithelial and Bowman’s layer measurements were limited to nasal and temporal locations along the horizontal meridian. Given that keratoconus cones are most commonly located in the inferior or inferotemporal region, the selected measurement sites may not fully capture structural changes in the most affected area. Future studies incorporating a wider sampling area, including the inferior and inferotemporal quadrants, are warranted to comprehensively evaluate regional variations in corneal remodeling. Additionally, because the Medmont topographer does not natively report the skew percentage of the steepest radial axis (SRAX), we derived an equivalent SRAX value by manual measurement from the axial maps. This method, although practical, may introduce minor measurement variability and should be considered when interpreting the KISA% results.

Based on an 18-month follow-up, this study preliminarily confirms that overnight Ortho-K is an effective and relatively safe non-surgical option for rigorously selected patients with stable, mild-to-moderate keratoconus. Its primary mechanism involves reversible epithelial remodeling—central thinning and peripheral thickening—which improves uncorrected visual acuity and refractive error without significantly affecting Bowman’s layer integrity. Although some patients withdrew due to visual symptoms or poor tolerance, overall adverse events were mild and manageable. The systematic imaging evidence provided here offers new insights into the mechanism and safety boundaries of Ortho-K in structurally compromised corneas. Successful application relies on personalized fitting, strict patient selection and education, and standardized follow-up including corneal structural monitoring. Further long-term prospective studies are needed to establish its role in the comprehensive management of keratoconus.

Author contributions

All authors contributed intellectually to this work. Specifically, C.Z. and B.W. were responsible for the study conception and execution. Data acquisition and curation were handled by C.Z., Z.X.H., and X.Z.. The interpretation and analysis of data were performed by C.L.H. and C.Z. The initial manuscript was drafted by C.Z. and Z.X.H., and subsequently reviewed and approved by C.Z. and X.Z. All authors have read and endorsed the final version of the manuscript.

Funding

This work was supported by a grant from the Zhejiang Provincial Administration of Traditional Chinese Medicine (2026ZL0233).

Data availability

The supporting data and materials for the findings of this study are accessible from the corresponding author upon reasonable request, by contacting zhangchang654@163.com.

Declarations

Conflict of interest

The authors have no financial interest in Euclid Systems Corporation, and the lenses used were part of standard clinical care, not provided by the manufacturer for research purposes. The authors declare no competing interests.

Ethics approval and consent to participate

The research protocol has received approval from the Ethics Committee at Tongde Hospital, with approval number: Zhetongderunshen 2024 Research No. 184-JY, and it adheres to the guidelines outlined in the Helsinki Declaration. Written informed consent was obtained from all participants (and from their legal guardians for participants under the age of 18) prior to enrollment.

Consent for publication

Not applicable.