Significant Improvement in High Myopic Anisometropic Amblyopia Treated With Repeated Low-Level Red-Light (RLRL) Therapy: A Case Report

Introduction

High anisometropia is defined as a spherical equivalent difference greater than 3 diopters (D) between the 2 eyes, which can cause binocular vision imbalance and amblyopia if not treated promptly [1–3]. Traditional treatments include spectacle correction, occlusion therapy, and atropine penalization, but their efficacy in highly myopic anisometropic cases is limited [2,4–6]. Moreover, no current nonsurgical treatment can reduce high anisometropia, which is the most significant cause of spectacle discomfort and binocular vision dysfunction [7–9]. Repeated low-level red-light (RLRL) therapy, a novel approach for myopia management, has shown potential in controlling myopia progression by regulating choroidal thickness and scleral remodeling [10–13]. Historically, red light was used as a therapeutic intervention for amblyopia [14]. In the present study, we explore whether red light therapy potentially offers unanticipated benefits for patients with high myopic anisometropic amblyopia, a population characterized by both myopia control and amblyopia treatment needs. We describe a pediatric case of high myopic anisometropic amblyopia in which RLRL therapy achieved significant refractive error reduction, axial length (AL) shortening, and full visual acuity recovery – highlighting its potential as a noninvasive, dual-action intervention for a condition with limited therapeutic alternatives.

Case Report

In January 2020, a 5-year-old male child born in April 2014 was brought to the Lixiang Eye Hospital of Soochow University by his parents due to poor distance vision in the right eye. The child was brought for his first-ever ophthalmic evaluation after failing the vision-screening test performed at school entry. There was no history of premature birth, perinatal oxygen therapy, ocular trauma, or previous spectacle or patching treatment. No similar condition had been documented in family members, and the child had never undergone any prior refractive or ocular assessment. The Hirschberg test and cover test indicated no obvious strabismus. After 5 days of treatment with a 1% atropine agent to paralyze the ciliary muscle, cycloplegic retinoscopy revealed a refractive error of −7.75/−1.25×155 D in the right eye and −1.75 D in the left eye. Visual acuity testing using LogMar visual acuity charts with E labels (converted to Snellen expression) showed uncorrected visual acuity (UCVA) of 20/400 in the right eye and 20/50 in the left eye, while best-corrected visual acuity (BCVA) was 20/67 in the right eye and 20/25 in the left eye. AL measurements showed 27.79 mm in the right eye and 24.43 mm in the left eye. Anisometropic amblyopia was diagnosed as (1) inter-ocular spherical-equivalent difference ≥1.00 D, (2) BCVA in the affected eye ≥2 LogMar lines worse than the fellow eye, and (3) absence of strabismus or organic pathology. Full correction spectacles and part-time occlusion (4 hours per day) of the left (fellow) eye were prescribed, and RLRL therapy (2 times per day, 3 minutes per session) was initiated. In this case, a portable desktop device (Eyerising, Suzhou, China) was used, which incorporates a semiconductor laser diode and emits low-intensity red light at 650 nm wavelength and 1600 lux illuminance; this light traverses the pupil to reach the retina, yielding an approximate retinal power of 0.29 mW, assuming a 4-mm pupil diameter. Per our routine protocol, follow-up was scheduled every 3 months; however, several visits were delayed at the family’s request. At each visit, we recorded cycloplegic refraction, BCVA, and AL and performed funduscopy to rule out retinal adverse effects. Three months later, the child returned for follow-up. The refractive status remained unchanged, but AL of the right eye decreased to 27.53 mm, and BCVA improved to 20/50. The same treatment strategy was continued. Seven months after the initial visit, cycloplegic retinoscopy showed a refractive error of −6.50/−1.50×155 D in the right eye and −1.00/−0.50×180 D in the left eye. AL of the right eye further decreased to 27.04 mm, and BCVA improved to 20/29, while the left eye’s BCVA reached 20/20. The spectacles were adjusted according to the new refraction, and RLRL therapy was maintained. In February 2022, 25 months after the initial visit, the refractive error of the right eye was −5.00/−1.75×170 D, with AL of 26.79 mm and BCVA of 20/20. The left eye’s refractive error was −0.75/−0.50×180 D, and AL was 24.64 mm, with BCVA remaining 20/20. The spectacles were adjusted again, and the other treatment was continued without modification. In April 2023, 39 months after the initial visit, the refractive error of the right eye improved to −4.75/−1.75×175 D, AL stabilized at 26.78 mm, and BCVA remained 20/20. The left eye’s refractive error was −0.75/−0.50×10 D, AL was 24.72 mm, and BCVA was 20/20. AL was measured with the IOL-Master 500 (Carl Zeiss Meditec) under noncycloplegic conditions. Five acquisitions were averaged; only scans with signal-to-noise ratio ≥2.0 and standard deviation ≤0.03 mm were accepted. The same experienced operator performed all measurements. The flow chart showed in Figure 1.

Discussion

This case presents a notable improvement in the highly myopic eye of a patient with anisometropic amblyopia after RLRL therapy, which differs from the previous observation of rapid myopization in the fellow eye with atropine use [15]. The key findings include the significant reduction in refractive error (3.00 D) and AL (1.01 mm) in the highly myopic eye, accompanied by improved visual acuity, while the fellow eye exhibited more modest improvements.

The management of anisometropic amblyopia remains a significant challenge in ophthalmology. On one hand, the primary objective is amblyopia treatment to improve visual acuity, while on the other hand, refractive status management is essential; yet balancing these 2 goals is frequently problematic. Consequently, clinicians often focus predominantly on treating the amblyopia while overlooking refractive control in anisometropic amblyopia cases. This emphasis reflects both the widely held perception that amblyopia treatment takes precedence and the inherent difficulty of managing refractive status effectively. As a result, most research reports on anisometropic amblyopia primarily evaluate visual acuity improvement as the outcome measure [2,4,6,16]. A major reason for the difficulty in refractive management for patients with anisometropic amblyopia lies in the limitations of current interventions. Standard care typically involves spectacle correction; however, significant anisometropia often induces aniseikonia, leading to spectacle intolerance or non-adherence to therapy [17]. Crucially, spectacles themselves do not reduce the underlying anisometropia.

Presently, contact lenses or refractive surgery can reduce anisometropia, but both approaches carry unavoidable drawbacks. Contact lenses, particularly rigid gas permeable lenses, are employed in children with anisometropia, high refractive errors, or following congenital cataract surgery to better minimize aniseikonia and prevent or improve amblyopia [18]. However, some children may exhibit contact lens intolerance, manifested as ocular surface discomfort, conjunctivitis, or even corneal complications [19]. On the other hand, refractive surgery has also been considered a treatment option for high anisometropic amblyopia, encompassing both corneal refractive procedures – including photorefractive keratectomy, laser-assisted subepithelial keratectomy, laser in-situ keratomileusis, and small incision lenticule extraction – and intraocular refractive surgery. However, evidence for amblyopia improvement remained inconclusive, and refractive regression was more common among myopic eyes, younger patients, and with longer follow-up durations [20]. Additionally, adverse events such as corneal haze [21–24], striae [25], free flaps [26], radial tears at the incision site [27], and lenticule decentration [27] were reported. Additionally, several reports have documented the implantation of posterior chamber intraocular lenses for managing anisometropia, noting positive outcomes [28,29]. However, posterior chamber intraocular lens implantation carries inherent disadvantages, including risks of acute angle-closure glaucoma [30], cataract formation [31], lens decentration or rotation [32], and elevated intraocular pressure [33]. Moreover, given the incomplete ocular development in children, intraocular refractive surgery merits particularly careful consideration in this population Overall, refractive surgery is generally reserved by scholars as a strategy to be considered only when other treatment modalities have proven ineffective [34].

Conversely, during the management of high myopic anisometropic amblyopia, the fellow eye exhibits a tendency toward rapid myopic progression, while the refractive status in the amblyopic eye remains relatively stable, as consistently observed in our prior reports [15] and confirmed by Shih et al [1] and Caputo et al [35] within this population. Although this process reduces anisometropia, the accelerated myopization in the fellow eye represents an undesirable treatment outcome. Clear clinical evidence identifies AL >26 mm or myopic refractive error <−6.00 D as significant risk factors for sight-threatening complications associated with pathological myopia [36,37]. Furthermore, each 1-mm increase in AL confers an odds ratio (OR) of 2.94 (95% CI, 2.19–3.95) for developing myopic maculopathy, confirming AL as an independent and significant risk factor [38]. Meta-analyses also demonstrate that greater myopic refractive error correlates with a markedly increased relative risk of myopic macular degeneration (relative risk reaching 102.11 in high myopia), with risk escalating alongside increasing myopia magnitude [39, 40]. Additionally, patients experiencing rapid myopic progression face a substantially elevated risk of pathological myopic macular degeneration (OR 6.92; 95% CI, 1.07–44.60), leading to visual acuity deterioration [41]. Consequently, the rapid myopization of the fellow eye, occurring as a trade-off during therapy, is clearly an unsatisfactory consequence. This case report appears to represent an optimal outcome, characterized by a reduction in the degree of anisometropia alongside decreased myopia (or axial shortening), thereby resolving the question raised in our previous report: “vision improvement or myopia controlment” [15].

In China, red light was approved long ago by the National Medical Products Administration (equivalent to the US Food and Drug Administration) as a long-term, repeat-course therapy for amblyopia [42]. In 2021, a randomized controlled trial by Jiang et al [42] revealed a significant myopia control effect of RLRL therapy. Subsequently, multiple high-quality RCTs have validated the efficacy of RLRL therapy for controlling myopia progression [10,11,13,43,44]. In 2025, the World Society of Paediatric Ophthalmology and Strabismus issued a Myopia Consensus Statement (https://wspos.org/myopia/#) that dedicated substantial content to reviewing the role of RLRL therapy in myopia control.

Furthermore, myopia has traditionally been considered an irreversible condition characterized by progressive axial elongation [45,46]. However, in the study by Xu et al [10], involving 92 patients with high myopia undergoing 12 months of RLRL therapy, not only was high myopia progression significantly inhibited, but the mean AL change was −0.06 mm (95% CI, −0.10 to −0.02). Notably, 50.3% of patients exhibited AL shortening >0.05 mm, accompanied by a mild regression in spherical equivalent refraction of +0.11 D. This represents the first report of AL shortening in high myopia through intervention. Previously, He et al [42,47] had documented AL shortening in children with myopia receiving RLRL therapy. Among children with spherical equivalent refraction ranging from −1.00 D to −5.00 D treated with RLRL, the adjusted mean AL change was negative at both the 1-month and 3-month intervals. Interestingly, although the adjusted mean AL change became positive after 12 months, 21.6% of the RLRL group still demonstrated AL shortening exceeding 0.05 mm. The authors of these studies, however, could not definitively explain these phenomena. A key consideration is that the most relevant metric for myopia development is the distance from the cornea to the sclera, reflecting scleral ectasia. The IOL Master measures the distance from the cornea to the internal limiting membrane, and the reported small magnitudes of AL shortening could potentially be confounded by changes in choroidal thickness. Therefore, they may not accurately reflect true changes in the posterior segment axial dimension (cornea to sclera). In the present case, however, AL shortening exceeding 1 mm in the right eye far exceeds typical measurement error ranges. This substantial reduction strongly suggests that a process of scleral “anti-remodeling” occurred during RLRL therapy, whereby the sclera regained mechanical integrity and contracted, ultimately leading to AL shortening. This constitutes a significant clinical implication of the present case.

The mechanism by which RLRL therapy controls myopia progression is not fully understood. Building on in vivo evidence, Gawne et al [48] proposed a novel optical model wherein an imbalance between the statistical images formed by short- and long-wavelength light guides ocular growth until statistical equilibrium is achieved. For instance, under broadband (“white”) illumination, the eye is guided toward emmetropization, focusing at the midpoint of the visible spectrum [49]. During RLRL exposure, excessive activation of long-wavelength-sensitive cones occurs without concurrent stimulation of short-wavelength-sensitive cones, biasing ocular development toward hyperopia. This implies that light wavelength may function as an optical signal regulating axial elongation [50]. Furthermore, Gao et al [51] demonstrated that under hypoxic microenvironments simulating myopia progression, red-light irradiation inhibits extracellular matrix deposition via modulation of specific molecular pathways, such as the TGF-β/Smad signaling pathway. Consequently, low-level red light may slow myopia progression by suppressing fibrosis in human retinal pigment epithelial cells. Alternatively, it has been proposed that red light could enhance cellular metabolism and energy supply (eg, mitochondrial function), promoting repair in the retina and choroid, while improving local microcirculation and redox status, thereby mitigating myopia advancement [52]. A separate hypothesis suggests that red-light irradiation potentially inhibits axial elongation by increasing choroidal thickness and improving choroidal hemodynamics [53,54]. The present report offers a completely new perspective for managing myopic anisometropic amblyopia: RLRL not only treats the amblyopia but also induces axial shortening in the highly myopic eye, thereby delivering maximal benefit to these patients.

Although no retinal or choroidal adverse effects were detected on clinical examination during 36 months of follow-up, the safety data for greater than 1-year continuous RLRL exposure in children remains limited. The device delivers 0.29 mW retinal irradiance – 3 orders of magnitude below the ISO 15004-2 photochemical threshold for red light, and no photothermal risk is predicted with the 1600-lux divergent beam. Nevertheless, chronic photobiomodulation could theoretically alter retinal mitochondrial turnover, choroidal melanocyte function, or scleral fibroblast proliferation in ways that may not manifest until adolescence or adulthood. Systematic surveillance for subtle retinal dysfunction, choroidal thickness changes, and late-onset scleral thinning is therefore warranted in any future controlled trials that replicate our protocol.

We acknowledge several limitations. First, this is a single case without a control arm; therefore, causality between RLRL and AL shortening cannot be definitively established. Second, total follow-up after therapy cessation was only 3 years; whether the refractive and axial gains will persist, regress, or progress remains unknown. Third, choroidal thickness was not captured with optical coherence tomography, precluding quantification of choriocapillaris thickening that might explain the 1.01-mm AL reduction. Fourth, device-specific parameters (650 nm, 0.29 mW retinal irradiance) limit generalizability to other RLRL platforms. Fifth, the observed response may be limited to young, highly compliant children with unilateral high myopia and absence of posterior-segment pathology, and our patient’s good compliance may not reflect real-world adherence. Multicenter randomized trials with longer observation and standardized imaging protocols are therefore essential before RLRL can be routinely recommended for high myopic anisometropic amblyopia. Sixth, although a 1-mm AL reduction in 3 years is rarely seen and seems to contradict the expected trajectory of continued elongation in high myopia, it is the observation we documented and the core rationale for this report. The underlying mechanism – whether true scleral remodeling, pronounced choroidal thickening, or a combination – remains speculative and warrants dedicated investigation. Seventh, apart from potential scleral remodeling, we cannot exclude delayed emmetropization, asymmetric growth patterns inherent to anisometropic amblyopia, regression to the mean, or systematic measurement artifacts as alternative contributors to the observed AL reduction. Finally, the initial 0.17 LogMar gain at 3 months may partly reflect optical correction; however, the continued refractive and axial-length changes after the first 6 months are unlikely explained by spectacles alone.

Conclusions

In this case of high myopic anisometropic amblyopia, RLRL therapy induced unprecedented bilateral improvements: the amblyopic eye exhibited significant refractive error reduction (−3.00 D), AL shortening (−1.01 mm), and BCVA normalization (20/20), while the fellow eye exhibited a mild degree of myopic regression. This case demonstrates that adjunctive RLRL therapy can simultaneously restore normal visual acuity and reduce high anisometropia in children with high myopic anisometropic amblyopia. The marked AL shortening observed here suggests that scleral “anti-remodeling” may be achievable even in eyes with baseline AL of 26 mm or higher. RLRL may warrant further investigation as an adjunctive strategy for similar patients, provided that AL, refraction, and BCVA are monitored every 3 months. Shared decision-making should highlight the noninvasive nature of RLRL and the possibility of reducing amblyogenic refractive error and future pathologic-myopia risk. Larger, controlled studies are warranted before formal guideline revisions are made; however, this report already supports RLRL as a pragmatic, low-risk adjunct to standard spectacle correction in this challenging subset of amblyopia. Key unanswered questions include the mechanism of axial shortening, durability after cessation, optimal dosing, and interaction with conventional amblyopia therapies.