The Effects of Repeated Low-Level Red-Light Therapy on the Structure and Vasculature of the Choroid and Retina in Children with Premyopia

This was a single-center, single-masked, randomized controlled trial in which 94 children were enrolled between 12 August 2021 and 3 September 2022, at Tianjin Medical University Eye Hospital, Tianjin, China. Children were randomly assigned to the RLRL therapy group or the control group at a ratio of 1:1 using stratified block randomization with a block length of eight. Independent investigators conducted outcome measurements for masking purposes, and the statistician was also blinded to allocation. Written informed consent and verbal assent were obtained from a parent/legal guardian, and the study adhered to the principles of the Declaration of Helsinki. The institutional review board of Tianjin Medical University Eye Hospital approved the study (No. 2021KY-11). Investigators and key personnel at each site involved in the present study were trained and certified prior to study commencement. There were no changes in the protocol and methods after trial commencement. The trial was registered in the Chinese Clinical Trial Registry (http://​www.​chictr.​org.​cn/​) with the registration number ChiCTR2200062028.

A total of 94 children (94 eyes from 94 subjects) with premyopia were enrolled in the study, with the diagnosis of premyopia based on cycloplegic optometry results (− 0.50 D < spherical equivalent [SE] ≤  + 0.75 D). Children aged 7–12 years with regular follow-ups for at least 12 months were included. Exclusion criteria included astigmatism of ≥ 2 D, other ocular diseases (such as amblyopia, strabismus, congenital cataract, glaucoma, corneal scar, optic neuropathy, traumatic ocular injury, uveitis, and ocular tumor), history of any ocular surgery, and any systemic diseases or conditions that may affect visual function and development (including diabetes mellitus and chromosome anomaly).

Children in the RLRL therapy group received RLRL treatment twice daily, with each session lasting 3 min and at least 4 h between treatments, while those in the control group did not. Given that the treatment time for the two daily sessions needed to be separated by more than 4 h, one treatment session was planned to occur prior to the child going to school in the morning and the second session was planned that same afternoon after school. The RLRL intervention was provided by a desktop device (Eyerising, Suzhou Xuanjia Optoelectronics Technology, Kunshan city, China). This technology incorporates semiconductor laser diodes that emit low-intensity, single-wavelength 650 (10)-nm red-light laser beam at an illuminance level of approximately 1600 lx through the pupil to the fundus. This device was certified as a Class IIa medical device by the State Administration for Market Regulation of China. The light power of this RLRL device entering a 4-mm pupil (the maximum pupil size under the condition of bright-light exposure over 10 s) is 0.29 mW. These power levels are classified as Class 1, which is at a level considered safe for direct ocular exposure. Guardians/parents of the children received training to oversee the intervention program at home. To ensure accurate tracking and recording of treatment history, the device was integrated with an automated diary function and connected to the internet. In addition, two investigators were tasked with managing intervention compliance. To enhance treatment compliance, parents or legal guardians received weekly reminders about the intervention program in an attempt to maintain consistency and effectiveness in the intervention process.

All participants underwent a standardized examination at the baseline visit, followed by five scheduled follow-up visits at 1, 3, 6, 9, and 12 months. A cycloplegic refraction test was conducted both before enrollment and at every 6 months during the follow-up period. To achieve cycloplegia, four drops of compound tropicamide eye drops were administered at intervals of approximately 5 min. At 30 min after the administration of the final drop, cycloplegic autorefraction was assessed using the KR-8800 desktop autorefractor (Topcon Corp., Tokyo, Japan). This autorefractor obtained three readings, each separated by 0.25 D, for both spherical and cylindrical components which were then averaged. Subsequently, an experienced optometrist performed cycloplegic retinoscopy. Best-corrected visual acuity (BCVA) was measured at a distance of 5 m and converted to the logarithm of the minimum angle of resolution (logMAR) scale. Prior to initiating cycloplegia, several ocular parameters were measured using the IOL Master 700 swept-source optical coherence tomography (OCT)-based biometer (Carl Zeiss AG, Jena, Germany), including axial length (AL), anterior chamber depth (ACD), lens thickness (LT), and keratometry values (flat keratometry [Kf]). The Swept-Source Optical Coherence Tomography/Optical Coherence Tomography Angiography (SS-OCT/OCT-A) system (VG200S; SVision Imaging, Henan, China) was employed to evaluate macular structures and vascular characteristics in all eyes. To minimize the potential confounding effects of diurnal variations, all measurements were conducted between 12 a.m. and 3 p.m. The macular images were classified into three concentric circles according to the grid of the Early Treatment of Diabetic Retinopathy Study (ETDRS), centered on the foveal region (within central 1-mm diameter), the parafoveal (ParaF) region (from an inner diameter of 1 mm to an outer diameter of 3 mm), and the perifoveal (PeriF) region (between the 3 and 6 mm concentric ring). The major automated thickness outcomes recorded included ganglion cell complex (GCC) thickness (perpendicularly from the inner edge of the nerve fiber layer [NFL] to the outer edge of the inner plexiform layer [IPL]; see Electronic Supplementary Material [ESM] Fig. S2A); the inner nuclear layer (INL) thickness (perpendicularly from the inner to the outer edge of the INL; see ESM Fig. S2B); the outer retinal layer (ORL) thickness (measured from the inner edge of the outer plexiform layer [OPL] to the outer edge of the retinal pigment epithelium (RPE)/Bruch’s membrane complex; see ESM Fig. S2C); retinal thickness (RT) (measured from the inner edge of the nerve fiber layer [NFL] to the outer edge of the RPE/Bruch’s membrane complex; see ESM Fig. S2D); and choroid thickness (CT) (defined as the vertical distance between the outer edge of the retinal pigment epithelium [RPE] and the outer edge of the choroid; see ESM Fig. S2E). Retinal microvascular parameters included superficial retinal vascular density (SRVD) (ESM Fig. S3A) and deep retinal vascular density (DRVD) (ESM Fig. S3B). The SRVD was automatically defined by the system as microvessels located 5 μm above the inner limiting membrane to the one-third interface of the ganglion cell layer and inner plexiform layer (GCL + IPL). Likewise, the DRVD was defined as the microvessels located from the one-third interface of GCL + IPL to 25 μm below the lower border of the INL. For choroidal microvascular parameters, choriocapillaris perfusion area (CCPA) was measured, which involved maximum projection 20 μm deep to Bruch’s membrane (ESM Fig. S3C). Choroidal vessel volume (CVV), also known as choroidal luminal volume, was defined as the volume of the large and medium choroidal vessels (ESM Fig. S4).

The minimum sample size required for the analysis was conservatively determined to be N = 42 per cohort of participants who completed the trial. This sample size was considered sufficient to detect a small treatment effect, specifically a difference of 0.25 D in refraction, with a standard deviation (SD) of 0.35 D. The type I error rate was set at 0.05, and the statistical power was set at 80%. Balanced recruitment by age between 7 and 12 years (inclusive) was assumed for this sample size estimation. Only the right eye of each participant was analyzed in this study. During the treatment phase, both endpoints were summarized with mean value ± SD or median with interquartile range at every visit. To ensure the accuracy of the results, only SE data with complete cycloplegia were included in this analysis. A change in a parameter was determined as the difference between the baseline value and the corresponding follow-up value. Changes in AL, SE, retinal, and choroidal parameters at each follow-up time point were compared using a generalized linear mixed model (GLMM). Baseline age, gender, group, visit, group-by-visit interaction, and corresponding baseline parameters were fixed factors in the models, and subjects were included as random factors. The estimated mean differences, 95% confidence intervals, and two-sided P values were calculated.

Linear regression models were utilized to evaluate the factors associated with changes in AL and SE over a period of 12 months in the RLRL group. The analysis encompassed baseline age, gender, CT, AL/SE, and change in fovea CT over 12 months into the models. To facilitate visualization of the results, a 1-unit change in the regression models was deemed equivalent to a 10-mm change in fovea CT. A statistical significance level of 0.05 was consistently applied throughout the analysis. All statistical tests were performed with SPSS software (version 26.0; SPSS IBM, Armonk, NY, USA).

A total of 94 consecutive children met the prescribed criteria and were consequently enrolled in the study. Among these children, seven did not have a completed follow-up dataset, one had segmentation errors, and one had poor image quality. These children were excluded from the analysis, leaving 85 children (90.4%) in the final analysis (43 in RLRL group; 42 in control group) (Fig. 1). The distribution of baseline age, gender, AL, and SER was well balanced between the RLRL group and control group (P > 0.05; Table 1).

During the initiation phase of treatment, two children reported intolerance to bright light and dry eye symptoms. However, these side effects were transient and resolved either spontaneously or with minimal intervention. Over the subsequent 12 months of treatment, no participants withdrew from the study due to intolerability or discomfort. Moreover, there were no recorded side effects in terms of complaints, functional loss (BCVA), or anatomical changes (OCT/OCTA scans) during the 12-month follow-up period.

As presented in Table 2 and Fig. 2, the thickening of the choroid was observed across the entire macular region against different time points in the RLRL group. Foveal CT relative to baseline CT significantly increased at the 1-month follow-up with RLRL therapy, with a mean (± SD) adjusted change of 16.96 ± 19.87 μm. The greatest magnitude in foveal CT change was observed at the 3-month visit (an increase of 19.58  ± 20.59 μm), with a slight reduction in the extent of foveal CT increase at the 6-month visit (an increase of 15.85 ± 23.77 μm). The second greatest CT increase was observed at the 9-month visit (an increase of 19.57 ± 35.51 μm), after which the extent of CT increase gradually decreased until the end of the study at the 12-month visit (an increase of 11.99 ± 32.66 μm). We also observed a significant increase in choroidal thickness (CT) in the ParaF and PeriF areas in the RLRL group over 12 months. In contrast, CT across the entire macular region in the control group significantly decreased throughout the follow-up visits (all P  < 0.05).

Regarding the vascular parameters of the choroid, significant increases in CVV were observed primarily in the ParaF and PeriF regions of the choroid in the RLRL group. In contrast, the control group exhibited a decrease in CVV in the PeriF region over 12 months (P < 0.05). Furthermore, notable elevations in CCPA were detected in the PeriF area of the choroid in the RLRL group at the 1-month (an increase of 0.40 mm²), 3-month (an increase of 0.25 mm²), and 12-month visits (an increase of 0.42 mm²) (all P < 0.05) (Table 3).

As presented in ESM Tables S1 and S2, throughout the 12-month duration of RLRL therapy, no differences were observed between the RLRL group and the control group regarding foveal RT and retinal vascular parameters (P > 0.05). In the perifoveal region, however, a notable reduction in retinal thickness, particularly in the outer RT, was evident in the control group over 12 months (P < 0.05), while such alterations were minimal or absent in the RLRL group.

Comparison of myopia incidence rates between the RLRL group and the control group over 12 months based on cycloplegic retinoscopy in shown in ESM Fig. S1. The 6-month incidence rate of myopia was notably lower in the RLRL group than in the control group (9.3% [4/43 children] vs. 21.4% [9/42 children], respectively), resulting in a substantial 57.9% reduction in incidence. Furthermore, at the 12-month mark, the children in RLRL group maintained a lower myopia incidence rate (20.9% [9/43 children] vs. 38.1% [16/42 children], respectively). The absolute mean difference between the two groups was calculated to be 17.2%, representing a 45.1% reduction in incidence.

In the RLRL group, the mean (SD) change in SE over a period of 12 months was – 0.11 (0.44) D. In contrast, the control group exhibited a mean (SD) change of – 0.42 (0.36) D. The absolute mean difference in SE change between the two groups was found to be – 0.31 D (P < 0.001) (Fig. 3), indicating a notable 73.8% reduction in SER shift. The 12-month SE changes for each group, as well as the mean difference between the RLRL and control groups, are presented in ESM Table S3. Furthermore, the mean (SD) increase in AL over 12 months was 0.09 (0.21) mm for the RLRL group, compared to 0.28 (0.12) mm for the control group. This resulted in an absolute mean difference in AL change of 0.19 mm (P < 0.001), representing a substantial 67.9% reduction in AL change (Fig. 4). The AL change at the 1-, 3-, 6-, 9-, and 12-month visits for each group, as well as the mean difference between the two groups, are also presented in ESM Table S3. Regarding the anterior segment parameters, no significant differences were observed between the control and RLRL group in terms of anterior chamber depth, flat keratometry, and lens thickness at different time points.

Multiple regression analysis was performed to explore the factors associated with changes in AL and SE over the 12 months following RLRL therapy. The statistical analysis demonstrated that the changes in AL (standardized coefficient β = − 0.571, P < 0.001) and SE (standardized coefficient β = 0.467, P = 0.008) were significantly correlated with the thickening of the foveal choroid over 12 months but not with baseline AL/SE, baseline fovea CT, age, or gender (Table 4).