Abstract
Purpose
To determine the role of topical caffeine in slowing progression of myopia, both as a standalone treatment and in combination with atropine.
Methods
In a prospective, randomized, dispensing clinical trial, 96 children with myopia, aged 6–13 years, spherical equivalent (SE) from –0.50 diopters (D) to –6.00 D and astigmatism less than 2.00 D were randomly assigned to nightly use of either 2 % caffeine, 0.02 % atropine with 2 % caffeine (combination) or 0.02 % atropine eye drops. An additional 86 children with myopia were enrolled in a concurrent parallel group to wear single-vision (SV) spectacles. The primary outcomes were changes in SE and axial length (AL) over a period of 12 months for each group.
Results
All groups progressed in myopia. At 12 months, the mean change in SE/AL was –0.76 ± 0.51 D / 0.37 ± 0.20 mm and –0.70 ± 0.55 D / 0.35 ± 0.23 mm with SV and 2 % caffeine, respectively. In comparison, progression was slower at –0.46 ± 0.50 D / 0.24 ± 0.19 mm and –0.47 ± 0.38 D / 0.23 ± 0.18 mm with atropine monotherapy and combination groups, respectively. Compared to the change in AL with SV, the change in AL was significantly less with 0.02 % atropine and the combination group (post hoc analysis, P = 0.024 and 0.007, respectively). Similarly, the change in SE was significantly less with 0.02 % atropine compared to the SV group ( P = 0.027).
Conclusions
Used as a standalone treatment, topical 2 % caffeine did not slow myopia progression. When combined with atropine, caffeine had no impact on the efficacy of atropine in slowing myopia.
1
Introduction
Myopia, or near-sightedness, is a common refractive error characterized by the inability to see distant objects clearly while nearby objects remain in focus. As the prevalence of myopia and its associated burden continue to rise globally, effective strategies for myopia management are becoming increasingly essential. Over the past two decades, various optical and pharmacological interventions designed to attenuate myopia progression were developed. Among the approaches, topical atropine, an antimuscarinic receptor agent, was reported to be the most effective strategy. Even though the precise mechanisms underlying its efficacy remain unclear, atropine administered at various concentrations (ranging from 0.01 % to 1 %) has been demonstrated to slow myopia in individuals across different ethnicities. However, dose-dependent adverse effects such as photophobia and blurred near vision with higher concentrations of atropine present challenges and have limited its use. In this regard, although the Low-concentration Atropine for Myopia Progression (LAMP) Study found that children in Hong Kong tolerated 0.05 % atropine well over a three-year period, in another study, more than 60 % of German children reported blurred near vision with 0.05 % atropine. The highest clinically tolerable concentration for Caucasian children appeared to be around 0.02 %. Similarly, a study conducted on Vietnamese children evaluated three concentrations of atropine (0.01 %, 0.02 % and 0.03 %) and found 0.02 % to be the optimal concentration as it did not result in a substantial change in accommodative amplitude or pupillary diameter.
7-methylxanthine (7-mx), a nonselective adenosine antagonist has been previously explored for its role in slowing myopia. Oral 7-mx was found to exert inhibitory effects on axial growth and myopia progression in Danish children, and it was also found to influence the growth of sclera. Extending from this research, caffeine (1,3,7-trimethylxanthine), a xanthine derivative, was used topically in primates and demonstrated a reduction in vitreous chamber elongation, along with an increase in choroidal thickness. Furthermore, topical caffeine was observed to be permeable into anterior chamber tissues and lens capsule. However, there has been no study so far that assessed the efficacy of topical caffeine in slowing myopia progression in children with myopia. Therefore, we conducted a clinical trial to evaluate the efficacy of topical caffeine in slowing the progression of myopia in children and to assess whether a standalone treatment with topical caffeine or a combination therapy of atropine and caffeine provided superior outcomes.
2
Methods
2.1
Study design and population
In a prospective, dispensing trial conducted at An Sinh Hospital and Hai Yen Eye Center in Ho Chi Minh City, Vietnam, myopic children were enrolled and randomly assigned to one of three groups: once a day nightly administration of 2 % caffeine, 0.02 % atropine, or a combination of 0.02 % atropine with 2 % caffeine. Additionally, children with myopia were enrolled in a concurrent, parallel group to wear single-vision (SV) spectacles. Inclusion criteria for the trial included children 6–13 years of age, had myopia with spherical equivalent (SE) ranging from –0.50 diopters (D) to –6.00 D and astigmatism less than 2.00 D, vision correctable to at least 20/25 or better in each eye with single-vision spectacles. Exclusion criteria were presence or history of ocular conditions including strabismus, cataract, glaucoma, retinopathy of prematurity or other disorders affecting refractive development, systemic conditions such as Marfan syndrome affecting refractive development, current use of systemic or topical medications that might affect ocular health, history of eye surgery, history of use of myopia control interventions in 12 weeks before visit, other contraindications (e.g. respiratory and cardiac diseases, attention deficit hyperactivity disorder) that might affect the use of atropine or caffeine, known allergic reactions to atropine, xanthine and other antimuscarinic receptor agents and current participation in another clinical trial.
2.2
Ethical approval and participants’ consent
The study protocol was approved by the Institutional Research Ethics Committee of An Sinh Hospital (No. CS/AS/18/12), the Institutional Research Ethics Committee of the Ministry of Health of Vietnam (No. 84/CN-HDDD) and the Institutional Human Research Ethics Committee of University of New South Wales (HC200725) and adhered to the Declaration of Helsinki for experimentation on human subjects. The trial was registered on ClinicalTrials.gov (NCT04301323). Before data collection, written informed consent was obtained from the parents and/or carers.
2.3
Sample size
Prior studies indicated that the annual progression of refractive error in Asian children using single-vision spectacles was –0.70 ± 0.45 D. Consequently, a minimum of 35 subjects in each treatment group was required to detect a statistically significant 50 % difference in annual myopia progression (0.35 ± 0.45 D) between test and control groups at the 5 % level of significance with 80 % power, using a 2-tailed distribution and assuming 20 % dropouts. The detectable difference of 0.35 ± 0.45 equates to an effect size of 0.78. G*power was used to compute the sample. It was estimated that a minimum of 60 participants should be enrolled in the control group, which wore single-vision spectacles. The lack of randomization to the control group could result in differences in baseline demographics and consequently, a higher combined variance of the outcome variable. Furthermore, a higher drop-out may occur in the control group. This sample size for controls would ensure that a minimum power of 80 % was maintained even if the combined standard deviation increased by 20 % (0.45–0.55 D) due to lack of randomization and accounted for a higher dropout rate (25 %).
2.4
Participant grouping
Upon enrolment, eligible children and their parents were given the option to select either the treatment or control group. If they opted for the intervention group, participants were randomized in a 1:1:1 ratio to one of three treatment regimens: 2 % caffeine, 0.02 % atropine or a combination of 0.02 % atropine with 2 % caffeine (CustomCare Compounding Pharmacy, Dural, NSW, Australia). The randomization plan was generated from http://www.randomization.com/ . Randomization plan was for a minimum sample of 105 participants and used 7 blocks with 15 subjects per block. Children were dispensed with their allocated eye drops as single-use eye drop units on a monthly basis by an assigned clinic staff and they were required to be applied on a nightly basis at bedtime in both eyes. Children and their parents/carers were instructed to store the medication in a secure, cool place and to store both the used and unused vials in the clear plastic bags provided by the clinic and to return them at their next visit. The used vials were counted and the participants considered to be compliant when 85 % or more of the vials were used. Investigators, participants and their caregivers remained blinded to the identity of the dispensed drops throughout the study.
2.5
Data collection
The treatment group was examined at baseline, 2 weeks and subsequently at 3-month intervals, whereas the control group was examined at baseline and then every 6 months. Following a baseline examination including history, visual acuity and slit-lamp evaluation, axial length (AL) measurement was conducted using Lenstar 900 (Haag-Streit, Switzerland). Thereafter, eyes were cyclopleged with three drops of 1 % cyclopentolate (Cyclogyl, Alcon-Convreur, Rijksweg, Belgium) instilled three times at five-minute intervals. After approximately 30 min, pupils were checked with a pen torch (a fully dilated pupil that was unresponsive to light was considered the end point for dilation) and refractive error of the eye was measured using open-field autorefractometer NKVision 5001 (Shin-Nippon by Rexxam, Japan). For both AL and cycloplegic autorefraction, five measurements were conducted for each eye and the average was considered. Cycloplegic autorefraction was performed at baseline, 6 months and 12 months, while AL was measured at baseline and at 3-month intervals.
2.6
Statistical analysis
Statistical analysis was conducted using the Statistical Package for the Social Sciences software, version 25.0 (IBM Corp, Armonk, NY, USA). The primary outcome of the trial was the 12-month change in SE and AL. Data from both eyes were considered. In the model, the treatment group was factored as between-subject factor, while visits and eye were factored as within-subject factors. Interactions were tested. If there was a significant treatment x visit interaction, the data were analysed for each visit. The level of significance was set at 5 %. Comparison between study groups was performed using repeated measures Analysis of Variance (ANOVA). Post hoc multiple comparisons were corrected using Bonferroni correction. A linear mixed model that accounted for both fixed (study group) and random factors was used to analyze progression in a grouped format. The model was adjusted for confounders such as age, sex, parental myopia and baseline refractive error. Model-based estimated means for each group, with 95 % confidence limits, are reported.
3
Results
A total of 200 children were enrolled in the trial with 112 children opting to participate in the intervention group and 88 children in the control group. Following an assessment of the inclusion/exclusion criteria, 96 of the 112 participants were randomized to the intervention groups and 86 participants to the control group ( Fig. 1 ).
Before completion of the baseline visit, 5 participants from the intervention group and three participants from the control group were permanently discontinued. After the successful completion of the baseline visit, there were only a few discontinuations until the 12-month period (one participant each from 2 % caffeine and 0.02 % atropine groups and two participants from the combination group). In comparison, 12 participants were discontinued from the control group. The reasons for discontinuation were lost to follow-up, time conflict and disinterest (seven, four and one participant, respectively).
3.1
Characteristics of participants
Table 1 presents the demographic data of all participants who successfully completed the baseline visits. There were no statistically significant differences between the study groups regarding age, gender or parental myopia ( P > 0.05). A significant difference was observed in SE, with post hoc analyses revealing that the control group had less myopia compared to the atropine and combination groups. However, no significant differences were noted in AL among the groups.
| Variables | n | 0.02 % Atropine | n | 0.02 % Atropine + 2 % Caffeine | n | 2 % Caffeine | n | Control | P value |
|---|---|---|---|---|---|---|---|---|---|
| Age (year), mean ± SD | 33 | 10.1 ± 2.0 | 28 | 10.6 ± 1.9 | 30 | 10.3 ± 2.1 | 83 | 10.1 ± 2.1 | 0.657 |
| Female: Male ratio (%) | 33 | 72.7: 27.3 | 28 | 46.4: 53.6 | 30 | 53.3: 46.7 | 83 | 51.8: 48.2 | 0.145 |
| Parental Myopia, 0: 1: 2 | 33 | 14: 10: 9 | 28 | 9: 15: 4 | 30 | 15: 9: 6 | 83 | 32: 40: 11 | 0.220 |
| SE (D), mean ± SD (range) | 66 | −4.06 ± 1.32 (−4.39 to −3.74) | 56 | −4.39 ± 1.40 (−4.76 to −4.01) | 60 | −3.95 ± 1.17 (−4.25 to −3.64) | 164 | −3.21 ± 1.35 (−3.42 to −3.00) | 0.000 |
| AL (mm), mean ± SD (range) | 58 | 24.85 ± 0.81 (24.64–25.07) | 54 | 25.14 ± 0.74 (24.93–25.34) | 60 | 25.09 ± 0.90 (24.86–25.33) | 166 | 24.79 ± 0.82 (24.67–24.92) | 0.143 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
