The Ocumetra IMC 2024 showcase: myopia metrics, centiles and iris colour in atropine

The 2024 International Myopia Conference (IMC) brought together world-renowned experts in optometry and ophthalmology to share the latest breakthroughs and innovations in myopia research, to advance global understanding of myopia management. Ocumetra and the Centre for Eye Research Ireland (CERI) presented eight abstracts: here, we summarise this new research, which looks at metrics for monitoring myopia progression, axial length centiles, trends in vision impairment, ocular parameter changes in myopia progression, and the effects of iris colour on atropine efficacy. 

Pictured in the cover image is Professor Ian Flitcroft, Founder and Chief Innovation Officer of Ocumetra.  

Authors: Corrina McElduff,¹ Kate Loskutova,¹ James Loughman,^1,2 Gareth Lingham^1-3 and Ian Flitcroft^1,2,4

1. Ocumetra Ltd., Dublin, Ireland
2. Centre for Eye Research Ireland, Technological University Dublin, Dublin, Ireland
3. Centre for Ophthalmology and Visual Sciences (incorporating Lions Eye Institute), University of Western Australia, Perth, Australia
4. Children’s University Hospital, Dublin, Ireland

Centile-based metrics compares an individual's measurement (such as refractive error or axial length in myopia management) to a reference population, ranking it in terms of percentiles (or centiles). It provides a way to see how an individual's myopia progression compares to a typical range for their age, ethnicity, and other factors. In clinical practice, this helps tailor treatment by identifying how a patient’s progression aligns with or deviates from expected norms. This study discusses the shortcomings of current methods for monitoring myopia treatments and introduces new centile-based metrics—CURB (Cumulative Refractive Benefit), CALM (Cumulative Axial Length Modification), and the Treatment Efficacy Index (TEI)—designed specifically for use in clinical practice. Unlike existing metrics, which are primarily aimed at clinical trials, these new metrics are individualized, taking into account key factors like baseline refraction, age, and ethnicity. They enable direct comparisons between spherical equivalent refraction (SER) and axial length, and the TEI metric provides a personalized percentage-based measure that projects future progression. Data from the MOSAIC trial confirmed a strong correlation between SER and axial length efficacy, making these metrics more practical for personalized myopia management in clinical settings.

Purpose: To review the limitations of current metrics for monitoring myopia treatments, and provide new centile-based metrics for refractive and axial length changes optimised for application in clinical practice, not just clinical trials. Ideally such metrics should:

1. Be evidence based
2. Applicable to individual patients, not just clinical trials
3. Be applicable to both refractive error and axial length
4. Allow direct comparison between spherical equivalent refraction (SER) and axial length
5. Take into account the factors known to influence untreated myopic progression (i.e. baseline refraction or axial length, age and race) in an individual patient
6. Allow meaningful age-specific projections

Results: Currently used methods include absolute or percentage difference in refraction or axial length of a treatment group with a control group, Cumulative Absolute Reduction in axial Elongation (CARE, typically applied to clinical trial data), comparison of refractive or axial change against centile charts, and comparison with emmetropic growth rates (for axial length). For an individual patient in clinical practice (i.e. a clinical trial where n=1, with no control group), these approaches cannot provide adequate guidance as they fail to address factors individual patient factors known to influence myopia progression (such as baseline refraction or axial length, age and race). They are also not designed to provide future projections. CURB (CUmulative Refractive Benefit) and CALM (Cumulative Axial Length Modification) are derivatives of our previously described (2020) metric of Treatment Efficacy Index (TEI = 100*(1- ([(biometric change over time in patient)-( biometric change expected in an age and sex matched emmetropic eyes over time)]/[(biometric change over time period in untreated age in subjects of identical baseline characteristics )-( biometric change expected in an age and sex matched emmetropic eyes)]). Like TEI, CURB and CALM are based on an ethnicity, sex and baseline ser/axial length adjusted centile prediction model. Both are expressed in absolute terms as (biometric change over time in patient -  biometric change over time period in untreated age in subjects of identical baseline characteristics) to provide a better assessment of clinical relevance.

The TEI is expressed as a percentage, but by including all relevant individual progression factors, provides a personalised medicine approach and more accurate projections. SER and axial TEI values were calculated for all treated subjects in the MOSAIC trial for whom complete data was available at 12 and 24 months (n= 136). The TEI values for SER and axial length showed significant correlation (0.59 Spearman correlation, p < 10-8), with a linear relationship and a slope lose to unity (slope = 0.88; 95% CI 0.629, 1.131), validating the direct comparability of the SER and axial efficacy metrics in both responders and non-responders.

Conclusions: Existing metrics for monitoring myopia progression and treatment are mostly designed to describe clinical trials, or fail to address individual patient factors known to influence myopia progression (such as baseline refraction, age and race). CURB, CALM and TEI provide a novel, validated method of benchmarking an individual’s response to a myopia intervention for both SER and axial length.

Authors: Michael Moore,^1,2 Siofra Harrington,^1,2 Ian Flitcroft,^1,3,4 James Loughman^1,3

1. Centre for Eye Research Ireland, Technological University Dublin, Dublin, Ireland
2. School of Physics and Optometric and Clinical Sciences, Technological University Dublin, Dublin, Ireland
3. Ocumetra Ltd, Dublin, Ireland
4. Children’s University Hospital, Dublin, Ireland

Visual impairment has significant consequences on individual quality of life and imposes economic burdens due to healthcare costs and loss of productivity on a global scale.¹ This study investigated the link between increasing refractive error, and the development of visual impairment in individuals over 40, using data from 40 optometry practices in Ireland. Visual impairment was defined as logMAR 0.3 (equating to 6/12 or 20/40) or worse. The analysis found that myopic patients experienced visual impairment more quickly than those with emmetropic or hyperopic refractive errors. Among those aged 70-79, 10% of myopes developed visual impairment within 6.04 years, compared to 7.47 years for hyperopes and 9.01 years for emmetropes. Key factors influencing worse visual acuity included older age, male gender, higher refractive error, and increased myopic progression. The findings highlight that myopes face a higher and faster risk of visual impairment with aging: early detection and proactive myopia management in clinical practice are essential to reduce visual impairment and its associated burdens.

Purpose: The combination of increasing refractive error, particularly myopic refractive error, and age has been associated with higher rates of vision impairment. A limited number of studies have assessed this relationship, which is critical in understanding the potential public health implications of increasing myopia prevalence and the benefits of myopia control treatments.

Methods: Anonymized electronic medical record (EMR) data was sourced from 40 optometry practices in Ireland. Analysis was to those records that had complete refraction, visual acuity and demographic data and to those aged over 40. Visual acuity values were converted to LogMAR where they had not been originally recorded in LogMAR notation. A Kaplan-Meier survival analysis was performed to determine the time until hyperopic (≥ +0.50 D), emmetropic (> -0.50 D and < +0.50 D) and myopic (≤ -0.50) patients would become visually impaired (LogMAR < 0.3). Vision Impairment was assessed using linear mixed models with LogMAR visual acuity as the outcome, age, sex, year of exam, SER (Spherical Equivalent Refraction), final SER and total SER change as fixed effect covariates and random intercept terms for subject.

Results: In total 188,943 records contained complete data of which there were 235,555 visits of 112,542 unique patients aged over 40. The dataset was 58% female and had a mean age of 61.87 ± 12.57 years. There were 42,405 visits from myopes (17.7%), 72,245 visits from emmetropes (30.2%) and 124,615 visits from hyperopes (52.1%). Myopes were found to become visually impaired in a shorter time than either emmetropes or hyperopes. The time for 10% of those aged 70 – 79 to become visually impaired was 6.04 (95% CI: 5.05, 8.84) years for myopes, 7.47 (95% CI: 6.72, 8.18) years for hyperopes and 9.01 (95% CI: 6.56, 10.99) years for emmetropes. Worse visual acuity was predicted by older age (estimate = 0.002, p < 0.001), male sex (estimate = 0.004, p < 0.001), earlier year of exam (estimate = -0.001, p < 0.001), higher absolute SER (estimate = 0.006, p < 0.001), greater myopic change in SER (estimate = -0.009, p < 0.001) and more myopic final SER (estimate = -0.0004, p < 0.001).

Conclusions: Increasing age and increasing refractive error were associated with worse visual acuity and higher rates of visual impairment. Myopes had the worst outcomes in terms of visual impairment and progressed to visual impairment at the fastest rate.

Authors: Gareth Lingham,^1-3 James Loughman,^1,2 David A Mackey,³ Samantha Sze-Yee Lee,³ Emmanuel Kobia-Acquah,² Ian Flitcroft^1,2,4

1. Ocumetra Ltd, Dublin, Ireland
2. Centre for Eye Research Ireland, Technological University Dublin, Dublin, Ireland
3. Centre for Ophthalmology and Visual Sciences (incorporating Lions Eye Institute), University of Western Australia, Perth, Australia
4. Children’s University Hospital, Dublin, Ireland

While placebo-controlled trials remain important in clinical research, there is a growing trend towards using virtual control groups instead of placebos due to ethical concerns, particularly when proven effective treatments exist for the condition being studied.^2,3 This study evaluates the use of a centile-based method to create a virtual control group for assessing myopia control interventions. Data were drawn from three randomized controlled trials on atropine 0.01% for myopia progression (WA-ATOM, MOSAIC, and MTS1) and the proprietary axial length reference database from Ocumetra (Ireland) was used to generate age- and sex-matched baseline axial length centile projections for untreated axial length progression. Two virtual control models were tested—one basic and one adjusted for ethnicity. Among 590 participants (32.9% assigned to placebo), the median 24-month axial length change was 0.35mm. The basic model predicted a median change of 0.31mm, while the ethnicity-extended model predicted 0.34mm. Projections slightly underestimated growth in males and those with low baseline axial lengths. These findings show that reference axial length centiles can effectively predict axial length growth in untreated myopes, and that accounting for ethnicity improved accuracy; this can be useful in clinical practice when illustrating untreated myopia to patients.

Purpose: With acceptance of myopia control interventions into standard of care for myopia progression, placebo control groups in randomised controlled trials are being phased out. We aimed to evaluate the performance of a centile-based method for creating a matched, virtual control group in the evaluation of myopia control interventions.

Methods: Data for the analysis were taken from 3 randomised controlled trials of atropine 0.01% vs placebo eye drops for the treatment of myopia progression: the Western Australian Atropine Treatment of Myopia (WA-ATOM; Australia) study, the Myopia Outcome Study of Atropine in Children (MOSAIC; Ireland) and the Myopia Treatment Study (MTS1; USA). Using a proprietary axial length reference database (Ocumetra, Ireland), participant’s age- and sex-matched axial length centile at baseline was determined and projections for future, untreated axial length progression made by assuming participants maintain their baseline centile value. We tested this method to create two virtual control groups, one without (basic model) and one with additional adjustment to myopia progression rates based on ethnicity (ethnicity-extended model). Mean and standard deviation or median and interquartile range were used to described parametric and non-parametric variables, respectively. Confidence intervals were calculated using bootstrapping techniques, which are agnostic to the sample distribution.

Results: Data were available for 590 participants, of whom 194 (32.9%) were assigned to placebo eye drops. At baseline MTS1 control participants were slightly younger (mean age 10.3 years) and less myopic (median spherical equivalent refractive error [SER] -2.81 dioptres [D], mean axial length 24.44mm than the WA-ATOM (age 12.2 years, SER -3.50 D, axial length 24.90mm) and MOSAIC (age 11.8 years, SER -3.25 D, axial length 24.93mm) participants; however, sex was approximately evenly distributed 63%, 58% and 58% in MOSAIC, MTS1 and WA-ATOM, respectively. In the pooled sample, the distribution of actual 24-month change in axial length was positively skewed with a median of 0.35mm (interquartile range [IQR]: 0.17, 0.57). Using the basic model, projected median axial length was 0.31mm (95% confidence interval [CI]: 0.30 to 0.34mm), while projected median axial length using the ethnicity-extended model was 0.34mm (95% CI: 0.30 to 0.37mm). The errors of axial length growth predictions at 24-month were approximately normally distributed with mean (standard deviation) of +0.01mm (0.28). On simple linear regression, centile-based projections were more likely to underestimate 24-month axial length growth for males, relative to females, (estimate=0.09mm, p=0.03) and for participants with a low axial length at baseline (beta=-0.09, p>0.001).

Conclusion: Reference axial length centile databases can be used to predict 24-month axial length growth in untreated myopic children and adolescents. Adjusting for participant ethnicity improved the prediction. Adding systematic corrections for sex and baseline axial length or centile may further improve projections.

Authors:  Ernest Kyei Nkansah,¹ Gareth Lingham,^1-3 James Loughman,^1,2 Ian Flitcroft^1,2,4

1. Centre for Eye Research Ireland, Technological University Dublin, Dublin, Ireland
2. Ocumetra Ltd, Dublin, Ireland
3. Centre for Ophthalmology and Visual Sciences (incorporating Lions Eye Institute), University of Western Australia, Perth, Australia
4. Children’s University Hospital, Dublin, Ireland

This study used the refractive mechanism map (RMM) to classify myopia and analyzed two-year changes in ocular parameters in myopic children treated with atropine 0.01% versus placebo. The RMM is a model used to classify myopia based on the contributions of various ocular components, such as axial length, corneal power, and internal refractive power (primarily from the lens), to the overall refractive error of the eye. The study included 250 European children, aged 6-16, who participated in the MOSAIC clinical trial. Axial length, corneal curvature, anterior chamber depth, and lens thickness were measured over 24 months. Differences in changes to ocular structures between the atropine group and placebo group were statistically insignificant. At baseline, most participants were classified as axial myopes. Atropine 0.01% significantly slowed axial length growth in axial myopic children, with no significant impact on other ocular parameters such as anterior chamber depth, corneal curvature, or lens thickness. In contrast, there was no significant effect of atropine in corneal myopes. These results highlight atropine’s specific benefit in slowing axial length elongation in children with axial myopia. It also underscores the importance of classifying myopia based on ocular structure to ensure that treatments are targeted and effective for each type of myopia.

Purpose: Classify myopia using the refractive mechanism map (RMM) model and investigate two-year longitudinal changes in ocular parameters (axial length, corneal curvature, anterior chamber depth, lens thickness) in atropine-treated versus placebo-treated myopic children. 

Methods: The RMM calculates the contributions of each of the axial length, corneal power, and internal refractive power (predominantly lens) to the measured refractive error. The study included 250 European myopic children aged 6–16 years, who participated in The Myopia Outcome Study of Atropine in Children (MOSAIC) clinical trial from March 2019 to September 2023. Participants were randomly assigned 2:1 to use atropine 0.01% or placebo eye drops nightly for two years. Axial length, lens thickness, anterior chamber depth, and corneal curvature were measured using a Topcon Aladdin biometer and refractive error using cycloplegic autorefraction. Spherical equivalent refraction (SER) was measured by autorefraction 30 minutes after instillation of 1% cyclopentolate. Changes in ocular parameters were assessed at 12, 18, and 24 months. Statistical analysis included descriptive analysis and linear mixed models with a 3-way interaction between visit, treatment, and dominant myopia component, adjusting for age and sex. 

Results: At baseline, treatment groups were well balanced, and the mean (SD) age of participants was 11.8 (2.4) years and 62% were female. There was no statistical difference in ocular parameters [axial length (AL), anterior chamber depth (ACD), central corneal thickness (CCT), corneal curvature (CR), lens thickness (LT)] between treatment groups (p>0.05). Over the 2 years, in the atropine group AL and ACD increased by 0.33 ± 0.26mm and 0.04 ± 0.13mm, respectively, and LT decreased by 0.011 ± 0.14mm. In the placebo group, AL and ACD increased by 0.40 ±0.32mm and 0.03±0.11mm, respectively, and LT decreased by 0.006 ± 0.13mm (p>0.05). The mean corneal curvature and thickness did not change significantly in each treatment group. 

Using the RMM, at baseline 213 (85.2%) participants were axial myopes, 33 (13.2%) were corneal myopes, 3 (1.2%) had myopia from IRP and 1 (0.4%) had missing data. 204 (82%) attended the 24month visit (175 axial myopes, 28 corneal myopes, 1 myopia from IRP). In the axial group, atropine significantly slowed axial length elongation (mean effect=-0.07mm, p=0.026), but no significant effect on ACD (mean effect=0.025mm, p=0.145), LT (mean effect=-0.022mm, p=0.272), CCT (mean effect=-0.0008mm, p=0.431) and CR (mean effect=-0.0059mm, p=0.315). Atropine had no significant effect in the cornea group (AL: mean effect=-0.04mm, p=0.596; ACD: mean effect=-0.06mm, p=0.091; CCT: mean effect=0.0004mm, p=0.902; LT: mean effect=0.028mm, p=0.732; CR: mean effect=0.0008mm, p=0.936). 

Conclusions: Atropine 0.01% eye drops significantly slowed axial length elongation in axial myopic children while other ocular parameters showed no significant changes.

Authors:  Megan Doyle,^1,2 Veronica O’Dwyer,^1,2 James Loughman,^1,3 Emmanuel Kobia-Acquah,¹ Eoin Kerrin,¹ Síofra Harrington^1,2

1. Centre for Eye Research Ireland, Technological University Dublin, Dublin, Ireland
2. School of Physics and Optometric and Clinical Sciences, Technological University Dublin, Dublin, Ireland
3. Ocumetra Ltd, Dublin, Ireland

Choroidal thickness is of great interest in myopia management because it is hypothesized to play a role in myopia progression and control.^4,5 Changes in choroidal thickness can potentially serve as an early indicator of the effectiveness of myopia control interventions.³ This study examined the differences in choroidal thickness (CT) and axial length (AL) between myopic and non-myopic children aged 6-7 and explored the relationships between CT, spherical equivalent refraction (SER), and ocular biometry. The analysis included 149 children, with an average SER of -2.50 D for the myopes, and an average SER of +1.43 D for non-myopes. AL was longer in myopic children (mean 24.07 mm) compared to non-myopic children (mean 22.65 mm). They found that myopic participants had thinner choroids across all regions, with the largest difference in the inner-inferior region and the smallest in the outer-nasal region. AL was negatively correlated with SER and CT, meaning that as axial length increased, choroidal thickness decreased. For every 1mm increase in AL, subfoveal CT decreased by 32.07 µm, and for every 1D increase in SER, subfoveal CT increased by 12.75 µm. These findings suggest that choroidal thinning provides valuable insight into the structural changes associated with myopia.  

Purpose: Choroidal thickness (CT) and axial length (AL) are indicators of myopia progression and development. This study investigated differences in CT and AL between myopic and non-myopic 6–7–year–olds and explored the relationships between CT, cycloplegic spherical equivalent refraction (SER), and ocular biometry.

Methods: Baseline right eye macular scans (DRI-OCT Triton Plus, Topcon), SER (Shin-Nippon Auto Refkeratometer), AL (Aladdin Biometer, Topcon) were analysed for 149 children (mean (standard deviation (SD)) 7.25 (0.74) years; 53% male, 83% white). Cycloplegic myopia was defined as SER≤-0.50D (n = 48), and non-myopia as SER>-0.50D (n = 101). The Pearson correlation coefficient (r) evaluated the relationship between CT, SER, AL, and multivariate linear regression and quantified the effect (beta–coefficient) while controlling for confounders. Independent t-testing investigated differences in CT between myopes and non-myopes.

Results: The mean (SD) SER was -2.50 (1.38) D in myopic, and 1.43 (1.07) D in non-myopic participants. The mean (SD) AL was 24.07 (0.91) mm in myopes and 22.65 (0.73) mm in non-myopes. As expected, AL was negatively correlated with SER (r = -0.759, p<0.001). CT was negatively correlated with AL and positively correlated with SER in all quadrants (p<0.001). Mean (SD) CT was thinner in myopes across all quadrants (p<0.001); for example, subfoveal CT was thinner in myopes (252.47 (54.99) µm) than in non-myopes (293.05 (66.92) µm. The differences ranged from a minimum difference outer–nasally 28.86 (8.46)µm to a maximum difference inner–inferiorly 44.70 (9.95)µm. As CT was thinner in boys (β=-25.87, 95% Confidence Interval (CI): -5.15 to 46.58, p=0.015) and decreased with age (β=-16.83, 95%CI -2.72 to 30.95, p=0.02), both age and sex were controlled for in all analyses. 

For every 1mm increase in AL, CT decreased subfoveally by 32.07µm (β=-32.07, CI: -42.11 to -22.04), inner–inferiorly by -30.81(CI: -40.82 to -20.79) µm, and outer–nasally by 24.37 (CI:-32.06 to -16.67) p<0001 for all. For every 1D increase in SER, CT increased subfoveally by 12.75 (CI: 7.92 to 17.59)µm, inner–inferiorly by 11.46 (CI: 6.60 to 16.33) µm, and outer–nasally by 8.14 (CI: 4.34 to 11.35)µm, p<0.001 for all. 

Conclusions: Controlling for sex and age, myopic children had thinner choroids in all ETDRS regions than non-myopic 6–7–year–olds. The biggest difference between myopes and non-myopes was found in the inner–inferior region, and although still significant, the smallest difference was found in the outer–nasal region. Additionally, increased AL was a strong negative predictor of CT. Variations in CT related to SER and AL reflect tissue redistribution during childhood myopic axial elongation. Both AL and CT are essential for understanding myopia progression.

Authors: Ian Flitcroft,^1-3 Gareth Lingham,^1,2,4 Eoin Kerin,² Ernest Kyei Nkansah,² James Loughman^1,2

1. Ocumetra Ltd, Dublin, Ireland
2. Centre for Eye Research Ireland, Technological University Dublin, Dublin, Ireland
3. Children’s University Hospital, Dublin, Ireland
4. Centre for Ophthalmology and Visual Sciences (incorporating Lions Eye Institute), University of Western Australia, Perth, Australia

This study aimed to develop a clinical tool (a nomogram) to predict the risk of myopic visual impairment based on ChT. Although ChT is measurable in clinical practice, it is underutilized in decision-making for myopia management. Researchers analyzed 1,268 ChT measurements from three clinical trials (n = 1,268) and used risk factors like age, gender, AL, and SER to develop a predictive model. The cohort had a mean age of 13.9 years, with 63% female participants. Mean ChT was 238.63 microns, and results showed that ChT decreased by 21.16 microns for every 1 mm increase in AL and by 3.85 microns for every 1D increase in SER. Conversely, ChT increased by 2.55 microns per year of age and by 10.96 microns for males. Notably, 5% of participants had ChT measurements 100 microns thinner than expected, putting them at higher risk for myopic eye diseases such as myopic maculopathy. This information provided the basis for the clinical nomogram being developed that provides individualized expected ChT values for myopic children, offering a practical tool to help clinicians more accurately predict an individual child's risk of developing myopia-related complications, improving both early detection and the effectiveness of management strategies.

Introduction: Choroidal thickness (ChT) is a known experimental biomarker for myopia and can now be measured in clinical practice. However, it is not often employed to inform clinical decision-making in myopia. The aim of this study is to develop a clinical tool to predict increased or reduced risk of myopic visual impairment in individuals whose choroidal thickness is less than expected based on their age, sex, and refraction.

Method: A total of 1,268 choroidal thickness measurements from three myopic control clinical trials conducted at the Centre for Eye Research Ireland were included in a linear mixed model regression analysis. A clinical nomogram was developed using known myopic risk factors, including age, gender, axial length (AXL), and spherical equivalent refraction (SER), to predict expected choroidal thickness measurements. SER was measured by cycloplegic autorefraction, and AXL by partial coherence interferometry (Aladdin, Topcon, Japan). Choroidal thickness (ChT) was measured by Swept Source-OCT (Triton Plus, Topcon) and Spectral Domain-OCT (Spectralis, Heidelberg).

Results: The cohort had a mean age of 13.9 ± 4.7 years (range: 6.09 to 25.8 years), with 63% of participants being female. Mean choroidal thickness was 238.63 ± 82.15 microns (range: 63.25 to 459.1 microns), while the mean SER and AXL were -3.5 ± 1.9 D (range: +0.44 to -10.75 D) and 25.99 ± 4.55 mm (range: 22.09 to 28.87 mm), respectively. Regression analysis revealed a reduction in ChT of 21.16 ± 58.33 microns for every 1 mm increase in AXL and a reduction of 3.85 ± 1.45 microns for every 1 D increase in SER. There was an increase in ChT of 2.55 ± 0.45 microns per year of age and 10.96 ± 3.93 microns associated with male gender. Further analysis yielded a residual standard deviation of 62.63 microns, generating a 95% confidence interval of ± 123 microns from the expected ChT. This analysis showed that 5% of myopes had ChT measurements approximately 100 microns thinner than expected, placing them at higher risk for developing myopic eye diseases such as myopic maculopathy. This data was used to create a clinical ChT nomogram tool, which provides an individualized expected ChT for every myopic child, offering an additional means for risk profiling in myopia management.

Conclusions: ChT is a known myopia biomarker that is underutilized in clinical practice. The resulting clinical ChT nomogram is a simple, visual tool that provides an accurate and objective method for clinicians to assess and track a myopic child’s risk of complications, such as myopic maculopathy, based on the difference between their measured and expected choroidal thickness.

Authors: Eoin Kerin,¹ Gareth Lingham,^1-3 Ian Flitcroft,^1,2,4 Ernest Kyei Nkansah,¹ Samantha Sze-Yee Lee,³ David A Mackey,³ James Loughman^1,2

1. Centre for Eye Research Ireland, Technological University Dublin, Dublin, Ireland
2. Ocumetra Ltd, Dublinm Ireland
3. Centre for Ophthalmology and Visual Sciences (incorporating Lions Eye Institute), University of Western Australia, Perth, Australia
4. Children’s University Hospital, Dublin, Ireland

This study explores the impact of low-concentration atropine (0.01% and 0.05%) on pupil diameter and accommodation in relation to iris colour, using data from four clinical trials. Participants (total n = 646) were assigned to nightly atropine or placebo. The analysis showed no significant differences in near accommodation between iris colour groups for atropine 0.01% after 12 or 24 months, and similar changes in pupil diameter across trials, except for a slight difference in the 0.05% atropine group. However, atropine 0.01% was more effective in controlling myopia progression in participants with lighter irises compared to those with brown irises. These results suggest that while low-concentration atropine is generally well-tolerated across different iris colours, higher concentrations may be more effective for those with darker irises to achieve optimal myopia control. In clinical practice, it is therefore important to consider iris colour among the factors that may influence the therapeutic effectiveness of atropine treatment.

Purpose: Low-concentration atropine is frequently used to treat myopia progression in children and adolescents. However, interaction between atropine and iris colour remains under-explored. This study investigates the impact of 0.01% and 0.05% atropine on pupil diameter and accommodative outcomes, while considering iris colour, utilising data from Myopia Outcome Study of Atropine in Children (MOSAIC; n=250, 11.8±2.37 years) and Treatment Optimisation of Atropine Study (TOAST; n=56, 21.4±1.88 years) in Ireland, Western Australian Atropine Treatment of Myopia (WA-ATOM; n=153, 11.51±2.69 years) study, and Pediatric Eye Disease Investigators Group's (PEDIG) Myopia Treatment Study (MTS1; n=187, 10.00±1.78 years). 

Methods: Participants were assigned to nightly atropine 0.01%, atropine 0.05%, or placebo. Pupillometry was conducted in MOSAIC (Topcon Aladdin), WA-ATOM (Neuroptics NPi-200), and TOAST (Diagnosys Colourdome), although WA-ATOM only assessed pupils under scotopic conditions. Accommodative amplitude was measured in MOSAIC, WA-ATOM, and MTS1 using near point rule. Iris colour was categorised as brown or not brown. Analysis employed linear mixed models with random intercept terms for eye nested within participant. Treatment effects were modelled using treatment by visit interaction, and differences across eye colour groups were assessed with three-way interaction among eye colour, treatment, and visit. Interactions were analysed with pairwise comparisons (p<0.05). Analyses were conducted in R (version 4.3.2). 

Results: 424 participants were assigned to atropine 0.01%, 194 to placebo, 28 to atropine 0.05% and an additional 66 and 60 MOSAIC participants reassigned to atropine 0.05% and placebo, respectively, on cross-over. In pooled MOSAIC, WA-ATOM and MTS1 0.01% groups, there were no significant differences in near point of accommodation between iris colour groups at 12months (p=0.14) and 24months (p=0.86). In the MOSAIC, WA-ATOM and TOAST trials, separately, iris colour groups showed similar changes in pupil outcomes, with only significant treatment by colour difference in MOSAIC 0.05% group after 12 months treatment (not brown vs brown, difference=+0.51mm, p=0.048). In the pooled MOSAIC, WA-ATOM and MTS1 0.01% vs placebo, estimated treatment effect of atropine 0.01% on SER and axial length varied significantly with both eye colours (p=0.004 and p=0.004, respectively). At 24 months, treatment effect of atropine 0.01% was greater in not brown iris group (SER: difference=+0.16D, p=0.003; AL: difference=-0.09mm, p<0.001), compared to brown group (SER: difference=-0.07D, p=0.24; AL: difference=+0.02mm, p=0.31). 

Conclusions: Low-concentration atropine (0.01% and 0.05%) generally exhibits similar effects on pupil diameter and accommodation across iris colours. Atropine 0.01% had greater treatment effect among participants with lighter irises, compared to brown. These findings support tolerability of low-concentration atropine in diverse populations, but higher concentration atropine may be needed among participants with brown iris colour to achieve adequate efficacy

Authors: Siofra Harrington,^1,2 Michael Moore,^1,2 James Loughman,^1,3 Ian Flitcroft,^1,3,4 Veronica O'Dwyer^1,2

1. Centre for Eye Research Ireland, Technological University Dublin, Dublin, Ireland
2. School of Physics and Optometric and Clinical Sciences, Technological University Dublin, Dublin, Ireland
3. Ocumetra Ltd, Dublin, Ireland
4. Children’s University Hospital, Dublin, Ireland

This study aimed to identify effective non-cycloplegic screening methods for pre-myopia in 6-7-year-olds, which increases the risk of developing myopia by age 10. Involving 728 children from 24 primary schools in Ireland, the study assessed visual acuity, spherical equivalent (SE), axial length (AL), and corneal curvature (CR) to determine screening accuracy. Myopia was defined as a cycloplegic SE of ≤ -0.50D, and pre-myopia was defined as > -0.50D but ≤ +0.75D. Results showed that uncorrected distance visual acuity (UDVA) alone had limited accuracy in detecting pre-myopia, but combining non-cycloplegic SE with ocular biometry (AL/CR) improved screening performance. The best method for myopia detection was the combination of non-cycloplegic SE, AL, and UDVA, achieving high sensitivity and specificity. Clinically, this means that relying on poor UDVA alone is insufficient for screening; rather, combining multiple measures enhances the effectiveness of early pre-myopia and myopia screening.

Purpose: Pre-myopia, characterised by less hyperopic refraction than typical for 6-7-year-olds, significantly increases the risk of developing myopia by age 10. This condition is becoming more prevalent in younger Asian and Caucasian children. While myopia management technologies exist, early identification of children at risk for myopia or pre-myopia is crucial. However, there is limited research on effective screening methods for pre-myopia. This study aimed to establish effective non-cycloplegic screening methods for pre-myopia in 6-7-year-olds.

Methods: This cross-sectional study involved 24 primary schools in Ireland, with 728 participants (mean age 7.08 ± 0.46 years; 48.2% girls). Uncorrected distance visual acuity (UDVA) was assessed using a logMAR chart. Cycloplegic spherical equivalent (SE) classified refractive status (myopia: SE ≤ -0.50 D; pre-myopia: SE > -0.50 D but ≤ +0.75 D). Pre-cycloplegic SE was measured with the Welch Allyn Spot Vision Screener, and post-cycloplegic SE with the Dong Yang Rekto ORK-11 Auto Ref-Keratometer. Axial length (AL) and corneal curvature radius (CR) were recorded using the Zeiss IOL Master. Logistic regression models identified effective non-cycloplegic screening methods for myopia and pre-myopia. Receiver Operator Characteristic (ROC) curves assessed sensitivity and specificity trade-offs in screening tests.

Results: Pre-myopia prevalence was 32.4% (n = 236) [95% CI: 29.3 to 36.2], and myopia prevalence was 3.7% (n = 27) [95% CI: 2.5 to 5.5]. UDVA-based screening had an area under the ROC curve (AUC) of 0.45 for detecting myopia and pre-myopia combined. Non-cycloplegic SE, AL, AL/CR, and parental myopia had AUCs of 0.69, 0.67, 0.71, and 0.57, respectively. The best screening method combined non-cycloplegic SE and AL/CR (AUC = 0.75). Including UDVA or parental myopia did not improve results.

For detecting myopia alone, the AUCs were UDVA: 0.74, non-cycloplegic SE: 0.88, AL: 0.84, AL/CR: 0.82, and parental myopia: 0.61. AL plus UDVA provided an AUC of 0.89. The best combination of AL and non-cycloplegic SE achieved an AUC of 0.94. Adding parental myopia did not improve this (AUC = 0.93), but adding UDVA achieved an AUC of 0.95.

Conclusion: While UDVA alone provides acceptable discrimination for myopia, it is insufficient for screening pre-myopia. Non-cycloplegic SE alone has relatively poor discrimination for pre-myopia, but its accuracy improves when combined with ocular biometry. The best results for myopia discrimination are achieved by combining non-cycloplegic SE, biometry, and UDVA measures.