Response of the human choroid to short-term changes in eyelid and periocular temperature

ABSTRACT

Clinical relevance

Choroidal thickness measurement is gaining popularity in clinical practice and research as an early indicator of myopia progression. Understanding the influence of temperature on choroidal thickness changes will improve the reliability of the measures.

Background

It has been suggested that environmental temperature may affect choroidal thickness and blood flow, with potential implications for ocular disease and refractive development. This study investigates the effect of changes in eyelid/ocular adnexa temperature on choroidal thickness.

Methods

In a paired-eye study, 20 young, healthy subjects received a warm stimulus (heat pack) over one closed eye and simultaneously a cold stimulus (ice pack) over the other for 10 min. Eyelid temperatures were monitored with thermal probes, and optical coherence tomography scans of the retina and choroid were taken before and after heating and cooling, and then every 5 min during a 15-min recovery period. Retinal and choroidal thicknesses were measured across the macular region (6 mm), including the subfoveal (1 mm), parafoveal (1–3 mm), and perifoveal (3–5 mm) regions, and compared between the cooled and warmed eyes.

Results

When the thermal stimuli were applied, eyelid surface temperatures changed predictably and remained significantly different (by approximately 10–15°C) between the eyes after 2 min (p < .001). Relative to the warmed eye, macular choroidal thickness in the cooled eye increased significantly after 10 min of treatment (p = .004). This choroidal thickening response occurred in the subfoveal, parafoveal, and perifoveal regions (all p < .05). Upon removal of the thermal stimuli, choroidal thickness rapidly returned to the baseline and was no longer different between the cooled and warmed eye (p = .641).

Conclusion

Cooling the anterior eye by application of a cold stimulus directly onto the closed eyelid caused a small but significant increase in choroidal thickness relative to warming the anterior eye, demonstrating that the choroid can modulate its thickness rapidly and transiently in response to local temperature changes.

KEYWORDS:

• Blood flow
• choroid
• myopia
• optical coherence tomography
• temperature
• thickness

Introduction

In humans, the outer retina, including the retinal pigment epithelium, is avascular and receives its blood supply entirely from the choroid. Choroidal blood flow accounts for approximately 85% of ocular blood flow¹ and far exceeds the metabolic requirements of the retina,² as the oxygen content of venous blood remains at approximately 95% of arterial blood.³ This exceedingly high blood flow suggests that the choroidal circulation may have additional roles other than basic metabolic exchange.

Maintaining a stable retinal temperature is one proposed physiological function of seemingly excessive choroidal blood flow.⁴ Such a thermoregulatory action might be physiologically necessary since the retina and choroid continuously absorb a large proportion (~25–33%)⁵ of the light energy incident on the cornea. Choroidal circulation could regulate temperature by acting both as a heat sink and a heat source.^6,7 Evidence for the choroid’s thermoregulatory role comes from studies on experimentally induced short-term temperature changes on the surface of the human eyes. In one study, a warming lamp held over closed eyelids for 10 min caused the ocular temperature to rise significantly but choroidal blood flow, as measured with a laser Doppler flowmeter, decreased rapidly, presumably to maintain a stable retinal temperature.⁸

Despite the potential role of the choroid in retinal thermoregulation, the underlying mechanism remains largely unknown. The choroid is a dynamic structure with the capacity to change its thickness in response to various factors,⁹ including exercise,¹⁰ accommodation,¹¹ and retinal image defocus.¹² In humans, exposing one eye to a light source increases choroidal blood flow in the contralateral eye with a corresponding increase in the contralateral tissue temperature,¹³ which suggests a centrally mediated response to temperature change.

However, a local thermoregulatory mechanism may also be involved, as choroidal thickness varies according to the heat flux density on the surface of the cornea,¹⁴ and thickness and blood flow are positively correlated in both health,^15–17 and diseases (e.g., diabetic retinopathy, high myopia).^18–20 To the extent that changes in choroidal thickness and blood flow mediate the thermoregulatory function of the choroid, this could have important implications in ischaemic retinal disease, as reduced blood flow could compromise the ability of the choroid to regulate retinal temperature.

However, little is known about choroidal thickness responses to locally induced changes in ocular surface temperature. While obtaining direct measures of choroidal blood flow is challenging,²¹ choroidal thickness can be readily measured using optical coherence tomography (OCT).

This study tested the hypothesis that changing the surface temperature of the anterior eye and adnexa alters choroidal thickness. A paired-eye, within-subject study design was used to stress test the ocular thermoregulatory system by simultaneously cooling one eye, warming the other, and comparing the choroidal thickness response between the cooled and warmed eye.

Methods

Study design

This prospective paired-eye study included 20 healthy adult subjects (14 females, age ± SD: 23 ± 3 years, spherical equivalent refraction (SER): Right eye: −1.93 ± 1.97D; Left eye: −1.99 ± 1.93D). Subjects were recruited primarily from staff and students of the Faculty of Medical and Health Sciences at the University of Auckland through advertisements. The study adhered to the tenets of the Declaration of Helsinki and received ethics approval from the University of Auckland Human Participants Ethics Committee (ref: 019994). All subjects provided informed consent in writing, and their data were de-identified.

The primary outcome was the relative change in choroidal thickness between the eyes receiving the two treatments: warm vs cold. Since temperature changes were induced via conduction (i.e., thermal stimuli touching the closed eyelid), potential mechanical effects were controlled by having both eyes receive the same mechanical stimulus but different thermal stimulation. However, as a result, there was no untreated thermal control eye. Consequently, relative effects between warmed and cooled eyes are reported. The secondary outcomes were comparisons of changes in eyelid surface temperature, forehead surface temperature, retinal thickness (RT), and intraocular pressure (IOP).

Experimental protocol

Before enrolment in the study, all subjects underwent an initial eye examination to establish eligibility. This included testing of visual acuity and measurement of refractive error with an open-field auto-refractor (NVision-K 5001, Shin-Nippon, Japan; http://www.shin-nippon.jp/) with subjects fixating a target at 6 m. Subjects were excluded if they had a best-corrected visual acuity greater than 0.00 LogMAR, high refractive error (SER of ±6.00D or greater), anisometropia greater than 1.00D, any history of ocular or cardiovascular diseases and medications, ocular surgery or trauma.

Before baseline measurements were made, subjects viewed a video movie for 20 min at 6 m with their habitual optical correction while seated and as still as possible, to minimise the influence of previous tasks on choroidal thickness. In addition, all subjects were asked to avoid cigarette smoking and caffeine intake for 6 h before starting the experiment. The room temperature was between 19.8°C and 20.7°C, and the ambient lighting was 10 lux. All experimental sessions were conducted between 1 and 5 PM to control for diurnal variation in choroidal thickness across subjects.

Following the 20-min stabilisation period the video was paused, and baseline measurements of eyelid surface temperatures and IOPs were recorded, and two OCT scans were taken of each eye. Each eye was then subjected to either warm or cold treatment (decided by a coin toss) for 10 min, using a bespoke heating/cooling device. The device consisted of an elastic headband with pockets sewn on, which rested on the surface of the closed eyelids, with holes so that temperature probes could be attached.

A heat pack (a pocket hand warmer, 96 g, https://www.kathmandu.co.nz/) was inserted into one pocket and an ice pack (~120 g) into the other. During the 10-min treatment period, eyelid surface temperatures were recorded every 2 min. After the 10-min treatment period, the temperature sensors, headband, and heat/ice packs were removed, and OCT images and then IOP measurements (TA01i, https://www.icaretonometer.com) were taken of both eyes.

Subjects then watched the video for further 15 min (the recovery period, with habitual correction as for the stabilisation period), during which both the eyelid surface temperature and OCT images were taken in both eyes every 5 min.

Temperature measurements

Eyelid surface temperatures were measured using a digital thermometer (QM1601, DIGITECH, China) with an operating range of −50 to 1300°C and a resolution of 0.1°C, via two external K-type thermocouple probes. Thermocouples were placed in contact with the upper eyelids of the closed eyes, during the application of both heat and cold stimuli giving simultaneous measures of temperature from the two eyes.

Optical coherence tomography

Cross-sectional chorioretinal images of both eyes were obtained using Topcon Deep Range Imaging Swept Source OCT (DRI SS-OCT Triton, Topcon Corp., Tokyo, Japan; http://www.topcon.co.jp/) in horizontal line scan mode (6 mm length, 128 averaged samples per line, 1024 intervals) centred on the fovea (Figure 1) while the subjects viewed the internal fixation light. The SS-OCT has a scan acquisition rate of 100,000 A-scans per second with a lateral resolution of 20 microns and depth resolution of 8 microns and uses a sweeping laser (tuneable range of approximately 100 nm) as a light source centred at a wavelength of 1050 nm.

Figure 1. Examples of (A) an OCT fundus image overlaid with a six mm horizontal line scan in a temporal (T) to nasal (N) direction, (B) a B-scan image from the OCT after averaging 128 individual scans from the same location, (C) a B-scan image after correction for tilt and layer segmentation. The software automatically segmented the internal limiting membrane (red), retinal pigment epithelium – choroid interface (green), and choroidoscleral interface (blue) for choroidal thickness (CT) and retinal thickness (RT) measurements.

For each subject, after the first scan, subsequent scans were performed under follow-up mode, in which the system automatically aligns the scan with the previous scan position using the infrared retinal image. In rare cases, where any of the 128 scan lines were missed (e.g., due to blink), or when the image quality score was poor (IQ score < 60), the scan was immediately repeated.

All B-scans, in their original resolution, were exported for further analysis using OCT image segmentation software developed at the Queensland University of Technology (OCT-Tool, ver. 4.0.4).²² All the scans (n = 240) were de-identified and shuffled using a custom-written script in MATLAB (2017b; The MathWorks, USA) before horizontally aligning and processing through the OCT segmentation tool (Figure 1). The software automatically placed a foveal reference line, but this line could be manually adjusted to ensure it passed through the centre of the foveal pit.

Each segmentation line (n = 128 per scan) was examined to ensure correct segmentation, and manually corrected with the ‘Correct Layer’ tool included in the software if required. Following segmentation, the image identifiers were restored, and the segmented data was imported into MATLAB. The macular choroidal thickness and retinal thickness (RT) were calculated as the mean of thickness measures along the 6 mm lateral width of the scan (Figure 1C).

To investigate whether the effect was focal or distributed throughout the choroid, regional choroidal thicknesses were calculated as the mean of the thickness measures in the subfoveal (1 mm), parafoveal (1–3 mm) and perifoveal (3–5 mm) regions along the line. No attempt was made to correct images for magnification effects because the primary parameter of interest was within-subject effects.

There was excellent repeatability of choroidal thickness between the two baseline measurements in both the cooled eye (e.g., macular choroidal thickness: r = .987, p < .001, mean [±95% CI] difference [second minus first] = −2.98 [−7.58, 1.61] µm) and the warmed eye (r = .998, p < .001, mean difference = 1.54 [−3.60, 0.51] µm).

Statistical analysis

Statistical analyses were carried out using the Statistical Package for the Social Sciences (SPSS 24.0, SPSS Inc. Chicago, USA) and MATLAB (2017b; The MathWorks, USA). Separate analyses were conducted for the treatment and the recovery phase of the experiments. Some changes in eyelid surface temperature were not normally distributed and were compared using the non-parametric Friedman Test, with post-hoc pairwise comparisons using the Wilcoxon signed-rank test.

Normally distributed data are reported as mean (95% CI), while non-normally distributed data as median and interquartile ranges (25th percentile, 75th percentile). Changes (post-minus pre-treatment) in thickness and IOP following the treatment period were compared between the cooled and the warmed eye using two-sided paired t-tests. Similarly, changes in thickness during the recovery period were analysed with a two-way repeated-measures ANOVA using the type of eye (warmed vs cooled) and time points (pre-treatment baseline, 5, 10, and 15 min of the recovery period) as factors.

Post hoc multiple comparisons used Holm–Sidak corrections where appropriate. Correlation analysis was performed using Pearson’s R. Results were considered statistically significant at p < .05.

Results

Eyelid surface temperature

The surface temperature of the eyelid increased rapidly with the application of the warm stimulus (χ² (5) = 71.56, p < 0.001) and decreased rapidly with the application of the cold stimulus (χ² (5) = 94.90, p < .001) during the treatment period (Figure 2a). There was an increase in surface temperature in the warmed eye after just 2 min of treatment (difference = +5.25°C [4.65, 6.65], Z = −3.92, p < .001). Likewise, the surface temperature in the cooled eye decreased significantly after just 2 min of treatment (difference = −5.30°C [−8.45, −4.25], Z = −3.92, p < .001). Temperature changes continued throughout the 10-min treatment period, reaching a maximum increase of 7.25°C in the warmed eye and −8.60°C in the cooled eye.

Figure 2. Time course of the changes in (a) eyelid surface temperature (b) choroidal thickness (CT) and (c) macular retinal thickness with a warm stimulus (orange) and a cold stimulus (blue). The yellow-shaded region indicates 10 min of the treatment period; the grey-shaded region indicates 15 min of the recovery period following the removal of the stimuli. *Indicates significantly different changes in choroidal thickness between the eyes. Data are presented as the median in (a) and mean in (b) and (c), and the error bars represent the interquartile range in (a) and ±1 standard errors of the mean in (b) and (c).

Following the removal of the warm stimulus, the surface temperature rapidly returned to baseline temperatures after just 5 min of recovery (difference from baseline = 0°C [−0.28, 0.60], Z = −1.09, p = .276), and remained stable throughout the recovery period (10 min: difference = 0.30 [−0.40, 0.58], Z = −1.11, p = .267; 15 min: difference = −0.20 [−0.45, 0.23], Z = −1.02, p = .306). The eyelid surface temperature of the eye with the cold stimulus also rapidly returned to the baseline after just 5 min of recovery (difference from baseline = 1.1°C [−1.60, −0.90], Z = −3.41, p = .001); however, there was a small but significant change from the baseline that persisted throughout the recovery period (10 min: difference = −0.80 [−1.10, −0.63], Z = −2.88, p = .004; 15 min: difference = −0.80 [−1.10, −0.45], Z = −3.18, p = .001).

During the treatment period, there was a significant difference in temperature changes between the cooled eye and the warmed eye at every time point (all p ≤ .001). During the recovery period, small but statistically significant differences in eyelid temperature between the eyes remained after 5 min (Z = −3.33, p = .001) and 10 min (Z = −3.24, p = .001) but not after 15 min of recovery (Z = −1.95, p = .051).

Choroidal thickness

Choroidal thickness in the cooled eye increased significantly after 10 min of treatment (difference = +3.78 µm [+1.07 to +6.89], t₁₉ = 2.42, p = .026) but not in the warmed eye (difference = −1.65 µm [−3.63 to 0.49], t₁₉ = −1.56, p = .135). After 10 min of thermal stimulation, there was a significant difference in the change in choroidal thickness between the eyes, with the change in choroidal thickness in the cooled eye higher compared to that in the warmed eye (difference = +5.43 µm [+1.93 to +8.92], t₁₉ = 3.25, p = .004) (Figure 2b). Orthogonal regression analyses revealed that these changes in choroidal thickness were independent of baseline choroidal thickness in both the cooled eye (r = .03, p = .442) and the warmed eye (r = .001, p = .890).

After the removal of the thermal stimuli, choroidal thickness rapidly returned to baseline levels, and there was no longer any significant difference between the eyes (F_{1, 19} = 0.23, p = .641) across the recovery time (F_3,57 = 1.14, p = .335). There was also no interaction between the eye and recovery time on changes in choroidal thickness during the recovery period (F_3,57 = 0.69, p = .560), suggesting similar rates of return to thermoregulatory homoeostasis.

Changes in choroidal thickness in the subfoveal, parafoveal, and perifoveal regions between the cooled and the warmed eye followed a similar pattern (Figure 3). The changes in choroidal thickness were significantly different between the eyes (F_{1, 57} = 23.2, p < .0001), and relative to the warmed eye, the choroid in the cooled eye was significantly thicker in the subfoveal (difference = +5.32 µm [1.21 to 9.43], p = .029), parafoveal (difference = +7.03 µm [1.65 to 12.41], p = .0046) and perifoveal (difference = +5.26 µm [1.69 to 8.83], p = .029) regions. This suggests that the observed effect was not localised but distributed throughout the choroid, consistent with overall changes in blood flow.

Figure 3. Changes in the thickness of the choroid (A) and retina (B) in the subfoveal, parafoveal, and perifoveal regions after application of the thermal stimuli (cooled eye, blue, warmed eye, orange) for 10 min. Data are presented as mean ± 1 standard error of the mean. *p < .05; **p < .005.

Retinal thickness

Compared to the baseline, the macular RT remained relatively stable, with no significant change even after 10 min of both the heat (difference = 0.38 µm [−0.65 to 1.42], t₁₉ = 0.77, p = .450) and the cold treatment (difference = 0.00 µm [−0.638 to 0.639], t₁₉ = 0.001, p = .999). Furthermore, there was no significant difference in the change of RT between the cooled and the warmed eye (difference = −0.38 µm [−1.71 to 0.95], t₁₉ = −0.60, p = .555, Figure 2c). Following the removal of the thermal stimuli, there were no changes in RT between the eyes (F_1,19 = 0.06, p = .803) or across the recovery time (F_3,57 = 1.50, p = .224). When stratified into regions (subfoveal, parafoveal, and perifoveal), again there were no significant differences between the cooled and the warmed eye (F_{1, 57} = 1.96, p = .17), nor across different regions of the retina (F_{2, 57} = 0.52, p = .60).

Intraocular pressure (IOP)

Following the application of the heat and the cold stimuli for 10 min, there was a significant decrease in IOP in both the cooled (difference = −2.35 mmHg [−3.43 to −1.27], t₁₉ = −4.57, p < .001) and the warmed eye (difference = −1.40 mmHg [−2.49 to −0.31], t₁₉ = −2.69, p = .014). However, the mean change in IOP between the two eyes was not different (difference = −0.95 mmHg [−2.34 to 0.44], t₁₉ = −1.43, p = .168). The changes in IOP did not correlate with changes in macular choroidal thickness in either the cooled eye (r = .15, p = .528) or the warmed eye (r = −.03, p = .902).

Discussion

Using high-resolution OCT imaging, this study has shown the effect and time course of recovery of choroidal thickness upon cooling and warming of the anterior eye. Cooling the anterior eye caused a rapid increase in choroidal thickness, resulting in a relative difference in choroidal thickness change between the cooled and the warmed eye after 10 min. The choroidal response to local ocular cooling was rapid and transient, returning to the baseline within 5 min after the removal of the stimulus. This finding suggests that the choroid can rapidly modulate its thickness in response to changes in locally induced surface temperature. Interestingly, warming the anterior eye, in contrast, did not affect the choroidal thickness, at least in the short term, and the choroidal thickness remained relatively stable during both the treatment and recovery period. This may suggest a greater capacity for the normal choroidal blood flow to remove excess heat by acting as a heat sink than to add additional heat to the ocular tissues.

The present findings are consistent with previous speculations that anterior eye temperature changes could influence the thickness of the choroid,^7,23 and perhaps expected as the choroid is thought to act as a temperature modulator to stabilise retinal temperature.²⁴ In monkeys, changes in choroidal blood flow (through changes in IOP) produce a reduction of retinochoroidal temperature under low ambient illumination and an increase in retinochoroidal temperature under high ambient illumination.⁶ The choroidal action of retinal thermoregulation cannot be attributed to autoregulation of choroidal blood flow.²⁵

The relative increase in choroidal thickness with the cold stimulus could be due to choroidal blood flow compensating for the decreasing retinal temperature through a local mechanism. In cats with experimentally cooled conjunctiva, the retinal temperature remained relatively constant, presumably due to the temperature stabilising effect of choroidal blood flow, but the compensatory response was no longer present when vasoconstrictors were used.²⁶ It could also be argued that the opposite effect should then occur with anterior eye warming. However, no changes were observed in choroidal thickness with warming, contrary to that noted by others.⁸ The neural mechanism by which the choroid senses changes in temperature is uncertain, and different mechanisms could underlie choroidal responses to increases and decreases in temperature. Moreover, anterior components of the eye could be better at dissipating excess heat, and so prevent conduction to the posterior retina.²⁴

Another possibility is that the application of a cold stimulus to the periorbital region could have induced a diving response, due to stimulation of the ophthalmic division of the trigeminal nerve,²⁷ as used in the cold-face test for assessing trigeminal-brainstem-vagal reflex integrity.^28–30 If this was the case, the observed increment in choroidal thickness could be due to the increased supply of blood to the central nervous system (including the eyes) while limiting the supply to peripheral organs. A centrally mediated response is plausible since light exposure in humans has been shown to change tissue temperature and choroidal blood flow in the contralateral eye.¹³ However, no changes were observed in choroidal thickness with the warmed eye despite applying a thermal stress of similar magnitude. Since the choroid showed relative thickening in response to the cold stimulus, a systemic diving response induced by the application of cold stimuli could have negated the choroidal thinning response of the contralateral warmed eye. Nevertheless, the relative thickening of the choroid in the cooled eye compared to the warmed eye likely suggests a local mechanism underlying choroidal thickness modulation in response to changes in temperature.

The magnitude of choroidal thickness changes in this study was small but similar to the experimentally induced changes that have been observed in several previous studies.^11,31–35 It is possible that changes in choroidal thickness were confined to sub-levels of the choroid, particularly the vascular luminal area, instead of the stromal area, resulting in only subtle changes in overall choroidal thickness. This is further supported by the present observation that the thickness changes were independent of the baseline choroidal thickness, and previous findings that changes in the luminal area were primarily responsible for diurnal changes in choroidal thickness.³⁶

Changes in choroidal thickness between the warmed and the cooled eye were consistent from the centre to the periphery, increasing with the cold treatment and decreasing with the warm treatment. This suggests that the choroidal responses to temperature changes are not localised but rather distributed throughout the choroid, and are presumably occurring via a global mechanism, such as changes in blood flow. This study is, however, unable to rule out a local mechanism since the thermal stimuli likely affected the entire anterior eye. It remains to be seen whether locally applied thermal stimuli would trigger localised choroidal responses to changes in temperature.

Interestingly, a reduction in IOP was observed in both the warmed and the cooled eyes. Alterations in IOP could cause changes in choroidal thickness, as previous reports have shown both a negative³⁷ and a positive correlation³⁸ between IOP and choroidal thickness changes. However, changes in IOP would not explain the selective increase in choroidal thickness with the cold stimulus, but not with the warm stimulus, as IOP was reduced in both cases. Rather, the removal of mechanical stimuli likely resulted in decreased IOP in both the cooled and the warmed eye. A reduction in IOP after ocular cooling has been previously noted, but unlike the present results, the study found no change in IOP after ocular warming.³⁹

This study lacks responses from control subjects in the absence of any thermal stimuli, which limits the absolute quantification of effects. Previous studies by the authors and other groups have found that the choroidal thickness remains relatively stable in control groups over similar periods,^40–43 although the observed change in choroidal thickness in response to thermal stimuli in this study appeared smaller than that exhibited by the control group in a previous study.¹² Nonetheless, the relative change in choroidal thickness observed between two contrasting thermal stimuli demonstrates the dynamic potential of the choroid to respond to temperature changes.

In this study, thermal stimuli were applied to the anterior eye, and measurement of retinal temperature was not possible. Therefore, the transference of temperature changes from the anterior eye to the retina is unknown. However, the anterior segment is also known to regulate the temperature of the posterior eye (although to a lesser extent).²⁴ Changes in surface temperature on the anterior eye observed in this study presumably influenced the temperature at the back of the eye. As the eyelids of participants were closed during temperature application, this would have prevented changes in tear evaporation, which also regulates eye temperature.⁴⁴ Although changes in tissue temperature may have influenced the reliability of OCT measures, no changes were observed in OCT measures of RT, suggesting that this effect, if present, was minimal and the observed changes were not artefacts induced by errors to the time-of-flight assumptions made by the OCT machine.

Another limitation is that the magnitude of the observed effect (average choroidal thickness change between the cooled and the warmed eye) was smaller than the axial resolution of the OCT instrument. Whether application of thermal stimuli for a longer duration produces a stronger choroidal response remains unknown. A higher resolution instrument, such as a low-coherence interferometer, may allow more precise quantification of axial measures as a surrogate for choroidal response to thermal stimuli.

Lastly, the lack of correction for lateral magnification of the OCT images might have affected inter-eye comparisons of parafoveal and peripheral choroidal thickness measurements. However, this is unlikely to influence the results since subjects with greater than 1.00D anisometropia were excluded, and there was a strong correlation in SER between the cooled and warmed eyes (r = 0.91, p < 0.0001).

In conclusion, changes in ocular surface temperature can produce rapid, local, and transient changes in choroidal thickness that are asymmetrical around heating and cooling. The potential clinical impact of these choroidal responses induced by relatively large magnitude changes in temperature remains unresolved. Nevertheless, consideration of changes in ambient or ocular surface temperature seems necessary when conducting studies on choroidal thickness.

Acknowledgements

The authors thank Joy Qin from the University of Auckland for assistance with data collection and Professor Scott Read from the Queensland University of Technology for providing the OCT segmentation software.

Disclosure statement

Safal Khanal: None

Philip RK Turnbull: None

Lucia Kim: None

John R. Phillips: Dr. Phillips is an inventor on patents related to contact lenses for myopia control

Additional information

Funding

This work was supported by the University of Auckland Summer Research Scholarship (LK), New Zealand Association of Optometrists Higher Degree Research Write-up Scholarship (SK), Reality Labs Research (SK), Meta (SK), American Academy of Optometry Foundation (SK), and E. Matilda Ziegler Foundation for the Blind Inc. (SK).