Variations in choroidal thickness between emmetropic and myopic eyes after caffeine intake

Abstract

Purpose:

To investigate the variations in choroidal thinning between emmetropic and myopic subjects after caffeine intake.

Methods:

Forty-five healthy participants (age mean ± standard deviation [SD]: 20.75 ± 1.6 years) recruited in a prospective cross-sectional study. They were divided into three groups, based on refractive error status, emmetropes (spherical equivalent [SE] -0.25 to +0.25D), myopes (SE ≥-0.50D to ˂-6.00D), and high myopes (SE ≥-6.00D). The participants underwent choroidal thickness [ChT] measurements using optical coherence tomography [OCT] (Topcon 3D OCT-1 Maestro System) at baseline, 20, 40, and 60 minutes after consuming 200 mg of caffeine. The measurements of ChT were taken from five different areas horizontally.

Results:

Compared to baseline measurement, the ChT significantly decreased after 200 mg caffeine intake in all conditions (P < 0.001). There was a statistically significant difference in ChT at baseline between the refractive groups (χ²(2)=6.548, P = 0.038) as the high myope group showed lesser horizontal ChT within Nasal at (1 mm) area compared to the myope group (P = 0.032), however, no other significant differences in mean scores of horizontal ChT at baseline were found. The thinning in ChT in the high myope group was lesser compared to both emmetrope and myope groups; however, it was not statistically significant (P > 0.05).

Conclusion:

The results of this study suggest that the reduction in ChT due to caffeine consumption is not significantly affected by the refractive status of the eye. Thus, it is safe to hypothesize that the vascular part of the choroid behaves identically to vasoconstrictive in both emmetropic and myopic eyes. The choroidal thinning found in myopic and high myopic eyes and/or the reduced choroidal vascularity index were not apparent in this study. Therefore, further longitudinal studies recruiting greater numbers of participants, including myopes and high myopes, and measuring both vascular and stromal layers to investigate such variations are warranted.

Myopia and high myopia are expected to affect 5 billion and 1 billion of the world population by 2050, respectively.[1] This significant increase will lead to a greater prevalence of vision-threatening conditions, including retinal detachment,[2] glaucoma[3] and myopic macular degeneration.[4]

Myopic eyes are structurally different from emmetropic and hypermetropic ones in many features, including axial length and anterior chamber depth,[5] choroidal thickness [ChT][6] and choroidal vascularity index [CVI].[7,8,9,10]

The choroid forms over 90% of the ocular blood flow and covers the outer two-thirds of the eye, and investigating the structure, thickness and functional variations of the choroids has been carried out extensively.[7,8,9] One of the extensively studied areas is the role of choroid in emmetropization. The choroid has shown significant influence on axial elongation.[11] Besides thickening and thinning of the choroid causing the retina to move towards the plane of focus, the choroid also influences ocular elongation as a result of defocus.[11] Additionally, it has been found that the choroid’s lateral extent might have a tangible impact on the sclera’s size, and therefore on the size of the globe.[12] Thus, the choroid, specifically ChT, shows an important role in the process of the eye developmental growth.

ChT is significantly affected by axial length, with an increase of 1 mm in axial length leading to a 58.2 µm reduction in ChT.[13] Among eyes with high myopia, axial length plays the most significant role in determining ChT.[6] It has been suggested that the axial length-dependent reduction in ChT is due to the progressive degenerative disease of myopia in combination with the loss of choroidal tissue in the nasal area due to peripapillary atrophy.[14] Additionally, myopia had a more significant impact on ChT than on retinal thickness.[15] Children with myopia had significantly thinner subfoveal ChT than emmetropes.[16] A highly myopic eye’s choroidal thinning can be attributed to the development of posterior staphyloma.[15] In contrast, children with hyperopia had significantly thicker subfoveal ChT than emmetropes.[17] Additionally, myopic eyes, particularly the high myopic, showed a significantly reduced volume of vascular area in relation to choroidal volume leading to a lower CVI compared to emmetropes.[7,8,9,18]

Other factors could affect the ChT besides refractive error and axial length such as age,[19] natural diurnal variations,[20] regional distribution[21] and diseases.[22]

Moreover, ChT has been found to decrease after caffeine intake; Caffeine causes a significant reduction in both subfoveal and extrafoveal ChT.[23,24] The effect of caffeine consumption starts 5 minutes after drinking and lasts for at least 6 hours.[24] These reductions in ChT were attributed to the significant increase of vascular resistance of the choroidal vessels after caffeine, which yielded to a decrease in choroidal blood flow.[23]

Although the effect of caffeine intake upon ChT has been addressed in the abovementioned studies, it has not been, to our knowledge, assessed in myopic eyes. The aim of this study is to investigate variations in choroidal thickness across five different areas in emmetropic and myopic subjects following caffeine intake.

Methods

Subjects

Forty-five visually healthy subjects (24 male and 21 female) participated in this prospective cross-sectional study. There were 13 emmetropic subjects, 27 myopic, and 5 high myopic, ranging in age from 19 to 28 years (mean ± SD, 20.76 ± 1.58 years). Ethical approval was obtained from the Institutional Review Board at King Saud University, and the study was conducted according to the tenets of the Declaration of Helsinki. Informed consent was obtained from each subject prior to beginning the study. Exclusion criteria were any ocular or systemic diseases, including previous history of ocular surgery, history of trauma, ocular or systemic medication, binocular vision abnormality, pregnancy, smoking, and astigmatism greater than 1.50D. Subjects with a range of SE refractive errors from −6.25D to +0.25D were recruited to the study.

Ocular parameters measurements

Refractive error was measured using standard subjective refraction methods, ophthalmic examinations were carried out by slit-lamp (Haag-Streit BQ 900 Slit-Lamp), axial length was determined by averaging three measurements using noncontact partial coherence interferometry (IOLMaster^®500 from ZEISS) and ChT measurements were obtained using noncontact spectral-domain optical coherence tomography (Topcon 3D OCT-1 Maestro System).

Study procedures

Subjects were asked not to consume any caffeine-containing product for at least 12 hours prior to experiment. The participants then were divided into three groups, based on refractive error status, emmetropes (SE -0.25 to +0.25D), myopes (SE ≥-0.50D to ˂-6.00D), and high myopes (SE ≥-6.00D).

Two blinded examiners, examiner 1 (S.K) and examiner 2 (T.M) measured the ChT at five areas, at baseline (before caffeine intake) and after caffeine intake. The caffeine was given as a cup of coffee (made from double espresso capsules from Nespresso) that contains 200 mg. The measurements were taken between 8:00 AM and 11:00 AM to avoid diurnal variations of choroidal thickness.

All reported data were for the right eye and for horizontal ChT. The ChT measurements were measured at baseline and repeated three times in 20-minutes intervals (at 20, 40 and 60 minutes) after drinking caffeine. Subjects were asked not to consume any caffeine-containing products in between measurements. ChT was obtained from five different areas, which are: subfoveal [SF], nasal parafoveal at 1 mm [N1], nasal perifoveal at 2 mm [N2], temporal parafoveal at 1 mm [T1], and temporal perifoveal at 2 mm [T2] [Fig. 1]. Scans with the best visualization of choroidal boundaries were chosen for analysis.

Figure 1.

ChT were measured at five different areas: the subfoveal [SF], the nasal parafoveal at 1 mm [N1], the nasal perifoveal at 2 mm [N2], the temporal parafoveal at 1 mm [T1], and the temporal perifoveal at 2 mm [T2]

Statistical analysis

Sample size calculations were conducted by using the mean difference and standard deviation of the differences from a previous study,[23] which required a sample size of 1 to achieve a power of 80% and a level of significance of 5%, for detecting a mean of the differences of 2.4 between pairs, assuming the standard deviation of the differences to be 0.8.

The data were analyzed offline using SPSS IBM software. Repeated measures ANOVA was carried out to evaluate the effect of caffeine on horizontal CT within the right eye at 4 different times (baseline, 20, 40, and 60 minutes) among the five areas (SF, N1, N2, T1, T2) between the three groups (Emmetrope, Myopes, and High myopes). A Bonferroni correction was applied for multiple comparisons when necessary. P values were considered statistically significant if they were less than 0.05. Linear regression analyses were performed to determine the associations between axial length and SE with ChT.

Results

Demographics

There were statistically significant differences in mean scores between the three groups in refractive error and axial length variables (x² (2) = 33.95, P < 0.001). Additionally, there were significant differences in axial length variables among the groups (F (2,42) = 12.855, P < 0.001). However, there was no statistically significant difference in mean scores between the three groups in age (P > 0.05). This information is summarized in Table 1.

Table 1.

Mean and standard deviation of demographic data at baseline

Variable		Group 1 (emmetrope SE -0.25 to +0.25D)		Group 2 (myope SE ≥-0.50D to <-6.00D)		Group 3 (high myope, SE ≥6.00D)		P
Age (Years) Range (Min., Max.) Mean±SD		(19, 28) 21.08±2.47		(19, 22) 20.56±1.12		(20, 22) 21.0±0.71		0.644a
SE of Refractive Error (D) Range (Min., Max.) Mean±SD		(-0.25, 0.25) -0.17±0.16		(-0.50, -4.00) -1.49±0.99		(-6.00, -6.25) -6.10±0.14		<0.001a
Axial length (Min., Max.) Mean±SD (mm)		(22.42, 25.56) 23.49±0.82		(22.97, 25.94) 24.35±0.77		(24.98, 25.72) 25.42±0.29		<0.001b
Total (n)		13		27		5

^aKruskal-Wallis test, ^bANOVA test

All measurements were performed horizontally for both eyes, and the paired sample t-test showed that there was no significant difference in measurements between the two eyes (P < 0.05). Therefore, only the measurements from the right eye were used for the analysis.

Reproducibility of TOPCON 3d OCT-1 Maestro system in measuring the horizontal ChT between two masked examiners

The intraclass correlation coefficients (ICC) between two examiners using an absolute agreement definition[25] showed excellent reliability of the instrument between the two examiners in measuring ChT at SF, N2, T1 and T2 as ICC values were 0.987, 0.911, 0.946, and 0.928 respectively, and good reliability at N1 area as ICC value was 0.862.

Relationship Between ChT and refractive error/axial length

The correlation between ChT at the five areas and refractive error at baseline was not statistically significant (P > 0.05) for all five areas [Fig. 2], and neither was the correlation between ChT and axial length at the same five areas at baseline (P > 0.05) [Fig. 3].

Figure 2.

The correlation between ChT and refractive error [SE] for all five areas at baseline. No significant correlation was found between ChT at any of the measured five areas and refractive error (P > 0.05). (a) subfoveal [SF], (b) nasal parafoveal at 1 mm [N1], (c) nasal perifoveal at 2 mm [N2], (d) temporal parafoveal at 1 mm [T1] and (e) temporal perifoveal at 2 mm [T2]

Figure 3.

The correlation between ChT and axial length for all five areas at baseline. No significant correlation was found between ChT at any of the measured five areas and axial length (P > 0.05). (a) subfoveal [SF], (b) nasal parafoveal at 1 mm [N1], (c) nasal perifoveal at 2 mm [N2], (d) temporal parafoveal at 1 mm [T1] and (e) temporal perifoveal at 2 mm [T2]

ChT differences between the groups at baseline

There was a statistically significant effect of area upon the horizontal ChT measurements, with (F (2.375, 99.767) = 63.032, P = 0.029). The resulted value of partial eta squared indicated that the area factor had a large effect on horizontal ChT among the three study groups. Post hoc analysis with a Bonferroni adjustment for the area factor showed that SF showed a statistically significantly greater ChT compared to all other 4 areas (P < 0.05). The N1 showed a greater ChT compared to N2 (P < 0.05) and thinner ChT compared to the T1 and T2 (P < 0.05). Similarly, the N2 area exhibited thinner ChT compared to T1 and T2 areas (P < 0.05). No statistically significant differences were found between the T1 and T2 areas (P > 0.05).

Moreover, the ANOVA and Kruskal-Wallis statistical tests showed that there was a statistically significant difference in mean scores of horizontal ChT within the right eye in the N1 area between the three groups with (χ² (2) = 6.548, P = 0.038). Bonferroni multiple comparisons showed that myope group (n = 27) reported a higher horizontal CT mean score within the N1 area compared to high myope group (n = 5) with P = 0.032. No other significant differences in mean scores of horizontal ChT at baseline within the other four areas were found between the three groups (P > 0.05).

ChT differences between the groups after caffeine consumption

Since the sphericity assumption was not met (χ²(5) = 24.591, P < 0.05), the results of tests within- subjects’ effects were taken from the Greenhouse-Geisser correction.

There was a significant reduction in ChT after taking 200 mg of caffeine in ChT measurements for all groups [Fig. 4] from 20 to 60 minutes (F (2.214, 92.978) = 25.372, P < 0.001). The largest reduction in ChT was after 40 minutes (differences in mean scores form baseline were 20.003, 34.881 and 29.425 µm for 20, 40 and 60 minutes respectively). Post hoc Bonferroni test showed that the decrease in ChT was statistically different between baseline and all other times (P < 0.001), between 20 and 40 minutes (P < 0.001) and 20 and 60 minutes (P = 0.002), and no statistically significant difference was found between 40 and 60 minutes (P > 0.05) [Fig. 4].

Figure 4.

The effect of caffeine consumption measured at three times upon ChT in all three refractive groups compared to baseline. The effect was greater after 40 minutes of consumption; however, there was no significant difference between the three refractive groups: (a) the emmetrope group, (b) the myope group and (c) the high myope group. The error bars represent the standard error

There was a statistically significant main effect of the interaction of time and refractive status factors on horizontal ChT measurements with (F (4.428, 92.978) = 2.575, P = 0.037). The resulted value of partial eta squared indicated that the interaction of time and refractive status factors has a medium effect size on horizontal CT. Although the thinning in ChT in the high myope group was lesser compared to both emmetrope and myope groups [Fig. 4], the between-subjects effects showed that there was no statistically significant effect found in horizontal ChT measurement mean scores among different groups with (F (2,42) =1.370, P > 0.05).

Discussion

Caffeine is known to have physiological effects on humans, including stimulating the central nervous system. It has been found to raise blood pressure, lower heart rate, and decrease cerebral blood flow, among other effects.[26] Caffeine’s uses include medicinal utilization as a form of asthma treatment.[27,28] More recently, Trier et al. (2023)[29] reported that an oral intake of 7-methylxanthine, which is a caffeine metabolite, could significantly reduce myopia progression and axial elongation in myopic children.

The ChT baseline measurements at SF in the current study showed similarities with earlier studies.[21,30,31,32,33,34,35,36] Additionally, the SF has a statistically significantly greater ChT than other quadrants, similar to those found in previous studies.[18,21,37,38] There are, however, some variations between this study’s findings and other previous studies in baseline measurements. One of these studies conducted by Flores-Moreno et al. (2013),[6] reported that SF ChT in myopic eyes could be as low as 131.3 ± 98.4 µm. Yet, however, the reason for such a significant reduction was that their study measured ChT for high myopic subjects with a mean SE was -14.34 ± 5.46D. Ho et al. (2013)[39] study showed that the SF ChT was 118 ± 68 μm for high myopic subjects, and the mean SE was −8.70D (interquartile range −6.10 to −11.00D). Besides the effect of refractive error upon SF ChT the further reduction in Ho et al. (2013)[39] study was due to the age factor. The age factor exhibited a statistically significant influence on SF ChT with an approximate reduction of 11.9 μm for each decade of life.

Generally, the reported lower or higher baseline values of ChT in the previous studies could be attributed to a variety of factors including the use of different OCT devices, scan patterns, and the distribution of participants by age, refractive status, ethnicity and, other factors.

Although it has been reported previously that ChT is significantly negatively correlated with axial length,[6,13] and decreased significantly with myopia,[14,15,16,17] our findings showed a weak correlation between the SE (as well as the axial length) and ChT, over the five areas, as the only statistically significant difference at the baseline was only at N1 area between myopic and high myopic groups. This may be explained by the unequal distribution of subjects between groups and the narrow range of SE for the high myope group reducing the impact of the SE factor upon ChT.

The findings of this study showed the expected thinning of the choroid after 200 mg of caffeine intake in agreement with previous studies.[23,24] Caffeine is one of the substances that could affect choroidal thickness among healthy individuals by inducing a temporary thinning of the choroid,[23,24] which could start after five minutes after caffeine consumption and could last for more than 6 hours.[24] This reduction in ChT is attributed to the vasoconstrictive effect of caffeine which leads to a significant increase in the vascular resistance of the choroidal vessels and a decrease in choroidal blood flow.[23] In the current study, the measurements were timed to take advantage of caffeine’s bioelimination properties, which reach peak plasma concentrations within 20-120 minutes.[40] In order to avoid diurnal fluctuations upon ChT, all OCT scans were conducted at the same time of the day, and the possible effect of age upon ChT was not present in the current study by recruiting younger participants.

ChT reduction after caffeine intake was lesser in the high myope group compared to the emmetrope and myope group; however, it was not statistically significant. This reduced could be a result of the lower CVI found in highly myopic eyes.[7,8,9,10,18] Gupta et al. (2014)[18] found that the subfoveal area had greater thickness compared to other choroidal areas and was significantly negatively correlated with axial length and SE, exhibiting that longer axial lengths could lead to a significantly smaller choroidal volume, resulting in lower CVI. However, a more recent study, conducted by Yazdani et al. (2021),[9] showed that lower CVI did not existed in lower myopic eyes, conversely, low myopes had a statistically significantly higher CVI, and lower stromal component, compared to emmetropes. Another recent study conducted by Wang et al. (2022),[10] found that CVI was not in a linear correlation with axial length, and the significant correlation was only apparent when the axial length reached approximately 27.26 mm. Both studies by Yazdani et al. (2021)[9] and Wang et al. (2022)[10] could explain the statistically insignificant difference between the refractive groups in ChT after caffeine consumption of our study due to two factors. First, 60% of our study’s participants were low myopes, which were not found to have lower CVI than emmetropes, and no statistically significant differences in ChT at baseline with other refractive groups. Second, the longest axial length in the current study was 25.94 mm which was not at the flexion point of axial length to affect CVI at 27.26 mm. Furthermore, there was an overlap between the three groups in axial length measurements [Table 1]. Thus, it could be that the insignificant variations in ChT and CVI among the three refractive groups were the reason for no significant differences in choroidal thinning after caffeine intake.

Although the acute transient effect of caffeine intake upon ChT has been established[23,24] the possible long-term effect, to our best knowledge, has yet to be evaluated. Such long-term studies will help in assessing the effect of vasoconstrictive upon ChT especially for subjects susceptible to choroidal damage due to a thinner choroid. Furthermore, in light of this study’s results and previous work, caffeine consumption up to an hour ahead of ChT examinations should be taken into consideration when measuring ChT for any purpose.

The findings of this study were subject to the limitation of a small sample size of forty-five participants with an unequal distribution of subjects between groups. Further studies should recruit a greater number of subject and should have equal or relatively equal number of participants in each group.

Another limitation could be the manual measurement of ChT which may raise the concerns of bias among examiners. However, the measurement was obtained from five different areas, and there was excellent reliability of the instrument between the two examiners in measuring ChT. Therefore, any bias resulting from manual measurement is not likely to be significant and might have only a minor effect upon the final results and conclusion of this study.

Conclusion

The results of this study suggest that the reduction in ChT due to caffeine consumption is not significantly affected by the refractive status of the eye. Thus, it is safe to hypothesize that the vascular part of the choroid behaves identically to vasoconstrictive in both emmetropic and myopic eyes. Choroidal thinning in myopic and high myopic eyes, as well as the reduced choroidal vascularity index, were not observed in this study. Therefore, further longitudinal studies recruiting greater numbers of participants, including myopes and high myopes, and measuring both vascular and stromal layers to investigate such variations are warranted.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.