Iris color and day–night changes in the sympathovagal ratio

Abstract

Iris melanocytes are innervated by parasympathetic and sympathetic nerve endings. Light affects autonomic nervous system activity via the retino-hypothalamic pathway. The hypothesis that the day-to-night variations in the sympatovagal ratio (LF/HF) may differ among individuals with different brown iris patterns was tested. A total of 621 healthy adults, aged between 16 and 50, with brown eyes and not diagnosed with a disease that might affect the autonomous nerve system were included in the study. A digital camera was used to acquire iris photos. Subjects were grouped into iris color groups (2–0 bg, 1–0 bg, 1–1 db, 1–1 lb, 2–0 b, and 1–0 b). Iris photos were analyzed with Picture Color Analyzer RBG software. The Central/Peripheral (R/RGB) ratio was used for objective distinction between the groups. Using 24-h Holter ECG monitoring, the change in the sympathovagal ratio from day (between 07:00 and 23:00 h) to night (between 23:00 and 07:00 h) was determined with the formula [(Day–Night) LF/HF)/Day LF/HF]. The frequency of subjects with a decrease in the LF/HF ratio from day to night was the highest in the 1–1 db group (65.7%), followed by the 1–1 lb group (56.4%). The highest increase was in the 2–0 bg group (76.5%), followed by the 1–0 B group (68.9%) (p < 0.001). Based on the findings of this study, iris color may be a predictive factor in diseases in which the circadian change of autonomic nervous system activity is effective.

Introduction

The iris not only adjusts, via parasympathetic and sympathetic stimulation of the muscles, the amount of light entering the pupil area; using melanin, which gives the eye its color, it also absorbs a portion of the near infrared, visible, and UV spectra and thus determines light quality [1]. Iris melanocytes are innervated by parasympathetic and sympathetic nerve endings [2]. The dihydroxyphenylalanine (DOPA) derivative melanin is found in melanocytes. The melanocyte lineage is derived from the neural crest [3].

Light acts on the autonomic system via the retinohypothalamic pathway [4]. The hypothalamic suprachiasmatic nucleus (SCN), the circadian center, is connected with the retinohypothalamic pathway and with autonomic circuits descending to peripheral organs such as the adrenal glands [5]. Day–night autonomic activity is affected by sleep, melatonin, and changes in hormonal levels and circadian rhythm. In the early morning hours, systemic blood pressure is higher, and myocardial ischemia, sudden cardiac death, and stroke attacks are more frequent. Platelet aggregation, plasma epinephrine levels, renin activity, and cortisol secretion rates are also higher at this time of day [6].

The sympathetic and parasympathetic activity ratio (LF/HF) can be determined using heart rate variability (HRV). A large LF/HF value indicates a high stress level and a strong influence of the sympathetic nervous system (NS), a weak influence of the parasympathetic NS, or both [7]. Sleep creates a unique environment for the measurement of autonomous activity due to the absence of factors including physical activity and high cortical function [8].

The change in the LF/HF ratio between day and night was examined to test the hypothesis that the effect of absorption of different wavelengths of light on the LF/HF ratio can vary between individuals with different brown iris patterns. Thus, the study aim was to determine whether iris color can be used in the risk estimation of diseases with increased activity of the autonomic nervous system, which have an increased frequency of occurrence during either day or night hours.

Methods

A total of 650 healthy individuals between the ages of 16 and 50 who applied for cardiac examination and had normal findings were recruited for the study. A cardiac clinical assessment was performed, including a medical history, physical examination, standard biochemical analyses, resting 12-lead ECG, and echocardiography. Resting 12-lead ECG intervals were normal. There were no structural or functional abnormalitis on echocardiography. The participants did not have any diseases that could affect the autonomic nervous system (obesity, obstructive sleep apnea syndrome, hypertension, diabetes mellitus, atrial fibrillation, coronary disease, heart failure, goiter, etc.). They were not consuming alcohol or caffeine or taking medications or vitamins. The subjects’ systolic and diastolic blood pressure, pulse rate, hemoglobin, creatinine, glucose, and LDL cholesterol values were within normal limits.

Iris color measurement was performed on the subjects. LF/HF measurements were performed with the 24-h Holter monitoring system. Twenty-nine patients were excluded due to excessive artefacts (20%) at this stage, so 621 subjects were included in the study.

The study was conducted according to the Helsinki declaration, approval was obtained from the ethical board (Ankara Keçiören Education and Training Hospital 9.3.2016/1099), and the subjects signed an informed consent form.

Iris color measurement

Photos were taken of the iris with a digital camera (10megapixel camera, 1.75 μm sizepixels, 5 × digital zoom, LED flash). In the iris, the area between the collaretta and the pupil was identified as the central iris and the area between the collaretta and the limbus as the peripheral iris. Although a widely accepted classification is not available for eye color, our classification rather resembles that of Mackey for brown eyes [9]. Subjects were grouped into six classes according to the iris structure observed with the naked eye or a simple light source as follows:

Group 1–2: Those with distinctively dark brown in the central area, slightly green in the peripheral area, 2–0 darkbown–green (2–0 bg) and those with slightly dark brown in the central area and green in the peripheral area, 1–0 lightbrown–green (1–0 bg). Group 3–4: Subjects with dark brown (1–1 db) and light brown (1–1 lb) eyes with the same color structure in the central iris and the peripheral iris. Group 5–6: Those with distinctively dark brown (2–0 b) and slightly dark brown (1–0 b) in the central iris area compared to the periphery (Fig. 1).

Fig. 1.

Iris color groups. First row: 1.1-1darkbrown (1-1db), 2.1-1lightbrown (1-1lb)[same color structure in the central iris and the peripheral iris], 3.1-0brown (1-0b) [slightly brown in the central iris area] Midddle row: 4. 2-0brown(2-0b)[distinctively dark brown in the central iris area], 5.1-0 lightbrowngreen (1-0bg)[slightly dark brown in the central area and green in the peripheral area], 6.2-0 darkbrowngreen (2-0bg)[distinctively dark brown in the central area, slightly green in the peripheral area]. Bottom row: Determination of central–peripheral–central iris borders and measurements of red,blue, and green color dimensions with Picture Color Analyzer Red Blue Green software

For the analysis of iris color, iris photos were analyzed with Picture Color Analyzer RBG (red–blue–green) software (downloadable from www.isao.com). The RGB system is a measure of the amount of red, green, and blue hues present on an image. The three color dimensions of the RGB system yielded numbers ranging from 0 to 255, with smaller numbers indicating darker colors. When the value of each component is 255, the result is pure white, and when all components have values of 0, the result is pure black. These numbers are called color values. The picture color analyzer automatically calculates the total and average RGB values. As the study covered brown eyes, the ratio of red color to the total of the tones of all colors (R/RGB) was taken for the central and peripheral parts as the color measurement indicator. The Central/Peripheral (R/RGB) ratio was used for objective distinction between the groups [10].

Intraobserver reliability studies were conducted by having the observers reanalyze 100 randomly selected iris photos. Interobserver reliability studies were performed by having a second observer reanalyze the same 100 photos immediately after they were analyzed by the first observer.

LF/HF measurement

Using the three-channel standard ambulatory Holter recording system (DMS Software Cardioscan II Holter monitoring system, version 11.4.0054a), after deleting the recorded ectopic beats, artifacts, low frequency (LF, 0.04–0.15 Hz) and high frequency (HF, 0.16–0.4 Hz) measurements were taken from day and night changes in the heart rate by using the frequency-domain analysis method. In accordance with the recommendations [11], only sinus-derived RR intervals were used to evaluate the mean RR interval and HRV parameters. For Holter ECG monitoring, the recording time was < 22 h, and if the number of artifacts and ectopic beats was > 20% of the total number of beats, it was invalidated. The change from day (between 07:00 and 23:00 h) to night (during sleep, between 23:00 and 07:00 h) in the LF/HF sympathovagal ratio was determined with the formula [(Day LF/HF – Night LF/HF)/Day LF/HF].

Statistical analyses

An a priori power analysis was conducted using G*Power version 3.1.9.7 (Faul et al., 2007) for sample size estimation (α = 0.05, power = 0.80, effect size: 0.5). Statistical analyses were performed using IBM SPSS Statistics version 17.0 software (Armonk, NY). The Kolmogorov–Smirnov test was used to check the data for the normal distribution. Numerical parameters showing a normal distribution were presented as mean ± standard deviation, and those that did not show a normal distribution were presented as median (min–max). Categorical variables were expressed as numbers and percentages. Comparison between two groups was performed using a Student’s t test or Mann–Whitney U test. Comparison between iris groups was performed with a Kruskal–Wallis test followed by Dunn's post hoc multiple comparisons test. Categorical variables were compared using the chi-square and Fisher’s exact chi-square test. Intraobserver and interobserver reliability analyses were performed. Differences were considered statistically significant at p < 0.05.

Results

The study population was comprised of 312 female and 309 male subjects. The subjects’ ages ranged between 16 and 50 years, and the mean (± SD) age was 31.6 ± 7.7. The subjects had a median (min–max) systolic blood pressure of 100 (80–130) mmHg, diastolic blood pressure 70 (50–90) mmHg, pulse rate 64 (60–100) bpm, haemoglobin 12 (10–18) g/l, creatinine 0.8 (0.4–1.2) mg/dl, glucose 85 (56–114) mg/dl, and LDL cholesterol 90 (56–129) mg/dl.

Color measurements

The median (min–max) color measurements taken in the central area in 621 subjects were red: 26 (11–60), green: 19 (6–40), blue: 5 (1–22), and R/RGB: 543 (430–675). In the peripheral area, they were red: 34 (14–63), green: 26 (7–47), blue: 9 (2–24), and R/RGB: 492 (372–660).

As a result of red, green, and blue (dimension) measurements in the central and peripheral areas of the visually defined groups, the median (min–max) Central/Peripheral (R/RGB) ratio in the entire population was 1.09 (0.9–1.4). The median (min–max)Central/Peripheral (R/RGB) ratio by group was 2–0 bg: 1.19 (1.1–1.4), 1–0 bg: 1.1 (0.99–1.15), 1–1 db: 1.02 (0.94–1.1), 1–1 lb: 1.03 (0.97–1.07), 2–0 b: 1.18 (1.08–1.4), and 1–0 b: 1.1 (0.97–1.16).

Reproducibility

The intraobserver and interobserver intraclass correlation coefficients for color measurements were 0.84 [confidence interval (CI) 0.81–0.86] and 0.86 (CI 0.83–0.88), respectively.

Change in the LF/HF ratio in the different iris color groups

The subjects’ median LF/HF ratio differed between day and night. The percentage of subjects with an increase and decrease in the LF/HF ratio and magnitude of change in the LF/HF ratio are shown in Table 1. The mean age, LF/HF ratios, and increase/decrease in the LF/HF ratio from day to night did not differ significantly by gender. The data are shown in Table 1.

Table 1.

Distribution of the LF/HF ratio in all patients and by gender

All patients, Age(year)*	31.6 ± 7.7
LF/HF**	Day 3.3 (2.3–4.5)	Night 2.6 (1.5–4.2)
Mean	2.8 (2–4.1)
Δ LF/HF (%)	Increase n:251 (%40.4)	Decrease n:370 (%59.6)
	31(18–42)	47 (34–59)
Gender groups	Female n:312(%50.2)	Male n:309(%49.8)	p
Age*	31.5 ± 7.7	31.6 ± 7.6	0.943
LF/HF**-Day	3.1 (2.2–4.4)	3.4 (2.4–4.6)	0.071
Night	2.3 (1.4–4.1)	2.6 (1.6–4.1)	0.311
Mean	2.6 (1.8–4)	3 (2.1–4.2)	0.077
LF/HF increase, n(%)	131 (42.0)	120 (38.8)	0,424
Δ LF/HF (%)**	28 (18–42)	32 (18–42)	0.588
Decrease, n(%)	181 (58.0)	189 (61.2)	0,424
Δ LF/HF (%)	47 (37–57)	49 (34–59)	0.933

*Mean ± SD

**Median (IQR) ΔLF/HF(%): LF/HF_(Day-Night)/LF/HF_Day.100

The mean age and gender ratios were not different between iris groups. The median day LF/HF ratio was the highest in the 1–1 db and 1–1 lb groups. The median night LF/HF ratio was the highest in 2–0 bg and 1–0 b groups. Of the subjects with an increase (n, %) in the LF/HF ratio from day to night, the highest ratio was found in the 2–0 bg, 1–0 b, 2–0 b, and 1–0 bg groups and the lowest in the 1–1 db group, followed by the 1–1 lb group. Of the subjects with a decrease (n,%) in the LF/HF ratio from day to night, the ratio was the highest in the 1–1 db group, followed by the 1–1 lb group. The data are shown in Table 2, Fig. 2.

Table 2.

LF/HF ratio and the change in iris color groups

Variables	2–0 BG	1–0 BG	1–1 DB	1–1 LB	2–0 B	1–0 B	p
	n = 102	n = 105	n = 108	n = 101	n = 102	n = 103
Age*	33.3 ± 7.2	32.9 ± 8.1	30.1 ± 6.9	30.9 ± 7.3	30.9 ± 8.8	31.3 ± 7.3	0.074
Gender, n(%) Female	51 [50]	55 [52.4]	58 [53.7]	46 [45.5]	50 [49.0]	52 [50.5]	0.897
Male	51 [50]	50 [47.6]	50 [46.3]	55 [54.5]	52 [51.0]	51 [49.5]
LF/HF ratio**—Day	2.1(1.3–3.4)¥§	2.3(1.33–4.15)¥§	3.5(2.2–5)ǂ†§ϕ҂	2.7(2–4.1)ǂ†¥ϕ҂	2(1.3–3)¥§	2.4(1.4–4.5)¥§	< 0.001
Night	3.47(2.6–5.2)¥§ϕ	3.4(2.4–4.97)¥§ϕ	3.1(2.2–4)ǂ†§҂	2.6(1.8–4.1)ǂ†¥ϕ҂	3.1(2.1–4.1)ǂ†§҂	3.6(2.4–5.1)¥§ϕ	< 0.001
Mean	2.93(1.79–4.27)	3 (2–4.1)	3.3(2.7–4)	2.6(2–3.8)	2.5(1.9–3.8)	3(2–4.7)	0.103
Increase n(%)	78(76.5)	70(66.7)	37(34.3)	44(43.6)	70(68.6)	71(68.9)
Δ LF/HF(%)	55(44–64)¥§	51(37–60)¥§	26(10–64)ǂ†ϕ҂	26(14–41)ǂ†ϕ҂	50(41–57)¥§	46(33–59)¥§	< 0.001
Decrease n(%)	24(23.5)	35(33.3)	71(65.7)	57(56.4)	32(31.4)	32(31.2)
Δ LF/HF(%)	32(23–43)¥	25(15–34)¥	54 (35–71)ǂ†§ϕ҂	18(9–31)¥	23(15–34)¥	24(11–34)¥	< 0.001

*Mean ± SD or **median (IQR). ǂ: 2–0 BG,†:1–0 BG,¥: 1–1 DB, §: 1–1 LB, ϕ: 2–0 B, ҂:1–0 B (p < 0,05 Dunn’s test adjusted)

Δ LF/HF: LF/HF_Day-Night / LF/HF _Day

Fig. 2.

LF/HF ratio change from day to night in iris color groups

Gender analysis

In each iris group, the mean ages of men and women were similar. In both genders, the median day LF/HF ratio was significantly higher in the 1–1 db and 1–1 lb groups. The decrease from day to night as a percentage of the total subjects was the highest in the 1–1 db and 1–1 lb groups and the lowest in the 2–0 bg and 2–0 b groups in females and the 2–0 bg and 1–0 b groups in males. The rate of increase in the nighttime LF/HF ratio was higher in the 2–0 bg, 1–0 bg, 2–0 b, and 1–0 b groups in both genders. The data are shown in Table 3.

Table 3.

LF/HF ratio distribution in female and male groups

Female	2–0 BG n = 51	1–0 BG n = 55	1–1 DB n = 58	1–1 LB n = 46	2–0 B n = 50	1–0 B n = 52	p
Age*	33.7 ± 7.7	32 ± 7.9	29.9 ± 6.8	31.7 ± 7.6	31 ± 8.7	31.1 ± 7.2	0.207
LF/HF** -Day	2.1 (1.4–3.5)¥§	2.0 (1.4–4.3)¥§	3.6 (2.1–5.4)ǂ†§ϕ҂	2.6 (1.8–4.1)ǂ†¥ϕ҂	1.7 (1.1–2.6)¥§	2.1 (1–4.9)¥§	< 0.001
Night	3.4 (2.4–4.6)§ϕ	3.4 (2.4–4.6)§ϕ	3.2 (2.1–4.2)§ϕ	2.7 (2–4.1)ǂ†¥҂	2.5 (2–3.7)ǂ†¥҂	3.1 (2.3–4.7)§ϕ	0.016
Mean	2.8 (1.8–4.2)¥ϕ	2.6 (2–4.1)¥ϕ	3.3 (2.2–4)ǂ†§ϕ҂	2.6 (2–3.8)¥ϕ	2.3 (1.8–4.7)ǂ†§¥҂	2.8 (1.8–4.7)¥ϕ	0.034
İncrease, n(%)	40 (78.4)	37 (67.3)	18 (31.0)	20 (43.5)	33 (66.0)	33 (63.5)
Δ LF/HF(%)	54 (43–63)¥§	46 (37–60)¥§	37 (26–46)ǂ†ϕ҂	32 (19–44)ǂ†ϕ҂	50 (43–54)¥§	52 (45–60)¥§	< 0.001
Decrease, n(%)	11 (21.6)	18 (32.7)	40 (69.0)	26 (56.5)	17 (34.0)	19 (36.5)
Δ LF/HF(%)	23 (18–45)¥	24 (15–34)¥	45 (32–61)ǂ†§ϕ҂	19 (8–25)¥	23 (15–32)¥	24 (9–34)¥	< 0.001

Male	n = 51	n = 50	n = 50	n = 55	n = 52	n = 51
Age*	32.9 ± 6.6	33.9 ± 8.2	30.3 ± 7.1	30.2 ± 6.9	30.9 ± 9	31.4 ± 7.3	0.078
LF/HF** -Day	2.1 (1.1–3.4)¥	2.7 (1.1–4)¥	3,6 (2.3–5)ǂ†§ϕ҂	2.6 (1,9–3.7)¥	2.4 (1.6–4.1)¥	2.6 (1.6–4.1)¥	0.013
Night	3.6 (2.4–5.2)¥§	3.6 (2.4–5.5)¥§	3 (1.8–3.8)ǂ†§ϕ҂	2.6 (1.9–3.5)ǂ†¥ϕ҂	3.7 (2.4–5.2)¥§	3.8 (2.9–5.8)¥§	< 0.001
Mean	3 (1.6–4.4)	3.2 (2–4.2)	3.3 (2.2–4.2)	2.6 (1.9–3.6)	3 (2.3–4.3)	3.2 (2.2–4.4)	0.362
İncrease, n(%)	38 (74.5)	33 (66.0)	19 (38.0)	24 (43.6)	37 (71.2)	38 (74.5)
Δ LF/HF(%)	56 (46–67)¥§	54 (40–60)¥§	39 (28–50)ǂ†ϕ҂	23 (6–32)ǂ†ϕ҂	50 (41–60)¥§	47 (35–58)¥§	< 0.001
Decrease, n(%)	13 (25.5)	17 (34.0)	31 (62.0)	31 (56.4)	15 (28.8)	13 (25.5)
Δ LF/HF(%)	33 (24–41)†¥§ϕ҂	28 (17–43)ǂ¥	64 (52–74)ǂ†§ϕ҂	16 (9–32)ǂ¥	23 (14–33)ǂ¥	23 (10–32)ǂ¥	< 0.001

*Mean ± Sd or **Median (IQR)

ǂ2–0 BG,†: 1–0 BG,¥: 1–1 DB, §:1–1 LB, ϕ:2–0 B, ҂: 1–0 B (p < 0,05 Dunn’s test adjusted)

Δ LF/HF (%): LF/HF_Day-Night/LF/HF_Day

Discussion

In our study, the daytime LF/HF ratio was determined to be greater in 1–1 db group, in which the quantity of melanin seems higher because of the dark color compared to other groups in our study, but there was no difference in the mean LF/HF ratio between the groups. The highest decrease in the LF/HF ratio from day to night was found in the 1–1 db group (65.7%) and the highest increase in the 2–0 bg group (76.5%). There was no gender-based difference in the change in the LF/HF ratio.

The fact that the LF/HF day–night variation interval is quite large suggests that there may be many unpredictable factors responsible for this variation, including differences in physical and emotional activity between subjects during the day. During sleep, vagal activity is high, and sympathetic activity is relatively low [8].

Cholinergic antagonists reduce the LF peak is by at least 50%. It has been reported that the LF/HF ratio increases by 1.1–8.4 when such agonists are administered in combination with selective parasympathetic denervation and beta-adrenergic receptor blockade and does not fully reflect the increase in sympathetic activity. Respiratory parameters can also alter the heart rate and R–R interval variability independently of cardiac autonomic regulation [12].

Iris color, melanin, and adrenergic innervation

It has been hypothesized that melanin content is associated with norepinephrine and cortisol generation, as dark colored eyes are more reactive than light colored ones and exhibit greater pupil dilatation in response to medication. Eye color darkness is associated with the nerve impulse rate, as the amount of iris melanin is embryologically related to the central nervous system. The melanin cover on neuronal axons functions as an insulator in the central nervous system, allowing nerve impulses to be transmitted faster [13–15]. Therefore, it can be hypothesized that sympathetic stimulation of the dilator muscle and cholinergic stimulation of the iris sphincter muscle may be related to the melanocyte concentration in the overlying stroma.

Alpha melanocyte stimulating hormone (αMSH), which stimulates melanocytes, adrenocorticotropic hormone, and endogenous opioids, is formed by cleavage of proopiomelanocortin [16]. Experimentally, the activity of tyrosinase, which converts tyrosine to DOPA in the iris, increased by 100% when the eye was exposed to sunlight. However, no significant change was detected in iris color [17]. It has been found that αMSH darkens the frog iris in vivo [18].

Mukuno et al. identified four different contacts, two of which had a cholinergic and the other two an adrenergic appearance, between melanocytes and nerve terminals in the stroma in human iridectomy materials [2]. Using protein structural data, a hypothesis was asserted that the structures and functions of cathecolamine receptors are similar to those of visual pigments [19]. Adrenergic innervation has been demonstrated to be important in iris pigmentation, as hypo-pigmentation develops in the iris after damage to postganglionic sympathetic fibers [20].

Melatonin, melanin, and the autonomous nervous system

The presence or absence of light, sleep hormone, and melatonin are expected to contribute to the change in the day–night LF/HF ratio. The photoentrainment of the circadian rhythm begins with the stimulation of melanopsin-containing retinal ganglion cells that respond directly to blue light [21]. Retinal light reaches the SCN through the retinohypothalamic pathway. Melatonin is secreted in the dark during sleep from the pineal gland, controlled by the SCN, which maintains the sleep–wake cycle and biorhythm. The superior cervical ganglion, the main sympathetic ganglion of the heart, stimulates mydriasis of the iris and is associated with the SCN, pineal gland, and hypothalamus.

Melatonin contributes to melanin synthesis by adjusting MSH secretion. A significant increase was determined in the number of melanocytes directly exposed to melatonin [22]. In rabbit iris-cilier processes, specific binding areas for I¹²⁵-iodo melatonin were identified [23]. Melatonin has a regulatory effect on the autonomous nervous system. Heart rate, blood pressure, and vascular tone are reduced during the night. This reduction in cardiovascular activity is partially correlated with autonomic nervous activity [24]. It seems likely that the melanin-stimulating and autonomic activity-regulating effects of melatonin at night were more effective in the 1–1 db group.

The stimulatory effect of wavelengths of visible light

Brown eyes (especially those in the 1–1 db group) have a lot of melanin in the iris stroma, while long wavelength rays, especially red, reflect and contribute to the eye color. This group absorbs short-wavelength light, especially the blue component of visible light, and the effects of this wavelength are more dominant. The opposite is true in the case of blue eyes [25].

Blue light (wavelength 460 nm) is more stimulating than green light (wavelength 550 nm) and has been shown to cause a decrease in EEG power density in the delta-theta range and a decrease in attention deficit. Exposure to blue light elicites responses in working memory and cognition via arousal adjustment and caused more emotional stimulation [26]. Lights in the blue–green–yellow region of the spectrum were compared in terms of their effect on subjective alertness, and the maximum effect was obtained with a very short wavelength (420 nm) [27].

Irises lighter in color transmit more light [28], exhibit greater light scattering, and might elicit more non-image-forming responses [29]. Non-visual light responses, such as melatonin suppression, circadian phase shifting, sleep latency, and elevation of the heart rate and core body temperature, are more sensitive to a short light wavelength (~ 480 nm, blue-opaque, sky color), whereas longer wavelengths of light do not cause significant melatonin suppression, and these responses might be regulated by melanopsin-based photoreception [30].

As a result, except for the possible complex mechanisms outside the scope of this study [1], there was more melanin in the stroma over the iris sphincter muscle in the 1–0, 2–0 b, and 1–0, 2–0 bg groups. Absorption of short wavelength light in the central area of the iris in these groups may have increased the parasympathetic effect in the daytime and thus may have affected the results. M3-muscarinic receptors are involved in the contraction of the central parasympathetic iris muscle region. The receptors mainly co-expressed with M₃ are M₂ receptors, which cause bradycardia [31]. M3 receptors adjust the secretion of melanin-concentrating hormone in the hypothalamic feeding center and thus regulate apetite [32].

In the 1–1 db and 1–1 lb groups, where there is possibly more melanin over the entire iris, short-wavelength light (blue) may have an effect on sympathetic activity to increase the daytime LF/HF ratio.

Limitations of the study

One limitation of the study is that HRV is an indirect measure of cardiac autonomic control. The second is that we did not measure sleep quality and duration, which can affect the measurement. Quantitative methods have been developed to determine iris color in photos taken under standard conditions. It has also been determined that RGB color space and digital imaging techniques can be used to estimate the melanin content of irises of different shades and colors [33, 34]. Other than melanin, the iris color seen from the cornea is affected by multiple light scattering in pigment granules, other components of the connective tissue constituting the iris stroma, light absorption by various chromophores, and other optic events. The third limitation is the possibility that even though the photos were taken with the same digital camera in the same location, the exact same lighting conditions might not be present on different days. Adequate discrimination of the Central/Peripheral (R/RGB) ratio was not always possible.

Conclusion

The frequency of subjects with a decreasing LF/HF ratio was highest in the 1–1 db and 1–1 lb groups at night. The greatest decrease in the magnitude of this ratio was in the 1–1 db group at night. A higher frequency of subjects with an increasing LF/HF ratio was detected in other groups at night.

Author contributions

ŞK: selection of subjects, their clinical evaluation, acquisition of Holter ECGs and their analysis, iris photo shoot, iris color measurement, interpretation of data for the work, and writing. AEK: data interpretation, iris color measurement, final approval of the version to be published, revision of the manuscript.

Funding

None.

Declarations

Conflict of interest

Authors declare that they have no conflict of interest.