Visual photosensitivity threshold and objective photosensitivity luminance in healthy human eyes assessed using an automated ocular photosensitivity analyzer: a step towards translation of a clinical tool for assessing photophobia

Abstract

Purpose:

To evaluate visual photosensitivity threshold and objective photosensitivity luminance in healthy eyes, thereby providing a normative dataset that will lead to a better understanding of diseases causing photophobia.

Methods:

It was a prospective cross-sectional study. Emmetropes whose visual acuity was better than 0.18 logMAR (6/9) with no other ocular abnormality were included. Headache Impact Test-6 questionnaire and visual light sensitivity questionnaire were administered. Visual photosensitivity threshold was measured subjectively using ocular photosensitivity analyzer. Objective photosensitivity luminance was assessed manually by evaluating the videos recorded using infrared camera and noting down the intensity of light at the first squeezing reflex.

Results:

A total of 75 normal subjects (age range, 7–71 years) were included. Median age was 32.7 years (inter-quartile range, 20.3–47.9 years). Forty subjects (53.3%) were males. Median Headache Impact Test score was 38 (inter-quartile range, 36–42) and visual light sensitivity questionnaire score was 11 (inter-quartile range, 8–15). Mean (standard deviation) right eye, left eye and binocular visual photosensitivity threshold was 3.34 (0.78), 3.33 (0.81) and 3.37 (0.78) loglux respectively. There was a significant negative correlation of visual light sensitivity questionnaire scores with right eye, left eye and binocular visual photosensitivity thresholds, and positive correlation of age with binocular visual photosensitivity thresholds. Mean (standard deviation) right eye, left eye and binocular objective photosensitivity luminance was 3.25 (0.55), 3.35 (0.47) and 3.15 (0.52) loglux respectively. Age was positively correlated with only binocular objective photosensitivity luminance and there was no correlation between age and right eye or left eye objective photosensitivity luminance.

Conclusions:

The study characterized, for the first time, objective photosensitivity luminance and established a normative data for both visual photosensitivity threshold and objective photosensitivity luminance. The data would help in understanding the pathophysiology of the diseases causing photophobia, monitoring the disease progression and thereby evaluating the treatment modalities.

Introduction

Photophobia is an abnormal intolerance or discomfort to the visual perception of light. Associated features include frequent blinking, squeezing and closure of eyelid and, if worse, can include tearing, headache, nausea and blurred vision.^1,2 The term photosensitivity usually refers to a physiologic response of closing eyes only on exposure to intense light, while photophobia is caused even by typical ambient room illumination. Photophobia can be due to underlying ocular disease conditions including aniridia, benign essential blepharospasm, dry eyes, keratitis, retinal dystrophies, etc.^2–5 Photophobia is also a common presenting symptom in neurological conditions such as migraine, traumatic brain injury and some psychiatric conditions.^6–8 The pathophysiology of photophobia is unclear.² In severe cases, photophobia limits activities of daily living including reading, driving and working.³ Given photophobia is a common symptom for a myriad of diseases ranging from a simple headache to a dangerous pituitary tumor, understanding and quantifying photophobia is important for proper diagnosis and management.⁹

Several attempts were made in the past to assess the severity of photophobia. Most of them included subjective responses from questionnaires and a majority were published as case reports.⁶ Development of objective assessments started with a consideration of light-induced responses associated with photophobia such as squinting and lacrimation.^4,10 But these could not give reliable results in certain conditions such as dry eye syndrome, lacrimal duct disorders and blepharospasm, where lacrimation and blinking are associated symptoms of the disease itself irrespective of photophobia. Hence, psychophysical tests measuring photosensitivity thresholds are important in understanding the severity of photophobia and help in disease monitoring, progression and evaluating the effect of treatment given.

In 1997, Vanagaite et al investigated light discomfort and pain in patients with migraine, tension-type and cervicogenic headaches and in normal controls. The setup included halogen lamp, grey glass and luximeter, and light intensity was controlled by a personal computer.^11,12 Using a very similar setup, Adams et al measured photophobia in patients with benign essential blepharospasm.³ An infrared camera and a binocular pupillometer were incorporated in the above-mentioned setup by Cortez et al to assess the pupillary responses in migraine patients.¹³ Stringham et al in 2003 measured the photophobic discomfort using three channelled Maxwellian optical viewing system with xenon. The authors also assessed squinting response corresponding to photophobia by electromyography.⁴ A similar setup was used to design a novel test for evaluating light-induced discomfort using red and blue lights.¹⁴ Although these studies helped in understanding basic design for testing photosensitivity thresholds, they were not easily acceptable in a clinical setting as they were dependent on bulky instruments creating heat and thermal fluctuations.

In view of above, the Ophthalmic Biophysics Center at Bascom Palmer Eye Institute (BPEI), Miami, USA, created the ocular photosensitivity analyzer (OPA) for the assessment of visual photosensitivity threshold (VPT) and Dr Byron Lam at BPEI developed the visual light sensitivity questionnaire (VLSQ-8). The reliability and repeatability of the instrument were validated on 35 healthy emmetropic subjects.¹⁵ Aguilar et al improved upon the first design and developed the second generation OPA to yield VPT reliably in healthy and retinal diseased subjects over time in a small sample of US population.¹⁶ The purpose of the present study was to determine VPT and a new parameter, objective photosensitivity luminance, on a large number of healthy eyes in the Indian population using the second generation OPA, thereby providing a normative dataset that will lead to a better understanding of diseases causing photophobia.

Methods

It was a prospective cross-sectional study conducted at a tertiary eye care center in 2018 and 2019. The study was performed with an approval from the Institutional Ethics Committee, L V Prasad Eye Institute, Hyderabad, India. All procedures were carried out in accordance with the tenets of the Declaration of Helsinki. Informed consent was obtained from each subject before participation. The subjects were emmetropes with visual acuity of better than 0.18 logMAR (6/9) in each eye and no other ocular condition. Subjects with known eye diseases such as dry eyes, a history of any intraocular surgery, systemic conditions such as diabetes mellitus which can cause ocular manifestations, vision worse than or equal to 0.18 logMAR (6/9) in each eye and neurological conditions including migraine were excluded from the study.

Each subject underwent regular eye examinations that include assessment of visual acuity and colour vision, slit lamp examination, intraocular pressure (IOP) measurement and Schirmer’s II test. Only subjects with normal color vision and normal anterior and posterior segments on examination were included. Headache impact test-6 (HIT-6) was administered to rule out patients with severe headache or migraine.¹⁷ The HIT-6 questionnaire measures the impact and effect of headache on the ability to function normally in daily life. The least possible score is 36 and highest is 78. Scores ≥60 are indicative of severe life impact, 56–59 of substantial life impact, 50–55 of some life impact, and ≤49 of little to no life impact. Subjects with HIT scores ≥50 were not included in the study to represent a normative study population. VLSQ-8 was administered to assess photophobia subjectively. The range of VLSQ score is 8–40 and a response that is higher than score 3 in each of the 8 questions is considered as having photosensitivity symptom. Subjects with a total VLSQ score of >24 were not included in the study.

Photosensitivity test was performed in a dedicated room for the OPA setup. OPA utilizes an automated method to measure VPT of an individual using Garcia-Parez staircase technique described elsewhere.^15,16 In brief, the light source contains 210 white light-emitting diodes (LEDs) assembled in bi-cupola curved surfaces mounted on an enclosure. Besides, it also consists of a blinking white LED of 0.14 lux for fixation. An infrared video camera is integrated into the LED array enclosure to record real-time video of the subject’s eyes. Synthesized speech incorporated in the touch screen laptop gives the instructions to the subject to carry out the test. The light stimuli are presented for a fixed duration of 2 seconds and the subject has to notify whether the light was uncomfortable by pressing the handheld push button. Based on the response of the subject, the stimulus of the light either decreases or increases. The VPT of the subject was calculated from the mean of 10 response reversals. An orthoptic eye patch for occlusion therapy (Opticlude, 3M India Limited, Pune, India) was used to occlude the non-tested eye. The order of testing was always the same i.e. binocular first, followed by right eye and finally left eye.

An objective assessment called objective photosensitivity luminance (OPL) was performed after the subjective assessment of thresholds. OPL is the light intensity at which there is a response from the subjects that can be attributed to light discomfort. The discomfort responses are one of the following: squeezing both eyelids, frowning, squinting the eyes or closure of the eyes while viewing the light. The evaluating investigator was provided with all experimental videos in a random order and was blinded to subject’s information. The assessment of OPL was done manually by playing the recorded videos since the beginning of the test, carefully looking for the discomfort responses and noting down the intensity of light when the first squeezing reflex was observed. It is a task that required a careful attention not to miss the squeezing reflex and not to be confused with an eyelid blinking. The investigator who evaluated the videos was masked to the luminance value by covering the video frame portion (with a sticky note) that displays it, while a second investigator noted down the displayed luminance value. Figure 1 (see also supplementary video) shows still images from a video that show the luminance that was considered a ‘squeezing reflex’ (Figure 1B) and the one at a slightly lower luminance that was not (Figure 1A). Videos which had frame pauses or freezes where OPL cannot be determined reliably were excluded.

Figure 1: Evaluation of objective photosensitivity luminance –

Figure 1 shows still images from a video that show the luminance (1085.33 lux) that was considered a ‘squeezing reflex’ (B) and the one at a slightly lower luminance (571.70 lux) that was not (A). The subjective visual photosensitivity threshold in this instance was 4514.20 lux.

The statistical analysis was performed using Origin v7.0 (OriginLab Corporation, Northampton, MA, USA) and STATA v14.2 (StataCorp, College Station, TX, USA). As photosensitivity bears a logarithmic relationship with the light intensity, the VPT values were converted to logarithmic values (log₁₀lux). Shapiro-Wilk test was performed to check for the normality of the data. Descriptive statistics included mean with standard deviation (SD) for parametric data and median with inter-quartile range (IQR) for non-parametric data. Mixed effects model applying maximum likelihood estimation was used to estimate marginal linear predictions of monocular and binocular OPL and VPT and make comparisons among them. Random intercept at the subject level was used to account for the correlation between fellow eyes of the same subject. The relationships between age, gender and VLSQ score with OPL or VPT were evaluated by mixed effects linear regression. Logistic regression analysis was performed to evaluate factors associated with squeezing reflex. A p-value of <0.05 was considered statistically significant.

Results

A total of 75 normal subjects aged between 7 and 71 years were enrolled in this study. Median age of the subjects was 32.7 years (IQR, 20.3–47.9 years). A total of 40 (53.3%) were males and 35 (46.7%) were females. Median HIT score was 38 (IQR, 36–42 and range, 36–49) and median VLSQ score was 11 (IQR, 8–15 and range, 8–24). The median (IQR) scores for each of the eight responses in the VLSQ questionnaire were 1 (1–2), 1 (1–3), 1 (1–3), 1 (1–2), 1 (1–1), 1 (1–3), 1 (1–1) and 1 (1–1) respectively. The VLSQ-8 results indicated that photosensitivity symptoms generally were low among the normal subjects with responses rarely being higher than 3. All eyes had a visual acuity of 20/25 or better. Anterior segment evaluation was within normal limits in all eyes. None of the patients had dry eyes as evaluated on clinical history, examination and by performing Schirmer’s test. None of the subjects had color vision deficiency. Mean IOP was 12.7 (SD, 2.3) mm Hg and 12.6 (SD, 2.0) mm Hg in the right eye and left eye respectively.

Monocular and binocular visual photosensitivity thresholds

The summary results of monocular and binocular VPTs are shown in the Table 1. Monocular and binocular VPTs were comparable among themselves (right eye vs left eye: p = 0.88; right eye vs binocular: p = 0.47; left eye vs binocular: p = 0.38). The effects of age, gender, HIT score and VLSQ-8 score on VPT are shown in the Table 2. Age was positively correlated with monocular and binocular VPTs (Figure 2). There was no effect of gender and HIT score on VPT. There was a significant negative correlation between VLSQ score and VPTs. As age and VLSQ score were significant predictors of VPT in bivariate analysis, a multivariate regression was performed. This showed a significant negative correlation of VLSQ scores with monocular and binocular VPTs, and positive correlation of age with binocular VPT.

Table 1:

Monocular and binocular visual photosensitivity thresholds

Visual photosensitivity threshold (loglux)	Monocular (Right eye)	Monocular (Left eye)	Binocular
Mean (standard deviation)	3.34 (0.78)	3.33 (0.81)	3.37 (0.78)
Minimum	0.88	0.88	0.86
Maximum	4.54	4.54	4.54

This table shows the summary results of monocular and binocular visual photosensitivity thresholds.

Table 2:

Effect of age, gender, headache impact test score and visual light sensitivity questionnaire score on visual photosensitivity threshold

Laterality		Right eye		Left eye		Binocular
Analysis		BV	MV	BV	MV	BV	MV
Age	P-value	0.03	0.13	0.04	0.21	0.003	0.02
	Co-efficient ± SE	0.014 ± 0.006	0.009 ± 0.006	0.013 ± 0.006	0.008 ± 0.006	0.017 ± 0.006	0.013 ± 0.006
Gender	P-value	0.78	NI	0.86	NI	0.82	NI
	Co-efficient ± SE	0.053 ± 0.193		−0.036 ± 0.201		0.044 ± 0.189
Headache impact test score	P-value	0.15	NI	0.17	NI	0.01	NI†
	Co-efficient ± SE	−0.031 ± 0.022		−0.030 ± 0.022		−0.052 ± 0.021
Visual light sensitivity questionnaire score	P-value	0.0008	0.004	0.001	0.005	0.0008	0.005
	Co-efficient ± SE	−0.071 ± 0.021	−0.062 ± 0.022	−0.070 ± 0.022	−0.063 ± 0.022	−0.068 ± 0.020	−0.057 ± 0.020

This table shows the effects of age, gender, headache impact test (HIT) score and visual light sensitivity questionnaire (VLSQ-8) score on visual photosensitivity threshold assessed by bivariate and multivariate mixed effects linear regression (BV: bivariate; MV: multivariate; NI: not included; SE: standard error; ^†HIT score was not included in the multivariate analysis as it was significantly correlated with VLSQ score; p = 0.02).

Figure 2: Visual photosensitivity thresholds as a function of age –

Figure 2 shows the positive correlation between both monocular and binocular visual photosensitivity thresholds with age (red squares: monocular right eye; blue circles: monocular left eye; green triangles: binocular).

Monocular and binocular objective photosensitivity luminance

Among 75 subjects, infrared videos were acquired on 66, 62 and 59 subjects in binocular, right eye and left eye testing respectively. Among these, 56 (84.9%), 47 (75.8%) and 45 (76.3%) subjects showed a squeezing reflex (binocular, right eye and left eye) respectively and OPLs were noted. The summary results of monocular and binocular OPLs are shown in the Table 3 for the subjects who showed a squeezing reflex. The marginal linear predictions (Table 3) showed that monocular and binocular VPTs were significantly higher than OPLs (binocular: p<0.0001; right eye: p<0.0001; and left eye: p<0.0001). This suggests that overall, a squeeze reflex was observed at a light intensity lower than the subjective sensitivity threshold. There was no significant difference between right eye OPL and left eye OPL (p = 0.31), and right eye OPL and binocular OPL (p = 0.18). However, a significant difference between left eye OPL and binocular OPL (p = 0.02) was noted. The effects of age, gender, HIT score and VLSQ-8 score on OPL are shown in the Table 4. Age was positively correlated with only binocular OPL and there was no correlation between age and right eye or left eye OPL. There was no effect of gender on OPL. VLSQ-8 score was negatively correlated with only binocular OPL and there was no correlation between VLSQ-8 score and right eye or left eye OPL. As age and VLSQ score were significant predictors of binocular OPL in bivariate analysis, a multivariate regression was performed. This showed a significant (positive) correlation of only age with binocular OPL.

Table 3:

Monocular and binocular objective photosensitivity luminance

Objective photosensitivity luminance (loglux)		Monocular (Right eye)	Monocular (Left eye)	Binocular
Mean (standard deviation)		3.25 (0.55)	3.35 (0.47)	3.15 (0.52)
Minimum		1.84	1.97	0.67
Maximum		4.38	4.35	4.33
Marginal linear predictions, mean (standard error)	VPT	3.34 (0.09)	3.33 (0.10)	3.37 (0.09)
	OPL	3.05 (0.10)	3.04 (0.10)	3.02 (0.09)
p-value, VPT vs OPL		<0.0001	<0.0001	<0.0001

This table shows the summary results of monocular and binocular objective photosensitivity luminance in subjects who showed a squeezing reflex (OPL: objective photosensitivity luminance; VPT: visual photosensitivity threshold).

Table 4:

Effect of age, gender, headache impact test score and visual light sensitivity questionnaire score on objective photosensitivity luminance

Laterality		Right eye		Left eye		Binocular
Analysis		BV	MV	BV	MV	BV	MV
Age	P-value	0.25	NI	0.79	NI	0.01	0.04
	Co-efficient ± SE	0.006 ± 0.005		0.001 ± 0.005		0.011 ± 0.004	0.009 ± 0.005
Gender	P-value	0.20	NI	0.72	NI	0.18	NI
	Co-efficient ± SE	0.215 ± 0.169		0.054 ± 0.147		0.191 ± 0.144
Headache impact test score	P-value	0.12	NI	0.35	NI	0.003	NI†
	Co-efficient ± SE	−0.029 ± 0.018		−0.016 ± 0.018		−0.048 ± 0.016
Visual light sensitivity questionnaire score	P-value	0.17	NI	0.86	NI	0.04	0.20
	Co-efficient ± SE	−0.026 ± 0.019		−0.003 ± 0.019		−0.034 ± 0.017	−0.022 ± 0.017

This table shows the effects of age, gender, headache impact test (HIT) score and visual light sensitivity questionnaire (VLSQ-8) score on objective photosensitivity luminance assessed by bivariate and multivariate mixed effects linear regression (BV: bivariate; MV: multivariate; NI: not included; SE: standard error; ^†HIT score was not included in the multivariate analysis as it was significantly correlated with VLSQ score; p = 0.02).

Squeezing reflex

The subjects were divided into two groups based on the presence or absence of squeezing reflex, and age, gender, HIT score, VLSQ-8 score and VPT were compared between the two groups. The results are shown in the Table 5. The VPT was significantly higher in both monocular and binocular conditions when squeezing reflex was observed. This suggests that a higher photosensitivity threshold (decreased photosensitivity) was found in individuals who displayed a squeezing reflex.

Table 5:

Squeezing reflex

Laterality	Variables	Squeezing reflex		P-value
		Present	Absent
Binocular OPL	Age (years), median (IQR)	37.8 (20.8–49.8)	24.5 (19.1–39.2)	0.20
	Gender (male:female)	31:25	4:6	0.50
	HIT score, median (IQR)	38 (36–42)	39 (37.5–42.5)	0.72
	VLSQ score, median (IQR)	11 (8–14)	12.5 (10.8–18.3)	0.10
	VPT (loglux), median (IQR)	3.61 (3.24–3.93)	2.72 (1.38–3.24)	0.0003
Monocular OPL (right eye)	Age (years), median (IQR)	36.4 (19.9–48.3)	23.9 (18.5–35.7)	0.21
	Gender (male:female)	29:18	5:10	0.10
	HIT score, median (IQR)	38 (36–44)	40 (38–42)	0.34
	VLSQ score, median (IQR)	11 (8–14)	12 (11–19)	0.07
	VPT (loglux), median (IQR)	3.63 (3.18–4.06)	2.26 (1.60–3.24)	<0.0001
Monocular OPL (left eye)	Age (years), median (IQR)	36.4 (20.3–48.0)	25.3 (17.5–48.6)	0.28
	Gender (male:female)	25:20	7:7	1.00
	HIT score, median (IQR)	38 (36–42)	39 (37.5–43.5)	0.30
	VLSQ score, median (IQR)	11 (8–14)	16.5 (11.8–19.3)	0.006a
	VPT (loglux), median (IQR)	3.66 (3.31–4.21)	2.20 (1.60–2.72)	<0.0001

This table shows the results of squeezing reflex analysis [HIT: headache impact test; IQR: inter-quartile range; OPL: objective photosensitivity luminance; VLSQ: visual light sensitivity questionnaire; VPT: visual photosensitivity threshold; ^aOn performing a multiple logistic regression analysis, only VPT was predictive of squeezing reflex (p = 0.004) in left eye and not VLSQ score (p = 0.54)].

Discussion

Ocular photosensitivity analyzer is a customized, fully automated instrument with known reliability and repeatability in evaluating photosensitivity thresholds in human eyes. A pilot study with the instrument had given promising results and laid a strong foundation for validating the work in a large sample of healthy eyes.¹⁵ Since visual photosensitivity threshold is a measure of the photosensitivity that is still based on the subjective responses, we have introduced an objective measure of quantifying photosensitivity called objective photosensitivity luminance in this study. We have also evaluated the effect of age, gender, headache impact test score and visual light sensitivity questionnaire response on VPT and OPL.

The pilot study evaluated a total of 35 healthy Caucasian subjects, whereas in the present study, photosensitivity testing was performed on 75 healthy individuals in the Indian population.¹⁵ In contrast to the pilot study (average VPT of 2.2 loglux), the VPT obtained in the present study (average of 3.3 loglux) appear to be significantly higher by an order of 1 in the logarithmic scale. The difference can be attributed to possible ethnic variations in ocular melanin pigmentation.¹⁸ In both studies, the monocular and binocular VPTs are comparable. Adams et al found the average binocular photosensitivity thresholds in the normal US population to be 3.16 loglux.³ The study by Vingen et al found, in Norway healthy controls, that monocular discomfort thresholds was a median of 3.9 loglux (IQR, 3.3–3.4 loglux) and binocular discomfort threshold was a median of 3.4 loglux (IQR, 3.0–4.0 loglux),¹² that were quite similar to our study results.

The pilot study was limited by small sample size across decade-wise age categories in the adults. In the present study, a significant positive relationship between age and VPT was found suggesting tolerance to light can increase with age. Similar to the pilot study, the subject’s gender did not have an effect on VPTs. In the present study, the subjective VLSQ scores were found to have a negative relationship with VPTs as expected, whereas HIT scores did not have an effect on VPTs.

As eyelid squeezing can happen due to a discomfort for light, we objectively evaluated the luminance when the subjects squeezed their eyes, by viewing the recorded videos. We call this measure objective photosensitivity luminance and we found OPLs to be significantly lower than VPTs. This suggests that, when subjects had a squeezing response, the response appears well below the illumination that the subjects feel uncomfortable. The squeezing response may be an involuntary one before the voluntary uncomfortable feeling to light. In contrast to the VPTs, OPLs are significantly different between monocular and binocular measurements. The binocular OPLs are lower than the monocular OPLs suggesting that the squeezing response to light can occur earlier when viewed binocularly than monocular. A possible reason for the binocular OPLs being lower than monocular can be that monocular eye closure may play a role in reducing light discomfort. At a given light intensity, binocular reaction times are faster than monocular.¹⁹ A study by Wiggins and von Noorden reported that the mean monocular photophobia threshold was significantly higher than the binocular threshold.²⁰ It was also found that there was a significant positive relationship between age and binocular OPL.

A study by Yuhas et al utilized examiner grading of videos of light avoidance responses in normal as well as photophobic subjects.²¹ The results of the present study cannot be compared directly with the study of Yuhas et al, as the authors used red and blue lights while white light was used in the present study. The study by Yuhas et al developed a light aversion grading based on manual analysis of videos with tearing, blinking and fixation losses, whereas in the present study, only squeezing reflex was considered. Blinking is a common involuntary action and tearing can be due to subjects focusing on light for a longer duration with lesser number of blinks. Reduced number of blinks can lead to tear film break up and cause watering. While the protocol in the study by Yuhas et al was different from that in the present study (red/blue lights versus white, responses to a given light intensity rather than threshold light intensity measured and participation of photosensitive subjects versus only normal subjects), the study by Yuhas et al also indicated that masked observers could not identify a difference between photophobic subjects and controls based on examiner grading of videos.

Nearly one-fourth of the subjects did not show an eyelid squeezing response. On comparing these subjects with those who showed response, only VPTs were significantly higher by an order of 1 in the logarithmic scale in those with squeezing reflex. The subjects without squeezing response have a significantly lower tolerance for light and can give a voluntary response earlier, and the involuntary squeezing response need not be the case in all. Further studies on why differences were found between the thresholds reported subjectively and those reported objectively need to be explored.

The strengths of the study include a large sample size, a uniform racial population and the evaluation of objective photosensitivity luminance. Only age and visual light sensitivity score explain the variability in the visual photosensitivity threshold and objective photosensitivity luminance in the present study. Since ocular dominance, pupil size and palpebral fissure height may influence photosensitivity, these need to be evaluated in the future studies for a better understanding of the visual photosensitivity threshold and objective photosensitivity luminance. Although the present study provides normative objective photosensitivity luminance values in a healthy control group, based on the caveat provided in the study by Yuhas et al,²¹ future studies also need to evaluate objective photosensitivity luminance in subjects with self-reported photophobia.

To summarize, this study characterized visual photosensitivity thresholds in a large number of human eyes and established a normative data for both visual photosensitivity threshold and objective photosensitivity luminance. This would help in understanding the pathophysiology of the diseases causing photophobia, monitoring the disease progression and thereby evaluating the treatment modalities.

Supplementary Material

Supplementary Video

Supplementary video: Evaluation of objective photosensitivity luminance – This supplementary video shows first a slightly lower luminance (571.70 lux) that was not considered a ‘squeezing reflex’ (Figure 1A) and then the luminance (1085.33 lux) that was considered a ‘squeezing reflex’ (Figure 1B). The subjective visual photosensitivity threshold in this instance was 4514.20 lux.

Three key points.

• A subjective (patient-reported luminance thresholds for visual comfort) and an objective (examiner-graded responses to light) measure of light sensitivity in healthy subjects were investigated.

• Marginal linear predictions showed that subjective monocular and binocular visual photosensitivity thresholds were significantly higher than the objective photosensitivity luminance.

• The study results suggest that, overall, a squeeze reflex was observed at a light intensity lower than the subjective photosensitivity threshold.