Skip to main content
NCBI home page
JAMA Network logoLink to JAMA Network
. 2022 Jul 27;158(9):1022–1030. doi: 10.1001/jamadermatol.2022.2823

Clinicophotobiological Characterization of Photoaggravated Atopic Dermatitis

Kirsty J Rutter 1,2,✉, Mark D Farrar 1,2, Elizabeth J Marjanovic 1,2, Lesley E Rhodes 1,2
PMCID: PMC9330275  PMID: 35895040

This case series study evaluates clinical and photobiological characteristics of photoaggravated atopic dermatitis.

Key Points

Question

What are the characteristic features of photoaggravated atopic dermatitis (PAD)?

Findings

In this case series of 120 patients with PAD, patients with different ages and skin phototypes were affected, usually presenting with sunlight-provoked or photodistributed eczema and having substantially impaired quality of life. Broadband provocation test results were positive for 93%, narrowband phototest findings were abnormal for 23%, and photopatch test reactions were positive for 15%; patients with darker skin were younger at onset and more frequently had abnormal narrowband phototest findings.

Meaning

The study revealed key features of PAD that may contribute to better disease recognition and diagnosis and supports development of personalized prevention and therapy.

Abstract

Importance

Photoaggravated atopic dermatitis (PAD) is estimated to affect 1.4% to 16% of patients with AD but remains poorly characterized with limited published data.

Objective

To provide detailed clinical and photobiological characterization of PAD.

Design, Setting, and Participants

This case series study used cross-sectional data collected from 120 consecutive patients diagnosed with PAD from January 2015 to October 2019 at a tertiary center referral unit for photobiology.

Main Outcomes and Measures

Routinely collected standardized clinical and photobiological data were analyzed using descriptive statistics, and regression analysis explored associations between demographic and clinical data.

Results

Of 869 patients who underwent photoinvestigation, 120 (14%) were diagnosed with PAD (69 female [58%]; median age, 45 [IQR, 31-61] years; range, 5-83 years; skin phototypes [SPTs] I-VI). Of these patients, 104 (87%) were adults. All patients had a history of AD, and most (62 of 104 [60%]) presented with sunlight-provoked or photodistributed eczema; median age at photosensitivity onset was 37 years (range, 1-72 years). Past-year Dermatology Life Quality Index score was greater than 10 for 80 of 103 adults (78%), and 82 of 119 (69%) had vitamin D (25-hydroxyvitamin D) level insufficiency or deficiency (<20 ng/mL; to convert ng/mL to nmol/L, multiply by 2.496). Broadband UV radiation provocation test results were positive for 112 patients (93%). In 28 patients (23%) with abnormal monochromator phototest findings, sensitivity occurred to UV-A, UV-B, and/or visible light, and UV-A of 350 ± 10 nm was the most prevalent wavelength. Photopatch test reactions were positive for 18 patients (15%). Patients with SPTs V to VI (31 [26%]) vs SPTs I to IV (89 [74%]) were younger at photosensitivity onset (median age, 24 years [IQR, 15-37 years] vs 40 years [IQR, 25-55 years]; P = .003), were more likely to be female (23 [74%] vs 46 [52%]; P = .03), and had a lower vitamin D status and a higher frequency of abnormal monochromator phototest findings.

Conclusions and Relevance

In this case series study, PAD affected patients with different ages and SPTs and was associated with substantially impaired quality of life. The findings suggest that confirming PAD through phototesting may provide better personalized care for patients through identification of provoking wavelengths, relevant photocontact allergies, and appropriate photoprotection advice.

Introduction

Photoaggravated atopic dermatitis (PAD) is a poorly understood subtype of AD that is worsened or provoked by exposure to UV radiation (UVR).1 Higher vs lower UVR exposure has been associated with reduced prevalence of AD,2,3 and patients may report benefit from sunlight exposure,4 with artificial UVR sources being used therapeutically.5 However, summer months and sunlight exposure have also been associated with worsening eczema in some patients.6,7 Historically, photoaggravation has been recognized to occur in a minority of patients with preexisting AD.8,9 The population prevalence of PAD has not been defined, but PAD is estimated to affect between 1.4% and 16% of people with AD.9,10,11,12 Nevertheless, studies focusing on this condition are scarce.

Some studies have described features of PAD,11,12,13 including eczema predominantly photodistributed over the face, v of the chest, and dorsal hands and/or a history of spring- or summer-related eczema aggravation. In those studies, conducted mainly with patients with light skin (skin phototypes [SPTs] I-III), female predominance was reported; the mean (SD) age at photosensitivity onset was 39 (18) years in the largest study, which included 44 patients with photosensitive AD.11 Phototesting in all 3 studies11,12,13 involved broadband UVR sources and supported sensitivity to UV-A and sometimes to UV-B. There is scant information on quality of life (QOL) in individuals with PAD.14

Using a clinicophotobiological data set, we sought to provide comprehensive information on demographic and clinical features of PAD. We also evaluated findings of detailed standardized photoinvestigation, including monochromator phototesting responses to UVR and visible light (VIS), to assess wavelength sensitivity, erythemal thresholds, and responses to solar-simulated radiation in patients with a range of SPTs.

Methods

In this case series study, we conducted a review of data collected from consecutive patients diagnosed with PAD by specialist photodermatologists (including K.J.R. and L.E.R.) in a tertiary center referral unit for photobiology between January 2015 and October 2019. Ethical approval was granted by the Health Research Authority (England) through the Integrated Research Application System and by Northern Care Alliance NHS Foundation Trust. Informed consent was waived by the Health Research Authority because the study was limited to secondary data analysis of previously collected, nonidentifiable information. This study followed the reporting guideline for case series.

Patients included in this study had a previous diagnosis of AD and were referred for assessment of suspected photosensitivity. The diagnosis of AD was made by the referring dermatologists and was reviewed for confirmation, for example, using the UK Working Party’s Diagnostic Criteria for Atopic Dermatitis.15 We defined PAD as a condition occurring in patients with diagnostic criteria for AD who showed evidence of photoprovoked eczema of constitutional origin but did not show the characteristic phototest findings of chronic actinic dermatitis.16,17,18,19 Assessments were conducted using standardized pro forma clinical and photobiological documentation for detailed history and examination (including rash appearance, triggers, and timing; patient-reported provocation through window glass; and physician-determined phototyping), QOL assessment, monochromator phototesting, broadband UVR provocation testing, photopatch testing, plasma and urinary porphyrin scan, serum autoantibody screening, and immunoglobulin E (IgE) and vitamin D (25-hydroxyvitamin D) measurement. Self-reported data on ethnicity were included to provide further demographic data. History of prescribed and nonprescribed drugs was obtained, and patients suspected of drug photosensitivity were not included in the analysis. Quality of life was assessed using the Dermatology Life Quality Index (DLQI) among adults (age ≥18 years) or the Children’s DLQI (CDLQI) among children (age, 5-17 years).20,21 Because evaluation of QOL based on scores over the past week may be inappropriate for intermittent, sunlight-provoked disease,14 a modified DLQI or CDLQI was also completed that provided scores over the past year.22

To determine erythemal thresholds using monochromator phototesting, skin on the middle of the back was exposed to narrow bandwidths of UV-B, UV-A, and VIS using a 1-kW xenon arc lamp coupled to a 0.25-m grating monochromator (Newport Spectra-Physics Ltd). Incremental dose series were used for 300, 320, 330, 350, 370, and 400 nm, and single doses were used for 500 and 600 nm (half-maximum bandwidth of 10 nm at 300-350 nm, 15 nm at 370-400 nm, and 20 nm at 500-600 nm). Reference ranges were originally established at another UK center.23

Photoprovocation testing was performed on 5-cm2 areas of ventral forearm skin for up to 3 consecutive days. One forearm was exposed to 15 J/cm2 broadband UV-A (320-400 nm) using a custom-built unit incorporating Cleo Performance bulbs (Philips Healthcare UK Ltd). The contralateral forearm was exposed to 10 J/cm2 solar-simulated radiation (290-400 nm) with a 1-kW xenon arc lamp plus an atmospheric attenuation filter (Newport Spectra-Physics Ltd). These doses represent physiological exposure approximating 30 minutes of local midday summer sun on a clear day in the UK. Mild erythema alone was not classified as abnormal because these doses could be associated with mild erythema in some individuals without photosensitivity.23

Photopatch and control patch testing was performed for 19 UV filters, 4 nonsteroidal anti-inflammatory agents (Chemotechnique Diagnostics), 3 prescribable sunscreen products, and patients’ own sunscreen products.24 Duplicate patches were applied to skin on the middle and lower back for 24 hours before 1 set was irradiated with 5 J/cm2 broadband UV-A (320-400 nm; UVAL 801; Waldmann GmbH & Co). Readings were obtained 24 and 48 hours after irradiation using International Contact Dermatitis Research Group grading.25

Statistical Analysis

On the basis of clinical suspicion that patients with darker skin had differing characteristics from those with lighter skin, a subgroup analysis was performed for SPTs I to IV and SPTs V to VI. Subgroup comparisons used unpaired t tests or nonparametric equivalents as appropriate. Multiple linear or logistic regression explored associations between demographic and clinical variables. Missing data were, at most, 5% for all variables except rash provocation through window glass (89 of 120 patients), specific referral indications (104 of 120), and CDLQI (12 of 16). Complete case analysis was conducted using Prism software, version 8 (GraphPad) and SPSS, version 23 (IBM UK Ltd). Two-sided P < .05 was considered significant.

Results

Study Population and Referral Indications

Of 869 patients who underwent photoinvestigation, 120 (14%) were diagnosed with PAD (Table 1). Median age at diagnosis was 45 years (IQR, 31-61 years; range, 5-83 years), including 104 adult patients (87%) and 16 patients younger than 18 years (13%); 69 (58%) were female. A total of 6 patients (5%) were Afro-Caribbean, 1 (0.8%) East Asian, 25 (21%) South Asian, and 84 (70%) White British. Indications for referrals were given for 104 patients; 62 patients (60%) were referred with sunlight-provoked and/or photodistributed eczema. Specific presenting features on referral included seasonally worse eczema (50 [48%]), photodistributed eczema (45 [43%]), history of eczema worsening or flaring after sun exposure (20 [19%]), flare of eczema during holiday (13 [13%]), flare of eczema with phototherapy (11 [11%]), and flare of eczema during photopatch testing (1 [1%]). (Some patients had >1 presenting feature on referral.)

Table 1. Clinical Characteristics of Patients With Photoaggravated AD.

Characteristic Patientsa P valueb
All (N = 120) SPTs I-IV (n = 89) SPTs V-VI (n = 31)
Age at diagnosis, median (IQR), y 45 (31-61) 52 (35-62) 34 (26-43) <.001
Age at onset of photosensitivity, median (IQR), y 37 (22-54) 40 (25-55) 24 (15-37) .003
Photosensitivity duration at diagnosis, median (IQR), y 5 (2-13) 5 (2-11) 6 (2-16) NA
Onset of AD <12 y of age 95 (79) 73 (82) 22 (71) .19
Photosensitivity concurrent with onset of ADc 16 (13) 12 (13) 4 (13) NA
Sex
Female 69 (58) 46 (52) 23 (74) .03
Male 51 (42) 43 (48) 8 (26)
Female to male ratio 1.35:1 1.07:1 2.88:1 NA
Ethnicityd
Afro-Caribbean 6 (5) 0 6 (20) NA
East Asian 1 (1) 1 (1) 0 NA
South Asian 25 (21) 1 (1) 24 (77) NA
White British 84 (70) 84 (94) 0 NA
Other or mixede 4 (3) 3 (4) 1 (3) NA
Serum IgE level, median (IQR), mg/dLf 0.168 (0.034-0.975) 0.158 (0.034-0.936) 0.185 (0.049-1.440) NA
25-Hydroxyvitamin D level, median (IQR), ng/mL 15.0 (9.25-22.0) 16.8 (10.9-23.3) 11.8 (7.1-18.3) .01
Vitamin D statusg
Deficiency 34 (29) 20 (23) 14 (48) .03
Insufficiency 48 (40) 36 (41) 12 (41)
Sufficiency 37 (31) 32 (36) 5 (17)
Quality of life scoreh
DLQI
Past week >10 44 (43) 34 (44) 10 (40) NA
Past year >10 80 (78) 58 (75) 22 (85) NA
CDLQI
Past week >12 4 (33) 3 (38) 0 NA
Past year >12 8 (62) 5 (63) 3 (60) NA
Reported rash provocation through window glassi 42 (47) 26 (38) 16 (76) .003
Open in a new tab

Abbreviations: AD, atopic dermatitis; CDLQI, Children’s Dermatology Life Quality Index; DLQI, Dermatology Life Quality Index; IgE, immunoglobulin E; NA, not applicable; SPTs, skin phototypes.

SI conversion factors: To convert serum IgE from mg/dL to mg/L, multiply by 10; 25-hydroxyvitamin D from ng/mL to nmol/L, multiply by 2.496.

a

Data are presented as number (percentage) of patients unless otherwise indicated.

b

SPTs I to IV vs SPTs V to VI.

c

Data were self-reported.

d

Ethnicity was self-reported.

e

Other or mixed level of detail was not available because data were taken from a database with no further breakdown.

f

To convert serum IgE from mg/dL to IU/mL, multiply by 4166.667. The reference range for IgE is less than 0.02736 mg/dL (<114 IU/mL).

g

Vitamin D deficiency was defined as a level below 10 ng/mL; insufficiency, 10 to 20 ng/mL; and sufficiency, greater than 20 ng/mL; n = 119.

h

Scores range from 0 to 30, with higher scores indicating a more impaired quality of life.

i

Data available for 89 patients.

History of Photosensitivity

Most patients (89 [74%]) could specify their age at onset of photosensitivity, which was a median of 37 years (range, 1-72 years); 12 (10%) were unaware they were photosensitive, 13 (11%) reported gradual onset of photosensitivity, and 6 (5%) did not have the information recorded. Onset of AD occurred before age 12 years in 95 patients (79%), and photosensitivity onset was concurrent with AD onset in 16 (13%). Delays in diagnosis were common, with median age at diagnosis 8 years later than median age at photosensitivity onset.

A total of 55 patients (46%) reported that exposures of less than 1 hour were sufficient to provoke or aggravate their condition; flares could last several days to weeks (eTable 1 in the Supplement). Provocation of rash through window glass was reported by 42 patients (47%). Description of rash volunteered by patients was assessed in a random sample of 52 patients and included redness (40 [77%]); dryness, scaling, or flaking (38 [73%]); itching (27 [52%]); weepy skin or tiny blisters (27 [52%]); swelling or tightness (24 [46%]); burning or pain (16 [31%]); bumpy skin (13 [25%]); and pigmentation (10 [19%]).

Laboratory Investigations

Serum IgE level was elevated in 97 patients (81%) (median, 0.168 mg/dL [IQR, 0.034-0.975 mg/dL]; to convert mg/dL to mg/L, multiply by 10). Connective tissue disease screen results were positive in 1 patient with an antinuclear antibody titer of 1:100. Only 37 patients (31%) had a serum vitamin D (25-hydroxyvitamin D) level of 20 ng/mL or greater (to convert ng/mL to nmol/L, multiply by 2.496), which is regarded as sufficient,26 whereas 82 patients (69%) had vitamin D level insufficiency or deficiency. Even in summer months, only 9 of 25 patients (36%) who had vitamin D status assayed had vitamin D sufficiency.

Quality of Life

Overall, 80 of 103 adults (78%) experienced very or extremely impaired QOL, with DLQI past-year score greater than 10 (Table 1 and Figure 1). Median past-week DLQI score of 9 (IQR, 3.25-14.25) was comparable with other chronic dermatoses,20,27 and median past-year DLQI score of 16 (IQR, 11.5-20.5) was significantly higher than past-week DLQI scores (P < .001). Similar patterns were observed for children, with median past-year and past-week CDLQI scores of 14 (IQR, 11-19) and 6.5 (IQR, 5-13), respectively, indicating highly impaired QOL.

Figure 1. Box Plots of Dermatology Life Quality Index (DLQI) and Children’s DLQI (CDLQI) Scores for Past Week and Past Year.

Figure 1.

Open in a new tab

DLQI and CDLQI scores range from 0 to 30, with higher scores indicating a more impaired quality of life. Horizontal lines inside boxes indicate medians; outer horizontal box lines, IQRs; and whiskers, ranges. Dashed lines represent boundaries for very large or extremely large effect on quality of life (top line, CDLQI; bottom line, DLQI). SPT indicates skin phototype.

aP < .001.

bP < .05.

The multiple linear regression model evaluating associations between past-week DLQI score and other variables showed that provocation through window glass and age at photosensitivity onset were significantly associated with past-week DLQI score (Table 2). Patients reporting provocation through window glass had covariate-adjusted past-week DLQI scores 4.7 points higher than patients without window provocation (β, 4.701; 95% CI, 1.247-8.156; P = .008). Age at onset of photosensitivity was associated with past-week DLQI score, although each subsequent year of onset was associated with only a 0.1-point higher past-week DLQI score (β, 0.115; 95% CI, 0.009-0.221; P = .03).

Table 2. Clinical Factors Associated With Age at Onset of Photosensitivity, Quality of Life, and Abnormal Monochromator Test Findings Among Patients With Photoaggravated Atopic Dermatitis.

Model β or OR (95% CI)a P value
Linear regression
Factors associated with age at onset of photosensitivity
Female −3.758 (−11.692 to 4.177) .35
SPTs V-VI −10.904 (−19.609 to −2.199) .02
Factors associated with past-week DLQI
Age at onset of photosensitivity 0.115 (0.009 to 0.221) .03
Female 1.428 (−1.929 to 4.785) .40
SPTs V-VI 1.363 (−2.954 to 5.680) .53
Abnormal monochromator test findings −0.283 (−4.839 to 4.272) .90
Positive window provocation results 4.701 (1.247 to 8.156) .008
Factors associated with past-year DLQI
Age at onset of photosensitivity −0.008 (−0.098 to 0.082) .86
Female 2.865 (0.028 to 5.702) .048
SPTs V-VI 0.930 (−2.718 to 4.578) .61
Abnormal monochromator test findings −1.534 (−5.383 to 4.578) .43
Positive window provocation results 4.220 (1.301 to 7.139) .005
Logistic regression
Factors associated with abnormal monochromator test findings
Female 0.883 (0.303 to 2.575) .82
SPTs V-VI 3.859 (1.325 to 11.238) .01
Age at onset of photosensitivity 0.990 (0.964 to 1.017) .47
Positive window provocation results 1.391 (0.502 to 3.851) .53
Open in a new tab

Abbreviations: DLQI, Dermatology Life Quality Index; OR, odds ratio; SPTs, skin phototypes.

a

Data are reported as β coefficients (95% CI) for multivariable adjusted linear regression and as OR (exponent of the β coefficient) (95% CI) for logistic regression (abnormal monochromator test findings) with all factors associated with each category listed under each analysis forced in the model. The exponent of the β coefficient is equivalent to the OR.

Positive window provocation results and female sex were significantly associated with past-year DLQI scores. Patients reporting provocation through window glass had covariate-adjusted past-year DLQI scores 4.2 points higher than those without window provocation (β, 4.220; 95% CI, 1.301-7.139; P = .005), and females had adjusted past-year DLQI scores 2.9 points higher than males (β, 2.865; 95% CI, 0.028-5.702; P = .048).

Phototesting

Broadband UVR provocation test results were positive for 108 of the 116 patients (93%) who were tested with both solar-simulated radiation and broadband UV-A, and all 4 patients tested with only broadband UV-A provocation had positive responses (Table 3 and eTable 2 in the Supplement). Solar-simulated radiation and broadband UV-A testing aided diagnosis; 20 patients (17%) showed a positive reaction to solar-simulated radiation alone, 10 of 116 patients (9%) showed a positive reaction to broadband UV-A alone, and 78 (67%) had a positive reaction to both. Morphologic features of 189 positive provocation reactions included erythema with dryness, scaling, and/or excoriations (74 [39%]); erythema with tiny papules (51 [27%]); moderate or severe erythema (42 [22%]); mixed pattern (14 [7%]); and erythema with edema (8 [4%]). A total of 37 of 108 patients (34%) who had positive provocation required 3 consecutive daily challenges to provoke a positive reaction. Eight of 116 patients (7%) who received provocations with both solar-simulated radiation and broadband UV-A (or 8 of the total 120 patients [7%]) developed only mild erythema and were diagnosed with PAD based on a history of sun-aggravated or photodistributed eczema, supported by examination or photographs. Two of these patients were phototested while taking immunosuppressing medication.

Table 3. Phototest Results in Patients With Photoaggravated Atopic Dermatitis.

Phototesting results Patients, No. (%)a P valueb
All SPTs I-IV SPTs V-VI
Abnormal monochromator test findingsc 28 (23) 16 (18) 12 (39) .02
≥1 Positive photopatch reaction 18 (15) 14 (16) 4 (13) >.99
≥1 Positive control patch reaction 20 (17) 17 (19) 3 (10) .28
Broadband provocation
Positive results
With both SSR and BBUV-A 78 (67) 59 (68) 19 (66) NA
Only with SSR 20 (17) 14 (16) 6 (21) .64
Only with BBUV-A 10 (9) 8 (9) 2 (7) NA
Negative results with SSR and BBUV-A 8 (7) 6 (7) 2 (7) NA
Open in a new tab

Abbreviations: BBUV-A, broadband UV-A; NA, not applicable; SPTs, skin phototypes; SSR, solar-simulated radiation.

a

Monochromator, photopatch, and control patch testing: all, n = 120; SPTs I-IV, n = 89; and SPTs V-VI, n = 31. Broadband provocation testing: all, n = 116; SPTs I-IV, n = 87; and SPTs V-VI, n = 29. Four patients received broadband provocation testing only with BBUV-A (not with SSR) for technical reasons; all 4 experienced positive provocation responses to BBUV-A.

b

SPTs I to IV vs SPTs V to VI.

c

Reduced minimal erythema doses in 1 or more waveband.

Monochromator phototest findings were abnormal for 28 patients (23%), who showed reduced erythemal thresholds in at least 1 narrow waveband. Reduced minimal erythema doses (MEDs) to UV-A wavelengths alone occurred in 18 patients (64%), to UV-A and UV-B in 5 patients (18%), to UV-B alone in 4 patients (14%), and to UV-A and VIS in 1 patient (4%). The most prevalent wavelength for lowered MEDs was 350 ± 10 nm (Figure 2 and eTable 3 in the Supplement). When abnormal UV-B thresholds occurred, MEDs were only mildly reduced, contrasting with those occurring in chronic actinic dermatitis.

Figure 2. Patients With Reduced Minimal Erythema Doses (MEDs) on Monochromator Testing.

Figure 2.

Open in a new tab

Wavelengths included UV-A, UV-B, and visible light. SPT indicates skin phototype.

aP = .009.

bP = .03.

Photopatch Testing

Positive photopatch test reactions occurred in 18 of 120 patients (15%). Ten patients (8%) showed 1 positive reaction, and 8 (7%) had more than 1 positive reaction. Most reactions involved sunscreen filters and were not thought to be of current clinical relevance, but 3 patients (3%) experienced reactions to their own sunscreen product. The agents most commonly associated with positive photopatch reactions were butyl methoxydibenzoylmethane (4 of 18 [22%]) and benzophenone-3 (4 of 18 [22%]). Twenty patients (17%) showed at least 1 positive contact reaction (ie, to control patch and photopatch). The agents most commonly associated with positive contact reactions were octocrylene (5 of 20 [25%]) and methylene bis-benzotriazolyl tetramethylbutylphenol (3 of 20 [15%]) (eTable 4 in the Supplement).

Darker vs Lighter Skin Type

Almost all patients with SPTs I to IV (84 of 89 [94%]) self-identified as White British, and patients with SPTs V to VI predominantly self-identified as Afro-Caribbean (6 of 31 [19%]) and South Asian (24 of 31 [77%]) (Table 1). Compared with patients with lighter skin (SPTs I-IV; 89 [74%]), patients with darker skin (SPTs V-VI; 31 [26%]) were younger at photosensitivity onset (median age, 24 [IQR, 15-37] vs 40 [IQR, 25-55] years; P = .003) and at diagnosis (median age, 34 [IQR, 26-43] vs 52 [IQR, 35-62]; P < .001) and were more likely to be female (23 [74%] vs 46 [52%]; P = .03). After adjusting for sex to account for unequal male to female ratios, the age difference remained statistically significant, with onset 11 years earlier in patients with darker skin (β, −10.904; 95% CI, −19.609 to −2.199; P = .02) (Table 2). Patients with darker skin had lower 25-hydroxyvitamin D levels and were more likely to report provocation through window glass. Similar proportions in the 2 groups had abnormal broadband provocation test results and a similar number of provocations for a response (Table 3). However, patients with darker skin were more likely to have abnormal monochromator MED test findings (12 [39%] vs 16 [18%]; P = .02); higher rates were seen across most UV-B and UV-A wavelengths (Figure 2).

In a logistic regression model examining the association between abnormal monochromator test findings and other characteristics, patients with darker skin had 3.9 times greater odds of abnormal monochromator test findings compared with those with lighter skin (odds ratio, 3.859; 95% CI, 1.325-11.238) after adjusting for age at photosensitivity onset, sex, and window provocation (Table 2). There was no significant difference in DLQI or CDLQI scores between the 2 groups.

Discussion

In this study, we provided detailed clinicophotobiological characterization of, to our knowledge, the largest reported case series of patients with PAD. Although there are few published studies, photoaggravation is estimated to affect 1.4% to 16% of patients with AD,10,12 and therefore PAD may be underreported and underrecognized. In our study, PAD occurred in patients with different ages, SPTs, and ethnicities and was associated with substantially impaired QOL. We found notable differences in features between patients with darker and lighter skin, including earlier onset of photosensitivity, greater female predominance, and more frequently reduced erythemal thresholds in patients with darker skin. Monochromator phototesting showed that sensitivity occurred principally to UV-A but could occur to UV-B and VIS; the most prevalent wavelength for reduced erythemal threshold was midrange UV-A at 350 ± 10 nm. Broadband UVR provocation results were positive for nearly all patients and confirmed the diagnosis. These findings may enhance the recognition, diagnosis, and consequently, management of patients with PAD.

The referral indications that we identified may alert physicians to potential PAD; these included seasonally worse, sunlight-provoked, or photodistributed eczema but also a history of eczema flaring during holiday or during phototherapy or photopatch testing. Reported provocation of rash through window glass in 47% of patients suggests UV-A (or VIS) sensitivity.28 Provocation through window glass was associated with worse QOL scores and may help identify patients who experience greater negative consequences.

Previous studies focusing on PAD have used broadband UV-A and UV-B sources.11,12,13 In contrast, we performed detailed phototesting to narrow bandwidths of UV-B, UV-A, and VIS, ranging from 300 to 600 nm, in 120 patients. This enabled examination of the action spectrum of PAD in patients with abnormal findings (23%). Photosensitivity was seen predominantly in the UV-A range and was sometimes accompanied by sensitivity to UV-B or VIS. Erythemal thresholds were modestly reduced with UV-A or UV-B, and in only 4 cases, monochromator phototesting indicated abnormality to UV-B alone. This finding contrasts with the exquisite sensitivity, especially to UV-B, that is typical of chronic actinic dermatitis.16,17,18,19 Understanding the action spectrum of PAD may assist future exploration of its pathogenesis and help optimize photoprotective management, including appropriate sunscreen use.

Furthermore, we found a high yield of positive broadband provocation test results (ie, in 93% of patients with PAD) when low-dose broadband UV-A and solar-simulated radiation were used. In previous studies,11,12,13 positive provocation was reported in nearly all patients, with varying provocation by broadband UV-A and/or UV-B. Some studies used higher erythema-inducing doses (eg, 2 MEDs of UV-A or UV-B with daily increments of 20% to 40%),11 whereas the doses in our study approximated 2 standard erythemal doses (ie, in keeping with casual exposure and usually below the personal erythemal threshold [MEDs]).29 Atopic dermatitis is well recognized to show polymorphic clinical features, particularly among individuals from different ethnic groups,30 and although the most common broadband response type in our study was clinically eczematous, we also observed papular erythema and mixed features. Although it has been suggested that papular photoprovocation responses reflect AD and coexisting polymorphic light eruption,11 histologic examination findings consistent with eczema have been found in such responses.12,13 Features discriminating PAD from polymorphic light eruption plus AD included patient history; examination findings; photographs showing eczematous features of a photodistributed rash, such as scaling, flaking, or dryness; and eczematous laboratory photoprovocation responses that could include papules, although papules in patients with PAD were smaller than those typical of polymorphic light eruption.

Our findings of positive photopatch and control patch reactions in a substantial proportion of patients (15% and 17%, respectively) highlights the importance of performing these tests in patients with suspected PAD. Multiple factors may contribute to an exposed-site eczema, and elimination of identified photocontact or contact allergens is an additional important component of management. Moreover, most positive reactions were to sunscreen filters, and because appropriate sunscreen use is a key component of management, it is important to identify these allergens. The findings were broadly in line with those reported in a photopatch study from Europe,31 in which the most common sunscreen filters associated with photoallergic contact reactions were octocrylene, benzophenone-3, and butyl methoxydibenzoylmethane.

Overall, our finding that most patients had childhood-onset AD and later onset of photosensitivity (median age, 37 years [range, 1-72 years]), with more female patients (69) than male patients (51), broadly agreed with previous, smaller case series studies of patients with PAD.11,12,13 However, in a subgroup analysis, we found that patients with darker skin (SPTs V-VI) were 16 years younger at photosensitivity onset (median age, 24 years) and were more likely to be female (74%) compared with patients with SPTs V to VI (median age, 40 years; 52% female). Interpreting phototests, in particular perception of erythema, can be more challenging in individuals with darker skin.32,33,34 Despite this, we found that a greater proportion of patients with SPTs I to IV than patients with SPTs V to VI had abnormal erythemal thresholds. Whereas previous studies of PAD predominantly involved patients with SPTs I to III, abnormal photoinvestigation findings were reported in 9 Japanese patients,13 and a case series study in India that included 22 patients (SPTs IV to VI) with PAD (although diagnosed without photoinvestigation) also showed female predominance (8:1).35 Although it is important in clinical practice to be alert to these differences, the reasons for these are unclear. However, there are well-recognized differences in clinical, genetic, and immunological features of AD among patients from different ethnic groups,30,36 and there may be differences in memory responses at least to some contact allergens.37,38 In addition, clinical and demographic differences in several other photodermatoses associated with skin color and ethnicity have been reported39,40 and reviewed,41 although reasons for these differences remain unexplained.

Vitamin D insufficiency was common, with a similar prevalence as that in reported data for patients with photodermatoses in the UK,42,43 where sun avoidance or protection was the principal factor associated with vitamin D insufficiency.43 Vitamin D status was significantly lower in patients with SPTs V to VI, potentially for multifactorial reasons, including physiological and dietary factors.44,45 In accordance with recommendations for a vitamin D status sufficient to maintain musculoskeletal health and with particular consideration for people with low exposure to sunlight, vitamin D supplementation should be considered for patients with PAD.26,46,47

Patients with photodermatoses contend both with their symptoms and the restrictions that avoidance of sunlight imposes, with a consequent reduction in QOL.14 The use of modified (past-year) DLQI appeared to have captured impact of an intermittent, sunlight-provoked condition more than standard (past-week) DLQI, as reflected in the higher scores. Poorer QOL in female patients with PAD is consistent with studies of AD at visible sites48 and may be particularly relevant to photodistribution.

Strengths and Limitations

Strengths of our study include the use of standardized, detailed data collection tools and photodiagnostic methods, enabling comprehensive data analysis. In addition, wider ranges of ages and SPTs were represented compared with previous studies of PAD.11,12,13

Limitations include the use of data derived from a single specialist center, which could limit generalizability. Some measures relied on self-reported items, which could be limited by recall bias.

Conclusions

This case series study characterized the clinical and photobiological features of PAD, a poorly recognized subtype of AD. Photoaggravated atopic dermatitis affected patients with different ages and SPTs and was associated with substantially impaired QOL. The study’s findings suggest that improved knowledge of PAD presentation, demographic aspects, and photoinvestigation can assist patient diagnosis and treatment and that more attention should be given to this condition.

Supplement.

eTable 1. Detail of Sunlight Provocation Responses Reported by Patients With Photoaggravated Atopic Dermatitis

eTable 2. Detail of Laboratory Broadband UVR Provocation Responses in Patients With Photoaggravated Atopic Dermatitis

eTable 3. Detail of Monochromator Phototesting in 120 Patients With Photoaggravated Atopic Dermatitis

eTable 4. Number of Patients With Photoaggravated Atopic Dermatitis With Positive Reactions to Agents in Photopatch Test Battery

Click here for additional data file. (136.3KB, pdf)

References

  • 1.Werth VP, Honigsmann H. Photoaggravated dermatoses. In: Lim HW, Honigsmann H, Hawk JL, eds. Photodermatology. CRC Press; 2007:251-267. doi: 10.3109/9781420019964-17 [DOI] [Google Scholar]
  • 2.Osborne NJ, Ukoumunne OC, Wake M, Allen KJ. Prevalence of eczema and food allergy is associated with latitude in Australia. J Allergy Clin Immunol. 2012;129(3):865-867. doi: 10.1016/j.jaci.2012.01.037 [DOI] [PubMed] [Google Scholar]
  • 3.Silverberg JI, Hanifin J, Simpson EL. Climatic factors are associated with childhood eczema prevalence in the United States. J Invest Dermatol. 2013;133(7):1752-1759. doi: 10.1038/jid.2013.19 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Patrizi A, Savoia F, Giacomini F, Tabanelli M, Gurioli C. The effect of summer holidays and sun exposure on atopic dermatitis. G Ital Dermatol Venereol. 2009;144(4):463-466. [PubMed] [Google Scholar]
  • 5.Garritsen FM, Brouwer MW, Limpens J, Spuls PI. Photo(chemo)therapy in the management of atopic dermatitis: an updated systematic review with implications for practice and research. Br J Dermatol. 2014;170(3):501-513. doi: 10.1111/bjd.12645 [DOI] [PubMed] [Google Scholar]
  • 6.Krämer U, Weidinger S, Darsow U, Möhrenschlager M, Ring J, Behrendt H. Seasonality in symptom severity influenced by temperature or grass pollen: results of a panel study in children with eczema. J Invest Dermatol. 2005;124(3):514-523. doi: 10.1111/j.0022-202X.2005.23625.x [DOI] [PubMed] [Google Scholar]
  • 7.Sargen MR, Hoffstad O, Margolis DJ. Warm, humid, and high sun exposure climates are associated with poorly controlled eczema: PEER (Pediatric Eczema Elective Registry) cohort, 2004-2012. J Invest Dermatol. 2014;134(1):51-57. doi: 10.1038/jid.2013.274 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Frain-Bell W, Scatchard M. The association of photosensitivity and atopy in the child. Br J Dermatol. 1971;85(2):105-110. doi: 10.1111/j.1365-2133.1971.tb07193.x [DOI] [PubMed] [Google Scholar]
  • 9.Dawe RS. Photo-aggravated dermatoses. In: Ferguson J, Dover JS, eds. Photodermatology. CRC Press; 2006:57-65. doi: 10.1201/b15138-12 [DOI] [Google Scholar]
  • 10.Williams JR, Burr ML, Williams HC. Factors influencing atopic dermatitis—a questionnaire survey of schoolchildren’s perceptions. Br J Dermatol. 2004;150(6):1154-1161. doi: 10.1111/j.1365-2133.2004.05869.x [DOI] [PubMed] [Google Scholar]
  • 11.ten Berge O, van Weelden H, Bruijnzeel-Koomen CA, de Bruin-Weller MS, Sigurdsson V. Throwing a light on photosensitivity in atopic dermatitis: a retrospective study. Am J Clin Dermatol. 2009;10(2):119-123. doi: 10.2165/00128071-200910020-00004 [DOI] [PubMed] [Google Scholar]
  • 12.Ellenbogen E, Wesselmann U, Hofmann SC, Lehmann P. Photosensitive atopic dermatitis—a neglected subset: clinical, laboratory, histological and photobiological workup. J Eur Acad Dermatol Venereol. 2016;30(2):270-275. doi: 10.1111/jdv.13451 [DOI] [PubMed] [Google Scholar]
  • 13.Tajima T, Ibe M, Matsushita T, Kamide R. A variety of skin responses to ultraviolet irradiation in patients with atopic dermatitis. J Dermatol Sci. 1998;17(2):101-107. doi: 10.1016/S0923-1811(97)00080-7 [DOI] [PubMed] [Google Scholar]
  • 14.Rutter KJ, Ashraf I, Cordingley L, Rhodes LE. Quality of life and psychological impact in the photodermatoses: a systematic review. Br J Dermatol. 2020;182(5):1092-1102. doi: 10.1111/bjd.18326 [DOI] [PubMed] [Google Scholar]
  • 15.Williams HC, Burney PG, Pembroke AC, Hay RJ. The U.K. Working Party’s Diagnostic Criteria for Atopic Dermatitis. III. Independent hospital validation. Br J Dermatol. 1994;131(3):406-416. doi: 10.1111/j.1365-2133.1994.tb08532.x [DOI] [PubMed] [Google Scholar]
  • 16.Menagé HD, Harrison GI, Potten CS, Young AR, Hawk JL. The action spectrum for induction of chronic actinic dermatitis is similar to that for sunburn inflammation. Photochem Photobiol. 1995;62(6):976-979. doi: 10.1111/j.1751-1097.1995.tb02396.x [DOI] [PubMed] [Google Scholar]
  • 17.Ferguson J. Chronic actinic dermatitis. In: Ferguson J, Dover JS, eds. Photodermatology. CRC Press; 2006:39-45. doi: 10.1201/b15138-8 [DOI] [Google Scholar]
  • 18.Hawk JL, Lim HW. Chronic actinic dermatitis. In: Lim HW, Honigsmann H, Hawk JL, eds. Photodermatology. CRC Press; 2007:169-184. doi: 10.3109/9781420019964-12 [DOI] [Google Scholar]
  • 19.Tan KW, Haylett AK, Ling TC, Rhodes LE. Comparison of demographic and photobiological features of chronic actinic dermatitis in patients with lighter vs darker skin types. JAMA Dermatol. 2017;153(5):427-435. doi: 10.1001/jamadermatol.2016.5861 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Finlay AY, Khan GK. Dermatology Life Quality Index (DLQI)—a simple practical measure for routine clinical use. Clin Exp Dermatol. 1994;19(3):210-216. doi: 10.1111/j.1365-2230.1994.tb01167.x [DOI] [PubMed] [Google Scholar]
  • 21.Lewis-Jones MS, Finlay AY. The Children’s Dermatology Life Quality Index (CDLQI): initial validation and practical use. Br J Dermatol. 1995;132(6):942-949. doi: 10.1111/j.1365-2133.1995.tb16953.x [DOI] [PubMed] [Google Scholar]
  • 22.Jong CT, Finlay AY, Pearse AD, et al. The quality of life of 790 patients with photodermatoses. Br J Dermatol. 2008;159(1):192-197. doi: 10.1111/j.1365-2133.2008.08581.x [DOI] [PubMed] [Google Scholar]
  • 23.Diffey BL, Farr PM. The normal range in diagnostic phototesting. Br J Dermatol. 1989;120(4):517-524. doi: 10.1111/j.1365-2133.1989.tb01325.x [DOI] [PubMed] [Google Scholar]
  • 24.Gonçalo M, Ferguson J, Bonevalle A, et al. Photopatch testing: recommendations for a European photopatch test baseline series. Contact Dermatitis. 2013;68(4):239-243. doi: 10.1111/cod.12037 [DOI] [PubMed] [Google Scholar]
  • 25.Wahlberg JE, Lindberg M. Patch testing. In: Frosch PJ, Menné T, Lepoittevin J-P, eds. Contact Dermatitis. Springer Berlin Heidelberg; 2006:365-390. doi: 10.1007/3-540-31301-X_22 [DOI] [Google Scholar]
  • 26.Institute of Medicine . Dietary Reference Intakes for Calcium and Vitamin D. The National Academies Press; 2011. [PubMed] [Google Scholar]
  • 27.Lundberg L, Johannesson M, Silverdahl M, Hermansson C, Lindberg M. Health-related quality of life in patients with psoriasis and atopic dermatitis measured with SF-36, DLQI and a subjective measure of disease activity. Acta Derm Venereol. 2000;80(6):430-434. doi: 10.1080/000155500300012873 [DOI] [PubMed] [Google Scholar]
  • 28.Tuchinda C, Srivannaboon S, Lim HW. Photoprotection by window glass, automobile glass, and sunglasses. J Am Acad Dermatol. 2006;54(5):845-854. doi: 10.1016/j.jaad.2005.11.1082 [DOI] [PubMed] [Google Scholar]
  • 29.Diffey BL, Jansén CT, Urbach F, Wulf HC. The standard erythema dose: a new photobiological concept. Photodermatol Photoimmunol Photomed. 1997;13(1-2):64-66. doi: 10.1111/j.1600-0781.1997.tb00110.x [DOI] [PubMed] [Google Scholar]
  • 30.Yew YW, Thyssen JP, Silverberg JI. A systematic review and meta-analysis of the regional and age-related differences in atopic dermatitis clinical characteristics. J Am Acad Dermatol. 2019;80(2):390-401. doi: 10.1016/j.jaad.2018.09.035 [DOI] [PubMed] [Google Scholar]
  • 31.Kerr AC, Ferguson J, Haylett AK, et al. ; European Multicentre Photopatch Test Study (EMCPPTS) Taskforce . A European multicentre photopatch test study. Br J Dermatol. 2012;166(5):1002-1009. doi: 10.1111/j.1365-2133.2012.10857.x [DOI] [PubMed] [Google Scholar]
  • 32.Diffey BL, Robson J. The influence of pigmentation and illumination on the perception of erythema. Photodermatol Photoimmunol Photomed. 1992;9(2):45-47. [PubMed] [Google Scholar]
  • 33.Lock-Andersen J, Wulf HC. Threshold level for measurement of UV sensitivity: reproducibility of phototest. Photodermatol Photoimmunol Photomed. 1996;12(4):154-161. doi: 10.1111/j.1600-0781.1996.tb00192.x [DOI] [PubMed] [Google Scholar]
  • 34.Shih BB, Allan D, de Gruijl FR, Rhodes LE. Robust detection of minimal sunburn in pigmented skin by 785-nm laser speckle contrast imaging of blood flux. J Invest Dermatol. 2015;135(4):1197-1199. doi: 10.1038/jid.2014.507 [DOI] [PubMed] [Google Scholar]
  • 35.Wadhwani AR, Sharma VK, Ramam M, Khaitan BK. A clinical study of the spectrum of photodermatoses in dark-skinned populations. Clin Exp Dermatol. 2013;38(8):823-829. doi: 10.1111/ced.12098 [DOI] [PubMed] [Google Scholar]
  • 36.Brunner PM, Guttman-Yassky E. Racial differences in atopic dermatitis. Ann Allergy Asthma Immunol. 2019;122(5):449-455. doi: 10.1016/j.anai.2018.11.015 [DOI] [PubMed] [Google Scholar]
  • 37.Robinson MK. Population differences in skin structure and physiology and the susceptibility to irritant and allergic contact dermatitis: implications for skin safety testing and risk assessment. Contact Dermatitis. 1999;41(2):65-79. doi: 10.1111/j.1600-0536.1999.tb06229.x [DOI] [PubMed] [Google Scholar]
  • 38.Dickel H, Taylor JS, Evey P, Merk HF. Comparison of patch test results with a standard series among White and Black racial groups. Am J Contact Dermat. 2001;12(2):77-82. [DOI] [PubMed] [Google Scholar]
  • 39.Kerr HA, Lim HW. Photodermatoses in African Americans: a retrospective analysis of 135 patients over a 7-year period. J Am Acad Dermatol. 2007;57(4):638-643. doi: 10.1016/j.jaad.2007.05.043 [DOI] [PubMed] [Google Scholar]
  • 40.Nakamura M, Henderson M, Jacobsen G, Lim HW. Comparison of photodermatoses in African-Americans and Caucasians: a follow-up study. Photodermatol Photoimmunol Photomed. 2014;30(5):231-236. doi: 10.1111/phpp.12079 [DOI] [PubMed] [Google Scholar]
  • 41.Sharma VK, Sahni K. Photodermatoses in the pigmented skin. Adv Exp Med Biol. 2017;996:111-122. doi: 10.1007/978-3-319-56017-5_10 [DOI] [PubMed] [Google Scholar]
  • 42.Reid SM, Robinson M, Kerr AC, Ibbotson SH. Prevalence and predictors of low vitamin D status in patients referred to a tertiary photodiagnostic service: a retrospective study. Photodermatol Photoimmunol Photomed. 2012;28(2):91-96. doi: 10.1111/j.1600-0781.2011.00644.x [DOI] [PubMed] [Google Scholar]
  • 43.Rhodes LE, Webb AR, Berry JL, et al. Sunlight exposure behaviour and vitamin D status in photosensitive patients: longitudinal comparative study with healthy individuals at U.K. latitude. Br J Dermatol. 2014;171(6):1478-1486. doi: 10.1111/bjd.13325 [DOI] [PubMed] [Google Scholar]
  • 44.Shih BB, Farrar MD, Cooke MS, et al. Fractional sunburn threshold UVR doses generate equivalent vitamin D and DNA damage in skin types I-VI but with epidermal DNA damage gradient correlated to skin darkness. J Invest Dermatol. 2018;138(10):2244-2252. doi: 10.1016/j.jid.2018.04.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Kift R, Berry JL, Vail A, Durkin MT, Rhodes LE, Webb AR. Lifestyle factors including less cutaneous sun exposure contribute to starkly lower vitamin D levels in U.K. South Asians compared with the white population. Br J Dermatol. 2013;169(6):1272-1278. doi: 10.1111/bjd.12518 [DOI] [PubMed] [Google Scholar]
  • 46.Scientific Advisory Committee on Nutrition . Vitamin D and health. Published July 2016. Accessed April 10, 2022. https://www.gov.uk/government/publications/sacn-vitamin-d-and-health-report
  • 47.Passeron T, Bouillon R, Callender V, et al. Sunscreen photoprotection and vitamin D status. Br J Dermatol. 2019;181(5):916-931. doi: 10.1111/bjd.17992 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Holm EA, Esmann S, Jemec GB. Does visible atopic dermatitis affect quality of life more in women than in men? Gend Med. 2004;1(2):125-130. doi: 10.1016/S1550-8579(04)80017-2 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement.

eTable 1. Detail of Sunlight Provocation Responses Reported by Patients With Photoaggravated Atopic Dermatitis

eTable 2. Detail of Laboratory Broadband UVR Provocation Responses in Patients With Photoaggravated Atopic Dermatitis

eTable 3. Detail of Monochromator Phototesting in 120 Patients With Photoaggravated Atopic Dermatitis

eTable 4. Number of Patients With Photoaggravated Atopic Dermatitis With Positive Reactions to Agents in Photopatch Test Battery

Click here for additional data file. (136.3KB, pdf)

Articles from JAMA Dermatology are provided here courtesy of American Medical Association

RESOURCES