Vitamin D deficiency and atopic dermatitis severity in a Bangladeshi population living in East London: A cross‐sectional study

Abstract Background Atopic eczema is a common, chronic, inflammatory skin condition with considerable heterogeneity. South Asian people living in the UK frequently have low serum vitamin D3 (25(OH)D3), and those with atopic disease can present with severe eczema. The association between vitamin D deficiency and eczema severity, and the role of vitamin D supplementation in atopic eczema is inconsistent, and under‐researched in people with Asian ancestry. Objectives This cross‐sectional study investigates the association between serum 25(OH)D3 and eczema severity in a cohort of South Asian children and young adults living in London. Methods Eligible participants were Bangladeshi children and young adults aged 0–30 years with eczema, living in London and participating in the Tower Hamlets Eczema Assessment study. Data was collected via parent/patient self‐reporting, clinical history and examination, and hospital databases. 25(OH)D3 levels were documented retrospectively, if available, from hospital databases. Eczema severity was classified by Eczema Area and Severity Index (EASI) score less than or greater than 10 (clear‐mild vs. moderate‐severe). Multivariate logistic regression was used to adjust for confounding factors. Results 681 participants were included in analyses. 25(OH)D3 results were available for 49.6% (338/681), 84.3% of which had deficient or insufficient lowest 25(OH)D3. Lowest 25(OH)D3 was inversely correlated with EASI score (Spearman's rank R 2 = −0.24, p < 0.001). 26.1% (178/681) had EASI >10 and a lower median lowest and nearest 25(OH)D3. After adjustment for confounding EASI > 10 was significantly associated with a lowest 25(OH)D3 < 25 (OR 3.21, 95%CI 1.35, 8.60), use of mild‐moderate potency topical steroid on the face and neck (OR 3.11, 95%CI 1.86, 5.31), calcineurin inhibitor on the face and neck (OR 2.79, 95% CI 1.26, 6.10) and potent – very potent topical steroid on the face and neck (OR2.23, 95%CI 1.02, 4.77) and body (OR 2.11, 95%CI 1.18, 3.87). Discussion Vitamin D plays a role in modulation of proteins required for skin barrier function and regulation of the innate immune system, suggesting 25(OH)D3 deficiency contributes to skin inflammation. This study demonstrates a relationship between 25(OH)D3 deficiency and worse eczema severity in a cohort of South Asian children and young adults.


| BACKGROUND
Human skin pigmentation plays a pivotal role in mediating penetration of ultraviolet (UV) radiation into the deep layers of skin and the cutaneous circulation. 1 Depigmented skin has evolved in high latitude regions with low average UV light, such as the UK, facilitating cutaneous vitamin D production. 1,2][5][6] Vitamin D plays a key role in the innate and adaptive immune systems.A growing body of research is investigating the relationship between vitamin D and atopic diseases.There is evidence of lower vitamin D levels in atopic eczema, 5 however, the association between vitamin D and eczema severity is conflicted.In addition, there is inconclusive evidence on the benefit of vitamin D supplementation in atopic eczema. 7,8he Tower Hamlets borough in London has a large Bangladeshi population, with 32% of the population identifying as Bangladeshi in 2018. 9Anecdotally, atopic eczema in this population is severe, often requiring systemic treatment, and it is frequently complicated by eczema herpeticum.Filaggrin (FLG) is the major atopic eczema risk gene in the Bangladeshi population, and there is increased variation in FLG mutations. 10We hypothesise that there is a relationship between eczema severity and vitamin D deficiency in this population.
The primary aim of this study was to determine the relationship between serum 25(OH)D 3 and eczema severity in a population of Bangladeshi children and young adults living in East London.

| Regulatory approval
Ethical approval was granted through the Health Research Authority after review by the Hampstead Regional Ethics Committee (REC 18/LO/0017; ReDA reference: 011978).

| Study and subjects
This cross-sectional study forms part of the Tower Hamlets Eczema Assessment (THEA) study.All recruitment took place through paediatric and adult dermatology clinics, and paediatric allergy clinics, at The Royal London Hospital, London, UK.Bangladeshi children and young adults aged 0-30 years, living in East London, and with a diagnosis of atopic eczema confirmed by two dermatologists, were eligible for the study.Participants were considered Bangladeshi if both parents were from Bangladesh or if both sets of grandparents were from Bangladesh.Patients with known congenital recessive or x-linked ichthyoses, mixed ethnicities or equivocal diagnosis of atopic eczema were excluded.Participation included a once only study visit where data collection including questionnaires and physical examination were performed, and other medical history data was collected retrospectively from hospital databases.
Participants were recruited between May 2018 and December 2021.Data were collected by a Bangladeshi research assistant or research nurse, and a dermatology consultant or registrar.Written, informed consent or assent was obtained from the patient and/or their Eczema severity was classified by Eczema Area and Severity Index (EASI) score, which is a validated tool for clinical studies investigating atopic eczema. 12,13ASI score was calculated during the hospital visit.The primary outcome was the probability of having an EASI score >10 with a history of vitamin D deficiency.EASI score was dichotomized as EASI ≤ 10 (clear-mild) and EASI > 10 (moderate-severe) based upon previous data generated by the cohort, as described by Thomas et al. 14 EASI scores are often dichotomised into EASI ≤ 7 and EASI >7. 12,15Analysis of the relationship between serum vitamin D3 and EASI scores ≤ 7 and > 7 are available in supplementary information (Tables S1, S2, S4 & S5, Figure S3).

| Saliva collection for genomic DNA and targeted FLG sequencing
A sample of saliva was used to collect DNA for the project (GeneFixTM DNA Saliva Collectors from Isohelix, Harrietsham, UK).

| Statistical analysis
Descriptive statistics are presented as counts and percentages for categorical variables and mean and standard deviation for continuous variables.Chisquared, Fisher's exact or Wilcoxon rank-sum tests were used for bivariate analysis.
A binomial generalized linear model was fitted with EASI score (EASI ≤ 10 and EASI > 10) as the dependent variable.Variables were selected for inclusion based upon demographic features (age, sex, BMI, socio-economic class 16 ) and plausible predictive variables: lowest serum 25(OH)D 3 , 17 personal history of atopic disease, 18 current topical anti-inflammatory treatment on head and body, current and previous use of immunosuppressive therapy, and FLG loss of function mutations (FLG LOFM). 19ll statistical analyses were performed using R (version 4.1.2). p-values ≤ 0.05 were considered statistically significant.Full details of collection of saliva samples, DNA extraction and analysis and handling of missing data is available in supplementary information (Appendix 1).

| Demographic data
682 participants were recruited to the THEA study between May 2018 and December 2021.One participant is excluded from analyses due to missing EASI score.Baseline demographics are summarised in Table 1.Mean age at recruitment was 9.5 years (interquartile range (IQR) 4.8, 14.5) and 57% (391/ 681) of participants were male.All participants were of Bangladeshi descent and 97% (660/681) were born in the UK.

| DISCUSSION
Our data has demonstrated a relationship between lowest 25(OH)D 3 and worse eczema severity in a Bangladeshi population living in East London.After adjusting for confounding factors including age, sex, BMI, socio-economic class, history of atopic disease, current and previous treatment for atopic eczema and presence of FLG LOFM, patients who were vitamin D deficient had an increased risk of having moderate-severe eczema, as classified by EASI score.
Our data showed after adjusting for confounding and for multiple testing, lowest 25(OH)D 3 was associated     with EASI score at recruitment but nearest 25(OH)D 3 was not.As this is a retrospective study, our participants' 25(OH)D 3 levels could be a biomarker for levels of UV exposure.In addition, low UV-light exposure in people with non-white skin phototypes may be related to higher EASI scores.Vitamin D supplementation for eczema is controversial, with several studies inconclusive.A recent systematic review and meta-analysis demonstrated an overall positive effect of vitamin D supplementation on atopic dermatitis severity. 7here is a lack of research on eczema severity in non-white populations, and most current studies only discuss white skin or do not mention ethnicity. 8,20The majority (84.3%) of our participants had a history of deficient or insufficient 25(OH)D 3 , evidenced by lowest 25(OH)D 3 , supporting that vitamin D is low in people with pigmented skin living in low UV regions. 4,6Vitamin D3 and its metabolites regulate skin barrier function through synthesis of proteins such as filaggrin, keratinocyte differentiation and regulation of antimicrobial peptides, such as cathelicidins. 21Oral vitamin D3 supplementation in children with atopic eczema reduces AD severity through increasing the epidermal expression of the vitamin D receptor and cathelicidin in lesional skin. 22NICE and RCPCH recommend treatment of vitamin D deficiency in children for 8-12 weeks (colecalciferol 3000IU daily aged 1-5 months, 6000IU daily aged 6months-11 years, 10,000IU daily aged 12-17 years), while they recommend supplementation of vitamin D (colecalciferol 400-600IU daily) for children with vitamin D insufficiency. 23he presence of FLG LOFM has been linked to higher vitamin D levels, with the suggestion that the presence of these variants favours intracutaneous vitamin D3 production in people with Northern European ancestry. 24,25We do not see this relationship in our cohort, despite an enrichment of FLG mutations with 37.5% (204/551) having at least one FLG LOFM.This may be due to our population being at high risk of vitamin D deficiency/insufficiency, or due to the presence of different FLG variants in Asian populations compared to white European populations. 10he EASI score has been validated for the assessment of atopic eczema extent and severity. 15ssessment of erythema in atopic dermatitis is more difficult in skin of colour, however the EASI score is recommended as the optimal core measure of eczema severity for patients with all skin colours. 26Our data supports that EASI score is associated with eczema severity.This is demonstrated by associations with higher EASI scores and use of potent or very potent topical corticosteroid on the body, or treatment of eczema on the face.Those with higher EASI scores were more likely to have had admissions with complications of eczema and were more likely to be on immunosuppression/biologic therapy, therefore they were more likely to have had blood testing, including 25 (OH)D 3 .
This study has the advantage of assessing the relationship between EASI score and serum 25(OH)D 3 in a large cohort of Bangladeshi patients, who are under-studied in existing research into the role of vitamin D supplementation in eczema.Due to the nature of using a specific cohort we cannot be sure this would be generalisable to other populations living in the UK, or in lower latitude regions.Limitations include a lack of data on supplementation and dietary intake of vitamin D, and whether vitamin D supplementation resulted in a decrease in EASI score.Eczema severity including EASI score, Patient Orientated Eczema Measure and Dermatology Life Quality Index/Children's Dermatology Life Quality Index were assessed in a single visit and may not be truly representative of each patient's average eczema severity as most of the patients were on topical treatment.

| CONCLUSION
Vitamin D deficiency is common in this cohort of Bangladeshi children and young adults living in East London, regardless of eczema severity.The findings demonstrate a weak, inverse correlation between serum 25(OH)D 3 and eczema severity in the THEA cohort.Further research on the role of vitamin D supplementation in atopic eczema, particularly in people with non-white skin phototypes living in low UV regions, should be performed.

T A B L E 1
Baseline characteristics of whole cohort and by clear-mild versus moderate-severe eczema.

T A B L E 2
Baseline characteristics of whole cohort and categorised by lowest vitamin D3.

1 F I G U R E 2
Correlation between Eczema Area and Severity Index (EASI) score and lowest vitamin D3 level.Lowest 25(OH)D 3 is inversely related to EASI score Spearman's rank R 2 = −0.22,p < 0.001.Box plots of (a) lowest and (b) nearest vitamin D3 levels in participants with Eczema Area and Severity Index (EASI) ≤10 and EASI >10.

What is already known about this topic?
What does this study add?�This study assesses the relationship between Eczema Area and Severity Index (EASI) score and serum vitamin D3 in a large cohort of Bangladeshi patients, who are understudied in existing studies of the role of vitamin D supplementation in eczema.
Odds ratio of EASI >10 using lowest vitamin D3.
T A B L E 3