Unraveling the diet-dermatitis connection: A systematic review of dietary exclusion in children with atopic dermatitis

INTRODUCTION

Pathophysiology of AD

The AD is attributed to a defective skin barrier and immune dysregulation driving cell-mediated cytokine-driven inflammation.^[1,36] Trikamjee et al.^[36] and Peng and Novak^[37] elucidated the intricate interplay among defective skin barriers, nutritional factors, environmental influences, and immune dysregulation in early life. Epidermal barrier dysfunction allows easier entry of microbes, allergens, and irritants triggering the immune system and releasing proinflammatory cytokines impairing the barrier and causing transepidermal water loss^[10] [Figure 1]. A major contributor to this barrier dysfunction is the filaggrin gene mutation located at the epidermal differentiation complex in chromosome 1, which decreases filaggrin expression essential for normal skin barrier functions. Filaggrin deficiency correlates with increased transepidermal water loss, reflecting AD severity.^[6]

Close

Figure 1:

Effects of skin barrier on the pathogenesis of atopic dermatitis. IgE: Immunoglobulin E, Th2: T helper 2, IL: Interleukin, TSLP: Thymic stromal lymphopoietin, TCR: T-cell receptors, MHC: Major histocompatibility test, TNF: Tumor necrosis factor, IF: Interferon, TWEAK: Tumor necrosis factor-like weak inducer of apoptosis.

Export to PPT

Filaggrin loss represents a genetic risk for AD implicating epidermal dysfunction and subsequent TH2 immune response.^[10] The World Allergy Association noted that immune system-dependent epidermal barrier function involves Th2 cytokines, like IL-4, inhibiting S100 and filaggrin expression further impairing the barrier. The Th2 cytokines, particularly IL-4 and IL-13, contribute to acute and chronic AD lesions suppressing epidermal differentiation and antimicrobial peptide production.^[6]

Physical skin irritation such as sweating, UV exposure, hot water use or mechanical damage from scratching exacerbates epidermal barrier impairment leading to increased transepidermal water loss and skin dryness.^[37] This emphasizes that acquired skin barrier dysfunction can result from mechanical disruption or exposure to irritants triggering inflammation through thymic stromal lymphopoietin release.^[6]

MATERIALS AND METHODS

The primary aim of this study was to critically examine current practices in dietary exclusion for children with AD, offering insights into rational dietary modifications and the intricate role of food in this dermatological condition. Through a comprehensive review, the research seeks to contribute valuable insights for more informed and effective dietary interventions in pediatric cases of AD.

A systematic review was conducted using available literature fulfilling the inclusion criteria from the year 2009 to 2023 to assess the current practices on dietary exclusion in children with AE. This review focused on the practices used in dietary exclusion given food hypersensitivity, food as a trigger to AE, and the coexistence of food allergy in children with AD.

The following terms/keywords syntax a search of studies was undertaken: “Atopic eczema,” “atopic dermatitis and food allergy,” “dietary exclusion and eczema,” “dietary exclusion and atopic dermatitis,” “dietary modification and eczema,” “food sensitization and eczema,” “food sensitization and atopic dermatitis,” and “food and eczema.”

The following literature search engines were used such as EBSCO host using MEDLINE database, Cochrane, PubMed, and Google Scholar. An individual literature search was also done using EMBASE for some of the literature, which was missing the full articles, which when manually added to the Covidence software was noted as duplicate.

Study screening and data extraction

Through the inclusion and exclusion criteria, the articles that merit critical appraisal from all identified citations stored in the electronic databases were selected. The screening of studies was done using the Covidence software, a screening and data extraction tool for conducting systematic reviews, which allows efficient and easily tracked screening. Using this software, 439 searched literatures from EBSCO host using MEDLINE database, Cochrane, PubMed, and Google Scholar were imported to the software where seven duplicates were removed. The screening of the titles and abstract was done for 432 citations. Among the 432 citations title and abstract, screening 56 studies warranted their full-text articles review to provide adequate information whereas further screening excluded 40 citations due to different reasons as shown in the PRISMA diagram below [Figure 2] and ended up with 16 citations.

Close

Figure 2:

Schematic flowchart of the study.

Export to PPT

Among the 16 citations, eight citations were then manually excluded (five were reviews, one each were a case report, a commentary, and a symposium article, which were not eligible for critical appraisal). Therefore, eight studies met the inclusion criteria and were used in this review where three studies were randomized controlled studies, three studies were systematic reviews, and two studies were cohort studies.

Quality assessment of the eight studies included in this review was done where through Covidence software; three randomized clinical trials were assessed using Cochrane risk of bias as shown in Table 3.

Table 3: Cochrane risk of bias for three randomized clinical trials.

Study	Allocation concealment	Blinding of outcome assessor for all outcomes	Blinding of outcome for all outcomes	Other sources of bias	Incomplete outcome data for all outcomes	Selective outcome reporting	Sequence generation
Mathew et al.[8]	Unclear	Unclear	High	Unclear	Unclear	Low	Unclear
Ridd et al.[38]	Unclear	Unclear	Unclear	Unclear	Low	Low	Low
Palmer et al.[39]	Low	Low	Low	Unclear	Unclear	Low	Low

Assessment of the methodological quality for three systematic reviews included in this review was done using the AMSTAR-2 tool, which is a 16-item tool that is reported to be effective for assessing the quality of systematic reviews that include non-randomized, randomized studies of a health intervention or both, and it is rated as high, moderate, low, and critically low quality as shown in Table 4.

Table 4: Quality assessment of included systematic reviews using the AMSTAR-2 tool.

Studies	Tsakok et al.,[42] 2016	Roberts et al.,[43] 2022	Das and Panda,[17] 2021
Q.1	Yes	Yes	Yes
Q.2	Yes	Yes	Partial yes
Q.3	Yes	Yes	No
Q.4	Yes	Partial yes	Yes
Q.5	Yes	Yes	Yes
Q.6	Yes	Yes	Yes
Q.7	Yes	Yes	Yes
Q.8	Yes	Yes	Partial yes
Q.9	Yes	Yes	No
Q.10	No	Yes	Yes
Q11	Yes	No meta-analysis conducted.	No meta-analysis conducted.
Q.12	Yes	No meta-analysis conducted.	No meta-analysis conducted.
Q.13	Yes	Yes	Yes
Q.14	Yes	Yes	No
Q.15	No	Yes	No meta-analysis conducted.
Q.16	Yes	Yes	Yes
Conclusion	Low	High	Critical low

The Newcastle-Ottawa scale was used to assess the quality of two cohort studies used in this review [Table 5].

Table 5: Newcastle-Ottawa scale was used to assess the quality of two cohort studies.

	Selection	Score	Comparability	Score	Outcome	Score	Risk of bias.
Eller et al.[19]	1. a) Truly representative.	1	1.a) The study controls for age, sex and marital status.	1	1.b) Record linkage	1	Good quality study-Low risk of bias
	2. b) Drawn from the same exposed cohort	1			2.Yes	1
	3. b) Structured interview.	1			3.b) Subject loss to follow up unlikely to introduce bias-number lost≤20% or description of those lost suggested no difference no different from those followed.	1
	4.Yes	1
Dhar et al.[44]	1.a) Truly representative	1	1.c) Cohort are not comparable on the basis of the design or analysis controlled for confounders.	0	1.b) Record linkage	1	Poor quality-High risk of bias.
	2.c) No description	0			2.a) Yes	1
	3.a) Secure records	1			3.d) No statement.	0
	4.Yes	1

RESULTS

To ensure transparency of the literature used in this systematic review, Table 6 has summarized this.

DISCUSSION

Through a comprehensive literature search, five databases were searched with a high yield in PubMed (278) followed by Cochrane (80) as seen in the PRISMA diagram and the articles were excluded for duplication. Eight studies met the eligibility criteria in this systematic review where three were randomized controlled trials (RCTs), three were systematic reviews, and two were cohort studies.^[38-45] Conflicts of interest were reported in 50% of the studies used in this review. Due to differences in quality and study designs of studies that met the inclusion criteria and increasing heterogeneity, a meta-analysis could not be performed.^[41] When assessing the severity of AD in studies that were used in this systematic review, different scoring system was used such as SCORAD, POEM, and EASI. In young infants, AD is reported to be the main risk for food sensitization, and this is due to the defective skin barrier, which predisposes to epicutaneous sensitization of food allergens.^[8] In this systematic review, the main reason for dietary exclusion in children with eczema was to alleviate the eczema symptoms where parents of children with eczema had tendencies to do dietary elimination, and some of them were advised from health-care facilities. Notably, Palmer et al.^[39] revealed the reason for dietary elimination in children with AD to be the prevention of food allergy. This was also stressed by Nosrati et al.^[45] where among the study participants, 87% reported following dietary elimination of different types of foods with an emphasis on empowering the patients on the role of food in the management of AE and the benefits of diet elimination since only 32.5% of patients visited a dermatologist. In their study, there were variations in the duration of dietary exclusion ranging from three weeks to six weeks.

Nearly, all the studies in this systematic review used patients with moderate-to-severe AE except a study that was done by Ridd et al.^[38] where study participants had mild-to-severe forms of eczema. Children with moderate-to-severe eczema or refractory eczema have a risk of developing food allergy, and hence, the recommendation is that they should be investigated for food allergy especially when eczema starts early in infancy (<3 months). However, mild cases of AD may not have such a risk.^[33] This further made it difficult to provide a generalization of risk since most of the studies used moderate-to-severe eczema, which is reportedly a risk of developing food allergy.

In this systematic review, foods that were excluded were cow’s milk, hen’s egg, peanut, codfish, rice, prawns, soybeans, seafood, and wheat. When a tailored elimination diet guided by a good history and confirmatory allergy tests is recommended, it helps to give a practical and rational dietary freedom to the patient, hence the importance of identifying the definite allergen instead of a blanket elimination.^[17] This is of paramount importance especially in children since an inadequate nutrient replacement following food elimination can result in nutritional deficiencies. As an example, children under two years of age with cow’s milk allergy developed protein energy malnutrition following the use of rice milk as a substitute for cow’s milk because rice milk could not meet the nutritional demands of these children under two years in this critical period for their growth and development.^[44,46] Paradoxically, most children develop food allergies <2 years of age; thus, a prompt action in diagnosing a specific food allergy is important since unnecessary food avoidance can put children at risk of nutrition deficiencies and failure to thrive.^[47]

Most parents suspect that all eczema symptoms are due to food allergies and implicate their role in the flares of AE. They often seek an allergy test to specifically guide them in their judicious dietary decision including dietary eliminations.^[38,46-51] Tait and Goldman,^[48] however, stressed that the parents’ report of eczematous reactions is sometimes a poor indicator of clinically relevant food allergy; hence, the emphasis should be on coupling the history, sensitivity tests, and confirming with the gold standard test for diagnosing specific food allergy. This is in keeping with what Tsakok et al.^[42] who postulated that food sensitization is likely to result from an inflamed skin barrier with the severity of eczema reported as a major risk factor for sensitization at an early age. This could support the fact that AD is thought to precede the development of food allergy due to epicutaneous sensitization caused by defective skin barrier in AE.^[8]

Approximately 80% of children with AD will have elevated food/allergen-specific IgE, but 35–40% will have confirmed food allergy.^[17] This is in keeping with a study done in South Africa by Gray et al.^[49] where they found that among 100 children with AD, 60% were sensitized to food/s, but only 40% had proven food allergy. Furthermore, Das and Panda^[17] emphasized that avoidance of proven food allergens can be practiced in the management of children with AD especially those who were previously sensitized. However, it should be guided, supervised, and must take into account the nutritional needs of the patient.

Delay in introducing allergenic foods such as oats, cow’s milk, fish, and eggs beyond six and nine months has been associated with an increased risk of allergenic diseases.^[39] Early introduction of such food is reported to promote food tolerance in young children; thus, parents and patients need to be educated about the effect of irrational avoidance of food since it can cause loss of food tolerance and cause subsequent food-induced immediate hypersensitivity reaction and anaphylaxis.^{[17,24,49-51]} An Australian health nut study revealed a significant risk of egg allergy in children, who were introduced to eggs between the age of 10–12 months or more when compared to those who were introduced at the age of four to six months of age.^[39]

Avoidance of food should be based on the history of exacerbation of symptoms and confirmed diagnosis of food allergy. When there is a consistent correlational symptom, a diagnostic elimination of up to 4–6 weeks can be done, and after an improvement of symptoms after dietary elimination, it should be confirmed by a double-blinded placebo control food challenge.^[8] Compared with IgE-mediated food allergy, which occurs immediately, non-IgE-mediated food allergies pose challenges due to the time delay between food contact and symptom (which normally takes 48–72 h).^[52] In such cases, no identifying test can be done, so the only diagnostic test will be to supervise dietary elimination of the suspected food allergen for four weeks expecting partial or complete resolution of symptoms such that the patient may present with abdominal cramps, vomiting, colic, diarrhea, failure to thrive, and blood in the stool.

Egg allergy and cow’s milk allergy are reported to be common in the younger age group when compared with fish allergy, tree nuts, and peanut allergy, which are reported to be common in older age groups.^[38,53] Furthermore, Ridd et al.^[38] reported that allergic reactions to different types of food may cause eczema symptoms and eczema is associated with food allergy. Tsakok et al.^[42] emphasized that 53–66% of people with eczema are food sensitized, while 15–81% report challenge-proven food allergy; this was also reiterated by Gray et al.^[49] who further emphasized that being sensitized to food does not necessarily mean allergic. There is a reported causal relationship between food allergy and AD, patients with AD were reported to be six times more likely to develop food sensitization when compared with their healthy controls.^[42] Severe persistent forms of AD and the early onset of AD are reported risk factors for food allergy.^[38,39,49] Early introduction of food has shown some benefits in developing oral tolerance.^[38,49] The defective barrier function in AD could explain the sensitization to food, which emphasizes the likelihood that AD is involved in the pathogenesis of food allergy rather than food exacerbating the AE, as epidermal barrier dysfunction is significant in the pathogenesis of AD.^[17,48,54] The blanket elimination of cows’ milk and eggs in all children with AE is injudicious; it was reported to be only beneficial in those patients, who were sensitized or allergic to cows’ milk and egg.^[48,54] Furthermore, Roberts et al.^[43] emphasized the rational use of dietary modification in children with AD since it can lead to long-term effects through malnutrition, discouragement of breastfeeding, and loss of oral tolerance to excluded foods. Unnecessary avoidance of food can cause a detrimental impact on the quality of life in patients with food allergies (FA).^[15,50] A diagnosis of FA is reported to have serious implications in patients’ life, limiting social activities and dietary freedom. Therefore, it is important to have a proper diagnosis since misdiagnosis can put a patient at risk of developing severe allergic reactions.^[25] A study done by Marklund et al.^[21] revealed that children, who were excluded from more than two diets were shorter compared to those who eliminated one diet. This shows that blanket dietary elimination can impair the growth and well-being of children. A study done among children aged 1–22 months in the US reported cases of kwashiorkor following dietary elimination, the main reasons being nutritional ignorance, food faddism or perceived milk intolerance.^[55] Tait and Goldman^[48] and Eigenmann et al.,^[50] also supported that the elimination of diet can jeopardize the nutritional status, and contribute to the development of clinical food allergy and failure to thrive. A study done in South Africa by Gray et al.^[49] revealed a high prevalence of food sensitization which was comparable with mixed races; however, the proven food allergy was less when compared to mixed races, and a significantly low peanut allergy in black children. This emphasizes the importance of test-guided dietary modification, instead of doing blanket or blind elimination of food/s; also, some of the allergens may be outgrown as children grow; therefore, it is important to be evidenced based and supervised and emphasize the proper interpretation of sensitivity tests since being sensitized does not necessarily mean allergic. Roberts et al.^[43] elaborated that there is a variation in healthcare practices on dietary elimination in established eczema. Ridd et al.^[38] emphasized the importance of clinicians focusing on promoting control of eczema and not cure messages when communicating with parents/patients by ensuring they emphasize the use of moisturizer, topical, and systemic medications as indicated. It is also important for a healthcare provider to explain to the caretaker/patient the relapsing nature of this chronic and relapsing inflammatory condition by avoiding the possible triggers and aiming at alleviating the presenting symptom instead of thinking of a cure. This was echoed by Eigenmann et al.,^[50] who emphasized the proper management of AE and looking for other possible triggers for AD since there are other triggers for eczema apart from food allergens. Only a third of patients with moderate-to-severe eczema are at risk of developing food allergy, and for those who are suspected to be allergic limited panel of food tests can be performed while explaining to the patient/parents the pros and cons of dietary elimination. Ensuring proper management of AD was also highlighted by Olabi and William^[2] who reported that underuse of topical therapy is the primary cause of treatment failure in AD, and this is chiefly due to the fear of toxicity of medications such as steroids. The AD is common in children; thus, the attending clinicians will regularly encounter children with food allergies or suspected FA; hence, the importance of educating healthcare providers to do a proper diagnosis of FA and rational avoidance of food, which is guided by proper investigation.^[52] Mehta et al.,^[47] further, stressed on ensuring proper dietary interventions for children with restricted foods due to food allergies to ensure the provision of adequate nutrients to avoid detrimental effects such as growth retardation. This systematic review provides a valuable opportunity for advancing the discourse surrounding the dietary restrictions in patients with AE emphasizing the need for timely medical intervention, and recognizing their profound relevance in effectively managing this prevalent childhood condition on a global scale.

CONCLUSION

This systematic review has unveiled a range of practices surrounding dietary exclusion in children diagnosed with AE revealing a common inclination among both parents and healthcare providers to eliminate specific foods due to concerns over exacerbating eczema symptoms. Parents often seek food allergy tests to guide their decisions regarding dietary modifications. It is important to educate parents and patients about the potential consequences of food avoidance, as it can lead to loss of food tolerance and subsequent immediate hypersensitivity reactions and anaphylaxis. While this review presents varying conclusions due to differences in study designs, a significant number of studies included in this review support evidence-guided food elimination as beneficial for children with moderate-to-severe AD. This emphasizes the need for robust, multi-center RCTs to establish a definitive evidence-based framework, aiding healthcare providers in making judicious decisions regarding food restriction in children with AD.